├── .gitignore ├── INSTALL.md ├── LICENSE-CC-BY-NC-4.0.md ├── README.md ├── demo.jpg ├── description ├── README.md └── robots │ ├── cfg │ ├── config.yaml │ └── robot │ │ └── unitree_g1_29dof_anneal_23dof.yaml │ ├── dtype.py │ └── g1 │ ├── 29dof_official.urdf │ ├── dof_axis.npy │ ├── g1_23dof_lock_wrist.urdf │ ├── g1_23dof_lock_wrist.xml │ ├── g1_23dof_lock_wrist_fitmotionONLY.xml │ ├── g1_29dof_rev_1_0.xml │ ├── g1_29dof_rev_1_0_with_toe.xml │ ├── meshes │ ├── head_link.STL │ ├── left_ankle_pitch_link.STL │ ├── left_ankle_roll_link.STL │ ├── left_elbow_link.STL │ ├── left_elbow_link_merge.STL │ ├── left_hand_index_0_link.STL │ ├── left_hand_index_1_link.STL │ ├── left_hand_middle_0_link.STL │ ├── left_hand_middle_1_link.STL │ ├── left_hand_palm_link.STL │ ├── left_hand_thumb_0_link.STL │ ├── left_hand_thumb_1_link.STL │ ├── left_hand_thumb_2_link.STL │ ├── left_hip_pitch_link.STL │ ├── left_hip_roll_link.STL │ ├── left_hip_yaw_link.STL │ ├── left_knee_link.STL │ ├── left_rubber_hand.STL │ ├── left_shoulder_pitch_link.STL │ ├── left_shoulder_roll_link.STL │ ├── left_shoulder_yaw_link.STL │ ├── left_wrist_pitch_link.STL │ ├── left_wrist_roll_link.STL │ ├── left_wrist_roll_rubber_hand.STL │ ├── left_wrist_yaw_link.STL │ ├── logo_link.STL │ ├── pelvis.STL │ ├── pelvis_contour_link.STL │ ├── right_ankle_pitch_link.STL │ ├── right_ankle_roll_link.STL │ ├── right_elbow_link.STL │ ├── right_elbow_link_merge.STL │ ├── right_hand_index_0_link.STL │ ├── right_hand_index_1_link.STL │ ├── right_hand_middle_0_link.STL │ ├── right_hand_middle_1_link.STL │ ├── right_hand_palm_link.STL │ ├── right_hand_thumb_0_link.STL │ ├── right_hand_thumb_1_link.STL │ ├── right_hand_thumb_2_link.STL │ ├── right_hip_pitch_link.STL │ ├── right_hip_roll_link.STL │ ├── right_hip_yaw_link.STL │ ├── right_knee_link.STL │ ├── right_rubber_hand.STL │ ├── right_shoulder_pitch_link.STL │ ├── right_shoulder_roll_link.STL │ ├── right_shoulder_yaw_link.STL │ ├── right_wrist_pitch_link.STL │ ├── right_wrist_roll_link.STL │ ├── right_wrist_roll_rubber_hand.STL │ ├── right_wrist_yaw_link.STL │ ├── torso_constraint_L_link.STL │ ├── torso_constraint_L_rod_link.STL │ ├── torso_constraint_R_link.STL │ ├── torso_constraint_R_rod_link.STL │ ├── torso_link.STL │ ├── torso_link_23dof_rev_1_0.STL │ ├── torso_link_rev_1_0.STL │ ├── waist_constraint_L.STL │ ├── waist_constraint_R.STL │ ├── waist_roll_link.STL │ ├── waist_roll_link_rev_1_0.STL │ ├── waist_support_link.STL │ ├── waist_yaw_link.STL │ └── waist_yaw_link_rev_1_0.STL │ ├── phc_g1_23dof.yaml │ └── smpl_humanoid.xml ├── example ├── motion_data │ ├── Bruce_Lee_pose.pkl │ ├── Charleston_dance.pkl │ ├── Hooks_punch.pkl │ ├── Horse-stance_pose.pkl │ ├── Horse-stance_punch.pkl │ ├── Roundhouse_kick.pkl │ └── Side_kick.pkl ├── pretrained_horse_stance_pose │ ├── config.yaml │ ├── exported │ │ └── model_50000.onnx │ └── model_50000.pt ├── pretrained_horse_stance_pose_2 │ ├── config.yaml │ ├── exported │ │ └── model_119000.onnx │ └── model_119000.pt └── pretrained_horse_stance_punch │ ├── config.yaml │ ├── exported │ └── model_33000.onnx │ └── model_33000.pt ├── humanoidverse ├── README.md ├── __init__.py ├── agents │ ├── __init__.py │ ├── base_algo │ │ └── base_algo.py │ ├── callbacks │ │ ├── __init__.py │ │ ├── analysis_plot_force.py │ │ ├── analysis_plot_force_estimator.py │ │ ├── analysis_plot_locomotion.py │ │ ├── analysis_plot_locomotion_estimate_vel.py │ │ ├── analysis_plot_motion_tracking.py │ │ ├── analysis_plot_motion_tracking_openloop.py │ │ ├── analysis_plot_template.html │ │ └── base_callback.py │ ├── mh_ppo │ │ └── mh_ppo.py │ ├── modules │ │ ├── __init__.py │ │ ├── data_utils.py │ │ ├── encoder_modules.py │ │ ├── modules.py │ │ └── ppo_modules.py │ └── ppo │ │ └── ppo.py ├── config │ ├── algo │ │ └── mh_ppo.yaml │ ├── base.yaml │ ├── base │ │ ├── fabric.yaml │ │ ├── hydra.yaml │ │ └── structure.yaml │ ├── base_eval.yaml │ ├── deploy │ │ ├── external.yaml │ │ ├── multiple.yaml │ │ └── single.yaml │ ├── domain_rand │ │ ├── NO_domain_rand.yaml │ │ ├── domain_rand_base.yaml │ │ ├── dr_nil.yaml │ │ └── main.yaml │ ├── env │ │ ├── base_task.yaml │ │ ├── legged_base.yaml │ │ └── motion_tracking.yaml │ ├── exp │ │ ├── base_exp.yaml │ │ ├── legged_base.yaml │ │ └── motion_tracking.yaml │ ├── obs │ │ ├── legged_obs.yaml │ │ └── motion_tracking │ │ │ ├── benchmark.yaml │ │ │ └── main.yaml │ ├── opt │ │ ├── record.yaml │ │ └── wandb.yaml │ ├── rewards │ │ └── motion_tracking │ │ │ └── main.yaml │ ├── robot │ │ ├── g1 │ │ │ ├── g1_23dof_lock_wrist.yaml │ │ │ └── g1_only_leg.yaml │ │ └── robot_base.yaml │ ├── simulator │ │ ├── genesis.yaml │ │ ├── isaacgym.yaml │ │ ├── isaacsim.yaml │ │ └── mujoco.yaml │ └── terrain │ │ ├── terrain_base.yaml │ │ ├── terrain_locomotion.yaml │ │ └── terrain_locomotion_plane.yaml ├── deploy │ ├── __init__.py │ ├── external │ │ ├── __init__.py │ │ └── core.py │ ├── mujoco.py │ └── urcirobot.py ├── envs │ ├── __init__.py │ ├── base_task │ │ ├── __init__.py │ │ └── base_task.py │ ├── env_utils │ │ ├── __init__.py │ │ ├── history_handler.py │ │ ├── terrain.py │ │ └── visualization.py │ ├── legged_base_task │ │ ├── __init__.py │ │ └── legged_robot_base.py │ └── motion_tracking │ │ ├── __init__.py │ │ └── motion_tracking.py ├── eval_agent.py ├── isaac_utils │ ├── isaac_utils │ │ ├── __init__.py │ │ ├── maths.py │ │ └── rotations.py │ └── setup.py ├── measure_traj.py ├── ratio_eps.py ├── sample_eps.py ├── simulator │ ├── base_simulator │ │ └── base_simulator.py │ ├── genesis │ │ ├── genesis.py │ │ ├── genesis_mjdebug.py │ │ ├── genesis_viewer.py │ │ └── tmp_gs_utils.py │ ├── isaacgym │ │ ├── isaacgym.py │ │ └── isaacgym_hoi.py │ └── isaacsim │ │ ├── event_cfg.py │ │ ├── events.py │ │ ├── isaaclab_cfg.py │ │ ├── isaaclab_viewpoint_camera_controller.py │ │ ├── isaacsim.py │ │ └── isaacsim_articulation_cfg.py ├── train_agent.py ├── urci.py └── utils │ ├── __init__.py │ ├── average_meters.py │ ├── common.py │ ├── config_utils.py │ ├── devtool.py │ ├── helpers.py │ ├── inference_helpers.py │ ├── logging.py │ ├── math.py │ ├── motion_lib │ ├── __init__.py │ ├── dof_axis.npy │ ├── motion_lib_base.py │ ├── motion_lib_robot.py │ ├── motion_lib_robot_WJX.py │ ├── motion_utils │ │ ├── __init__.py │ │ ├── flags.py │ │ └── rotation_conversions.py │ ├── skeleton.py │ └── torch_humanoid_batch.py │ ├── noise_tool.py │ ├── real │ ├── __init__.py │ └── rotation_helper.py │ └── torch_utils.py ├── motion_source ├── README.md ├── convert_lafan_pkl.py ├── count_pkl_contact_mask.py └── utils │ ├── __init__.py │ ├── rotation_conversions.py │ └── torch_humanoid_batch.py ├── overview.jpg ├── robot_motion_process ├── README.md ├── motion_interpolation_pkl.py ├── motion_readpkl.py ├── traj_vis.ipynb ├── vis_q_mj.py └── vis_rr.py ├── setup.py ├── smpl_retarget ├── README.md ├── mink_retarget │ ├── __init__.py │ ├── convert_fit_motion.py │ ├── retargeting │ │ ├── __init__.py │ │ ├── config.py │ │ ├── mink_retarget.py │ │ ├── torch_humanoid_batch.py │ │ └── torch_smplx_humanoid_batch.py │ └── shape_optimized_neutral.pkl ├── motion_filter │ ├── constants.py │ ├── model │ │ ├── __init__.py │ │ └── smpl.py │ └── utils │ │ ├── __init__.py │ │ ├── mesh_utils.py │ │ ├── motion_filter.py │ │ ├── part_volumes.py │ │ └── smpl_paraser.py ├── phc_retarget │ ├── fit_smpl_motion.py │ └── fit_smpl_shape.py ├── poselib │ ├── .gitignore │ ├── README.md │ ├── poselib │ │ ├── __init__.py │ │ ├── core │ │ │ ├── __init__.py │ │ │ ├── backend │ │ │ │ ├── __init__.py │ │ │ │ ├── abstract.py │ │ │ │ └── logger.py │ │ │ ├── rotation3d.py │ │ │ ├── tensor_utils.py │ │ │ └── tests │ │ │ │ ├── __init__.py │ │ │ │ └── test_rotation.py │ │ ├── skeleton │ │ │ ├── __init__.py │ │ │ ├── backend │ │ │ │ ├── __init__.py │ │ │ │ └── fbx │ │ │ │ │ ├── __init__.py │ │ │ │ │ ├── fbx_py27_backend.py │ │ │ │ │ └── fbx_read_wrapper.py │ │ │ ├── skeleton3d.py │ │ │ └── tests │ │ │ │ ├── __init__.py │ │ │ │ ├── ant.xml │ │ │ │ ├── test_skeleton.py │ │ │ │ └── transfer_npy.py │ │ └── visualization │ │ │ ├── __init__.py │ │ │ ├── common.py │ │ │ ├── core.py │ │ │ ├── plt_plotter.py │ │ │ ├── simple_plotter_tasks.py │ │ │ ├── skeleton_plotter_tasks.py │ │ │ └── tests │ │ │ ├── __init__.py │ │ │ └── test_plotter.py │ └── setup.py └── retargeted_motion_data │ └── phc │ └── shape_optimized_v1.pkl └── smpl_vis ├── README.md ├── import_motion_blender.py ├── smpl_render.py └── utils ├── __init__.py ├── body_model_smpl.py ├── geo_transform.py ├── renderer.py ├── renderer_tools.py └── smpl_utils.py /.gitignore: -------------------------------------------------------------------------------- 1 | __*/ 2 | 3 | 4 | 5 | isaacgym/ 6 | isaac-gym-preview-4 7 | 8 | *.pdf 9 | *.csv 10 | *.pkl 11 | *.npz 12 | *.tar.gz 13 | 14 | 15 | 16 | # Compiled Python bytecode 17 | *.py[cod] 18 | # Log files 19 | *.log 20 | # * ONNX files 21 | # *.onnx 22 | output* 23 | *.pdf 24 | 25 | playground/robot_description 26 | 27 | 28 | *.csv 29 | extracted_motions* 30 | *.lprof 31 | 32 | # JetBrains IDE 33 | .idea/ 34 | .vscode/ 35 | 36 | logs*/ 37 | # Generated by MacOS 38 | .DS_Store 39 | 40 | lafan 41 | SharedMotions/ASAP-motion 42 | SharedMotions 43 | PaperMotions 44 | 45 | humanoidverse/data 46 | # Generated by Windows 47 | Thumbs.db 48 | 49 | logs_real_motion* 50 | *.zip 51 | 52 | # Applications 53 | *.app 54 | *.exe 55 | *.war 56 | 57 | # Large media files 58 | *.mp4 59 | *.tiff 60 | *.avi 61 | *.flv 62 | *.mov 63 | *.wmv 64 | 65 | # logs 66 | logs 67 | logs_eval 68 | runs 69 | outputs 70 | results/ 71 | hydra_logs/ 72 | wandb/ 73 | # other 74 | *.egg-info 75 | __pycache__ 76 | # unitree_sdk2_python 77 | roboverse/data/motions 78 | *.pkl 79 | *.pt 80 | 81 | *.TXT 82 | 83 | sim2real/models/*.onnx 84 | sim2real/models/dec_loco/*.onnx 85 | 86 | roboverse/data/smpl 87 | data/shape 88 | data/motions 89 | roboverse/data/Recorded/ 90 | roboverse/data/AMASS/ 91 | AMASS/* 92 | 93 | motion_source/lafan_pkl/ 94 | motion_source/lafan_pkl_cont_mask/ 95 | smpl_retarget/smpl_model/ 96 | smpl_retarget/motion_data/ 97 | smpl_retarget/retargeted_motion_data 98 | 99 | *.txt 100 | 101 | -------------------------------------------------------------------------------- /INSTALL.md: -------------------------------------------------------------------------------- 1 | 2 | ### Hardware Requirements 3 | 4 | We test the code in the following environment: 5 | - **OS**: Ubuntu 20.04 6 | - **GPU**: NVIDIA RTX 4090, Driver Version: 560.35.03 7 | - **CPU**: 13th Gen Intel(R) Core(TM) i7-13700 8 | 9 | 10 | 11 | ### Environment Setup 12 | ```bash 13 | # Assuming pwd: PBHC/ 14 | conda create -n pbhc python=3.8 15 | conda activate pbhc 16 | 17 | # Install and Test IsaacGym 18 | wget https://developer.nvidia.com/isaac-gym-preview-4 19 | tar -xvzf isaac-gym-preview-4 20 | pip install -e isaacgym/python 21 | cd isaacgym/python/examples 22 | python 1080_balls_of_solitude.py # or `python joint_monkey.py` 23 | cd ../../.. 24 | 25 | 26 | # Install PBHC 27 | pip install -e . 28 | pip install -e humanoidverse/isaac_utils 29 | 30 | ``` 31 | 32 | ```bash 33 | # (Optional) Install additional dependencies for motion visualization with rerun (robot_motion_process/vis_rr.py) 34 | pip install rerun-sdk==0.22.0 trimesh 35 | ``` 36 | 37 | -------------------------------------------------------------------------------- /demo.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/demo.jpg -------------------------------------------------------------------------------- /description/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/README.md -------------------------------------------------------------------------------- /description/robots/cfg/config.yaml: -------------------------------------------------------------------------------- 1 | defaults: 2 | - _self_ 3 | # - env: env_im 4 | - robot: smpl_humanoid 5 | # - learning: im 6 | # - sim: default_sim 7 | # - control: default_control 8 | # - domain_rand: default_dr 9 | 10 | project_name: "PHC" 11 | notes: "Default Notes" 12 | exp_name: &exp_name humanoid_smpl 13 | headless: True 14 | seed: 0 15 | no_log: False 16 | resume_str: null 17 | num_threads: 64 18 | test: False 19 | dataset: False 20 | mlp: False 21 | output_path: output/HumanoidIm/${exp_name} 22 | torch_deterministic: False 23 | epoch: 0 24 | im_eval: False 25 | horovod: False # Use horovod for multi-gpu training, have effect only with rl_games RL library 26 | rl_device: "cuda:0" 27 | device: "cuda" 28 | device_id: 0 29 | metadata: false 30 | play: ${test} 31 | train: True 32 | collect_dataset: False 33 | disable_multiprocessing: True 34 | 35 | 36 | ####### Testing Configs. ######## 37 | server_mode: False 38 | has_eval: True 39 | no_virtual_display: False 40 | render_o3d: False 41 | debug: False 42 | follow: False 43 | add_proj: False 44 | real_traj: False 45 | 46 | hydra: 47 | job: 48 | name: ${exp_name} 49 | env_set: 50 | OMP_NUM_THREADS: 1 51 | run: 52 | dir: output/HumanoidIm/${exp_name} 53 | 54 | -------------------------------------------------------------------------------- /description/robots/cfg/robot/unitree_g1_29dof_anneal_23dof.yaml: -------------------------------------------------------------------------------- 1 | humanoid_type: g1_29dof_anneal_23dof 2 | bias_offset: False 3 | has_self_collision: True 4 | has_mesh: False 5 | has_jt_limit: False 6 | has_dof_subset: True 7 | has_upright_start: True 8 | has_smpl_pd_offset: False 9 | remove_toe: False # For humanoid's geom toe 10 | motion_sym_loss: False 11 | sym_loss_coef: 1 12 | big_ankle: True 13 | 14 | has_shape_obs: false 15 | has_shape_obs_disc: false 16 | has_shape_variation: False 17 | 18 | masterfoot: False 19 | freeze_toe: false 20 | freeze_hand: False 21 | box_body: True 22 | real_weight: True 23 | real_weight_porpotion_capsules: True 24 | real_weight_porpotion_boxes: True 25 | 26 | body_names: [ 'pelvis', 27 | 'left_hip_pitch_link', 'left_hip_roll_link', 'left_hip_yaw_link', 'left_knee_link', 'left_ankle_pitch_link', 'left_ankle_roll_link', 28 | 'right_hip_pitch_link', 'right_hip_roll_link', 'right_hip_yaw_link', 'right_knee_link', 'right_ankle_pitch_link', 'right_ankle_roll_link', 29 | 'waist_yaw_link', 'waist_roll_link', 'torso_link', 30 | 'left_shoulder_pitch_link', 'left_shoulder_roll_link', 'left_shoulder_yaw_link', 'left_elbow_link', 31 | 'right_shoulder_pitch_link', 'right_shoulder_roll_link', 'right_shoulder_yaw_link', 'right_elbow_link' ] 32 | 33 | limb_weight_group: 34 | - [ 'left_hip_pitch_link', 'left_hip_roll_link', 'left_hip_yaw_link', 'left_knee_link', 'left_ankle_pitch_link', 'left_ankle_roll_link' ] 35 | - [ 'right_hip_pitch_link', 'right_hip_roll_link', 'right_hip_yaw_link', 'right_knee_link', 'right_ankle_pitch_link', 'right_ankle_roll_link' ] 36 | - [ 'pelvis', 'waist_yaw_link', 'waist_roll_link', 'torso_link' ] 37 | - [ 'left_shoulder_pitch_link', 'left_shoulder_roll_link', 'left_shoulder_yaw_link', 'left_elbow_link' ] 38 | - [ 'right_shoulder_pitch_link', 'right_shoulder_roll_link', 'right_shoulder_yaw_link', 'right_elbow_link' ] 39 | 40 | dof_names: [ 'left_hip_pitch_link', 'left_hip_roll_link', 'left_hip_yaw_link', 41 | 'left_knee_link', 'left_ankle_pitch_link', 'left_ankle_roll_link', 42 | 'right_hip_pitch_link', 'right_hip_roll_link', 'right_hip_yaw_link', 43 | 'right_knee_link', 'right_ankle_pitch_link', 'right_ankle_roll_link', 44 | 'waist_yaw_link', 'waist_roll_link', 'torso_link', 45 | 'left_shoulder_pitch_link', 'left_shoulder_roll_link', 'left_shoulder_yaw_link', 'left_elbow_link', 46 | 'right_shoulder_pitch_link', 'right_shoulder_roll_link', 'right_shoulder_yaw_link', 'right_elbow_link' ] 47 | 48 | right_foot_name: 'r_foot_roll' 49 | left_foot_name: 'l_foot_roll' 50 | 51 | asset: 52 | assetRoot: "./" 53 | assetFileName: "../description/robots/g1/g1_23dof_lock_wrist_fitmotionONLY.xml" 54 | urdfFileName: "../description/robots/g1/g1_23dof_lock_wrist_fitmotionONLY.xml" 55 | 56 | extend_config: 57 | - joint_name: "left_hand_link" 58 | parent_name: "left_elbow_link" 59 | pos: [ 0.25, 0.0, 0.0 ] 60 | rot: [ 1.0, 0.0, 0.0, 0.0 ] # w x y z 61 | - joint_name: "right_hand_link" 62 | parent_name: "right_elbow_link" 63 | pos: [ 0.25, 0.0, 0.0 ] 64 | rot: [ 1.0, 0.0, 0.0, 0.0 ] # w x y z 65 | - joint_name: "head_link" 66 | parent_name: "torso_link" 67 | pos: [ 0.0, 0.0, 0.42 ] 68 | rot: [ 1.0, 0.0, 0.0, 0.0 ] # w x y z 69 | - joint_name: "left_toe_link" 70 | parent_name: "left_ankle_roll_link" 71 | pos: [ 0.08, 0., -0.03 ] # 0.08 0 -0.06 # amass 0.04, -0.02, -0.02 72 | rot: [ 1.0, 0.0, 0.0, 0.0 ] # w x y z 73 | - joint_name: "right_toe_link" 74 | parent_name: "right_ankle_roll_link" 75 | pos: [ 0.08, 0., -0.03 ] # amass 0.04, -0.02, -0.02 76 | rot: [ 1.0, 0.0, 0.0, 0.0 ] # w x y z 77 | 78 | base_link: "torso_link" 79 | joint_matches: 80 | - [ "pelvis", "Pelvis" ] 81 | - [ "left_hip_pitch_link", "L_Hip" ] 82 | - [ "left_knee_link", "L_Knee" ] 83 | - [ "left_ankle_roll_link", "L_Ankle" ] 84 | # - [ "left_ankle_pitch_link", "L_Ankle" ] # new 85 | - [ "right_hip_pitch_link", "R_Hip" ] 86 | - [ "right_knee_link", "R_Knee" ] 87 | - [ "right_ankle_roll_link", "R_Ankle" ] 88 | # - [ "right_ankle_pitch_link", "R_Ankle" ] # new 89 | - [ "left_shoulder_roll_link", "L_Shoulder" ] 90 | - [ "left_elbow_link", "L_Elbow" ] 91 | - [ "left_hand_link", "L_Hand" ] 92 | - [ "right_shoulder_roll_link", "R_Shoulder" ] 93 | - [ "right_elbow_link", "R_Elbow" ] 94 | - [ "right_hand_link", "R_Hand" ] 95 | - [ "head_link", "Head" ] 96 | - [ "left_toe_link", "L_Toe" ] 97 | - [ "right_toe_link", "R_Toe" ] 98 | 99 | 100 | smpl_pose_modifier: 101 | - Pelvis: "[np.pi/2, 0, np.pi/2]" 102 | - L_Shoulder: "[0, 0, -np.pi/2]" 103 | - R_Shoulder: "[0, 0, np.pi/2]" 104 | - L_Elbow: "[0, -np.pi/2, 0]" 105 | - R_Elbow: "[0, np.pi/2, 0]" 106 | -------------------------------------------------------------------------------- /description/robots/g1/dof_axis.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/dof_axis.npy -------------------------------------------------------------------------------- /description/robots/g1/meshes/head_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/head_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_ankle_pitch_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_ankle_pitch_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_ankle_roll_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_ankle_roll_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_elbow_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_elbow_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_elbow_link_merge.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_elbow_link_merge.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_hand_index_0_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_hand_index_0_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_hand_index_1_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_hand_index_1_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_hand_middle_0_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_hand_middle_0_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_hand_middle_1_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_hand_middle_1_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_hand_palm_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_hand_palm_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_hand_thumb_0_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_hand_thumb_0_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_hand_thumb_1_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_hand_thumb_1_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_hand_thumb_2_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_hand_thumb_2_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_hip_pitch_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_hip_pitch_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_hip_roll_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_hip_roll_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_hip_yaw_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_hip_yaw_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_knee_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_knee_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_rubber_hand.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_rubber_hand.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_shoulder_pitch_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_shoulder_pitch_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_shoulder_roll_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_shoulder_roll_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_shoulder_yaw_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_shoulder_yaw_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_wrist_pitch_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_wrist_pitch_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_wrist_roll_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_wrist_roll_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_wrist_roll_rubber_hand.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_wrist_roll_rubber_hand.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/left_wrist_yaw_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/left_wrist_yaw_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/logo_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/logo_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/pelvis.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/pelvis.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/pelvis_contour_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/pelvis_contour_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_ankle_pitch_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_ankle_pitch_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_ankle_roll_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_ankle_roll_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_elbow_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_elbow_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_elbow_link_merge.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_elbow_link_merge.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_hand_index_0_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_hand_index_0_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_hand_index_1_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_hand_index_1_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_hand_middle_0_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_hand_middle_0_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_hand_middle_1_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_hand_middle_1_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_hand_palm_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_hand_palm_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_hand_thumb_0_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_hand_thumb_0_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_hand_thumb_1_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_hand_thumb_1_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_hand_thumb_2_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_hand_thumb_2_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_hip_pitch_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_hip_pitch_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_hip_roll_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_hip_roll_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_hip_yaw_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_hip_yaw_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_knee_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_knee_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_rubber_hand.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_rubber_hand.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_shoulder_pitch_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_shoulder_pitch_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_shoulder_roll_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_shoulder_roll_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_shoulder_yaw_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_shoulder_yaw_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_wrist_pitch_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_wrist_pitch_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_wrist_roll_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_wrist_roll_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_wrist_roll_rubber_hand.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_wrist_roll_rubber_hand.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/right_wrist_yaw_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/right_wrist_yaw_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/torso_constraint_L_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/torso_constraint_L_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/torso_constraint_L_rod_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/torso_constraint_L_rod_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/torso_constraint_R_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/torso_constraint_R_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/torso_constraint_R_rod_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/torso_constraint_R_rod_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/torso_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/torso_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/torso_link_23dof_rev_1_0.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/torso_link_23dof_rev_1_0.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/torso_link_rev_1_0.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/torso_link_rev_1_0.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/waist_constraint_L.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/waist_constraint_L.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/waist_constraint_R.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/waist_constraint_R.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/waist_roll_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/waist_roll_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/waist_roll_link_rev_1_0.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/waist_roll_link_rev_1_0.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/waist_support_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/waist_support_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/waist_yaw_link.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/waist_yaw_link.STL -------------------------------------------------------------------------------- /description/robots/g1/meshes/waist_yaw_link_rev_1_0.STL: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/description/robots/g1/meshes/waist_yaw_link_rev_1_0.STL -------------------------------------------------------------------------------- /description/robots/g1/phc_g1_23dof.yaml: -------------------------------------------------------------------------------- 1 | humanoid_type: g1_29dof_anneal_23dof 2 | bias_offset: False 3 | has_self_collision: True 4 | has_mesh: False 5 | has_jt_limit: False 6 | has_dof_subset: True 7 | has_upright_start: True 8 | has_smpl_pd_offset: False 9 | remove_toe: False # For humanoid's geom toe 10 | motion_sym_loss: False 11 | sym_loss_coef: 1 12 | big_ankle: True 13 | 14 | has_shape_obs: false 15 | has_shape_obs_disc: false 16 | has_shape_variation: False 17 | 18 | masterfoot: False 19 | freeze_toe: false 20 | freeze_hand: False 21 | box_body: True 22 | real_weight: True 23 | real_weight_porpotion_capsules: True 24 | real_weight_porpotion_boxes: True 25 | 26 | body_names: [ 'pelvis', 27 | 'left_hip_pitch_link', 'left_hip_roll_link', 'left_hip_yaw_link', 'left_knee_link', 'left_ankle_pitch_link', 'left_ankle_roll_link', 28 | 'right_hip_pitch_link', 'right_hip_roll_link', 'right_hip_yaw_link', 'right_knee_link', 'right_ankle_pitch_link', 'right_ankle_roll_link', 29 | 'waist_yaw_link', 'waist_roll_link', 'torso_link', 30 | 'left_shoulder_pitch_link', 'left_shoulder_roll_link', 'left_shoulder_yaw_link', 'left_elbow_link', 31 | 'right_shoulder_pitch_link', 'right_shoulder_roll_link', 'right_shoulder_yaw_link', 'right_elbow_link' ] 32 | 33 | limb_weight_group: 34 | - [ 'left_hip_pitch_link', 'left_hip_roll_link', 'left_hip_yaw_link', 'left_knee_link', 'left_ankle_pitch_link', 'left_ankle_roll_link' ] 35 | - [ 'right_hip_pitch_link', 'right_hip_roll_link', 'right_hip_yaw_link', 'right_knee_link', 'right_ankle_pitch_link', 'right_ankle_roll_link' ] 36 | - [ 'pelvis', 'waist_yaw_link', 'waist_roll_link', 'torso_link' ] 37 | - [ 'left_shoulder_pitch_link', 'left_shoulder_roll_link', 'left_shoulder_yaw_link', 'left_elbow_link' ] 38 | - [ 'right_shoulder_pitch_link', 'right_shoulder_roll_link', 'right_shoulder_yaw_link', 'right_elbow_link' ] 39 | 40 | dof_names: [ 'left_hip_pitch_link', 'left_hip_roll_link', 'left_hip_yaw_link', 41 | 'left_knee_link', 'left_ankle_pitch_link', 'left_ankle_roll_link', 42 | 'right_hip_pitch_link', 'right_hip_roll_link', 'right_hip_yaw_link', 43 | 'right_knee_link', 'right_ankle_pitch_link', 'right_ankle_roll_link', 44 | 'waist_yaw_link', 'waist_roll_link', 'torso_link', 45 | 'left_shoulder_pitch_link', 'left_shoulder_roll_link', 'left_shoulder_yaw_link', 'left_elbow_link', 46 | 'right_shoulder_pitch_link', 'right_shoulder_roll_link', 'right_shoulder_yaw_link', 'right_elbow_link' ] 47 | 48 | right_foot_name: 'r_foot_roll' 49 | left_foot_name: 'l_foot_roll' 50 | 51 | asset: 52 | assetRoot: "./" 53 | assetFileName: "description/robots/g1/g1_23dof_lock_wrist_fitmotionONLY.xml" 54 | urdfFileName: "description/robots/g1/g1_23dof_lock_wrist_fitmotionONLY.xml" 55 | 56 | extend_config: 57 | - joint_name: "left_hand_link" 58 | parent_name: "left_elbow_link" 59 | pos: [ 0.25, 0.0, 0.0 ] 60 | rot: [ 1.0, 0.0, 0.0, 0.0 ] # w x y z 61 | - joint_name: "right_hand_link" 62 | parent_name: "right_elbow_link" 63 | pos: [ 0.25, 0.0, 0.0 ] 64 | rot: [ 1.0, 0.0, 0.0, 0.0 ] # w x y z 65 | - joint_name: "head_link" 66 | parent_name: "torso_link" 67 | pos: [ 0.0, 0.0, 0.42 ] 68 | rot: [ 1.0, 0.0, 0.0, 0.0 ] # w x y z 69 | - joint_name: "left_toe_link" 70 | parent_name: "left_ankle_roll_link" 71 | pos: [ 0.08, 0, -0.06 ] # 0.08 0 -0.06 # amass 0.04, -0.02, -0.02 72 | rot: [ 1.0, 0.0, 0.0, 0.0 ] # w x y z 73 | - joint_name: "right_toe_link" 74 | parent_name: "right_ankle_roll_link" 75 | pos: [ 0.08, 0, -0.06 ] # amass 0.04, -0.02, -0.02 76 | rot: [ 1.0, 0.0, 0.0, 0.0 ] # w x y z 77 | 78 | base_link: "torso_link" 79 | joint_matches: 80 | - [ "pelvis", "Pelvis" ] 81 | - [ "left_hip_pitch_link", "L_Hip" ] 82 | - [ "left_knee_link", "L_Knee" ] 83 | - [ "left_ankle_roll_link", "L_Ankle" ] 84 | # - [ "left_ankle_pitch_link", "L_Ankle" ] # new 85 | - [ "right_hip_pitch_link", "R_Hip" ] 86 | - [ "right_knee_link", "R_Knee" ] 87 | - [ "right_ankle_roll_link", "R_Ankle" ] 88 | # - [ "right_ankle_pitch_link", "R_Ankle" ] # new 89 | - [ "left_shoulder_roll_link", "L_Shoulder" ] 90 | - [ "left_elbow_link", "L_Elbow" ] 91 | - [ "left_hand_link", "L_Hand" ] 92 | - [ "right_shoulder_roll_link", "R_Shoulder" ] 93 | - [ "right_elbow_link", "R_Elbow" ] 94 | - [ "right_hand_link", "R_Hand" ] 95 | - [ "head_link", "Head" ] 96 | - [ "left_toe_link", "L_Toe" ] 97 | - [ "right_toe_link", "R_Toe" ] 98 | 99 | 100 | smpl_pose_modifier: 101 | - Pelvis: "[np.pi/2, 0, np.pi/2]" 102 | - L_Shoulder: "[0, 0, -np.pi/2]" 103 | - R_Shoulder: "[0, 0, np.pi/2]" 104 | - L_Elbow: "[0, -np.pi/2, 0]" 105 | - R_Elbow: "[0, np.pi/2, 0]" 106 | -------------------------------------------------------------------------------- /example/motion_data/Bruce_Lee_pose.pkl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/example/motion_data/Bruce_Lee_pose.pkl -------------------------------------------------------------------------------- /example/motion_data/Charleston_dance.pkl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/example/motion_data/Charleston_dance.pkl -------------------------------------------------------------------------------- /example/motion_data/Hooks_punch.pkl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/example/motion_data/Hooks_punch.pkl -------------------------------------------------------------------------------- /example/motion_data/Horse-stance_pose.pkl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/example/motion_data/Horse-stance_pose.pkl -------------------------------------------------------------------------------- /example/motion_data/Horse-stance_punch.pkl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/example/motion_data/Horse-stance_punch.pkl -------------------------------------------------------------------------------- /example/motion_data/Roundhouse_kick.pkl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/example/motion_data/Roundhouse_kick.pkl -------------------------------------------------------------------------------- /example/motion_data/Side_kick.pkl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/example/motion_data/Side_kick.pkl -------------------------------------------------------------------------------- /example/pretrained_horse_stance_pose/exported/model_50000.onnx: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/example/pretrained_horse_stance_pose/exported/model_50000.onnx -------------------------------------------------------------------------------- /example/pretrained_horse_stance_pose/model_50000.pt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/example/pretrained_horse_stance_pose/model_50000.pt -------------------------------------------------------------------------------- /example/pretrained_horse_stance_pose_2/exported/model_119000.onnx: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/example/pretrained_horse_stance_pose_2/exported/model_119000.onnx -------------------------------------------------------------------------------- /example/pretrained_horse_stance_pose_2/model_119000.pt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/example/pretrained_horse_stance_pose_2/model_119000.pt -------------------------------------------------------------------------------- /example/pretrained_horse_stance_punch/exported/model_33000.onnx: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/example/pretrained_horse_stance_punch/exported/model_33000.onnx -------------------------------------------------------------------------------- /example/pretrained_horse_stance_punch/model_33000.pt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/example/pretrained_horse_stance_punch/model_33000.pt -------------------------------------------------------------------------------- /humanoidverse/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/humanoidverse/__init__.py -------------------------------------------------------------------------------- /humanoidverse/agents/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/humanoidverse/agents/__init__.py -------------------------------------------------------------------------------- /humanoidverse/agents/base_algo/base_algo.py: -------------------------------------------------------------------------------- 1 | import torch 2 | 3 | from torch import nn, Tensor 4 | 5 | import time 6 | import math 7 | from pathlib import Path 8 | from typing import Optional, List, Tuple, Dict 9 | 10 | 11 | from hydra.utils import instantiate 12 | 13 | from humanoidverse.envs.base_task.base_task import BaseTask 14 | 15 | class BaseAlgo: 16 | def __init__(self, env: BaseTask, config, device): 17 | self.env = env 18 | self.config = config 19 | self.device = device 20 | 21 | def setup(self): 22 | return NotImplementedError 23 | 24 | def learn(self): 25 | return NotImplementedError 26 | 27 | def load(self, path): 28 | return NotImplementedError 29 | 30 | @property 31 | def inference_model(self): 32 | return NotImplementedError 33 | 34 | def env_step(self, actions, extra_info=None): 35 | obs_dict, rewards, dones, extras = self.env.step(actions, extra_info) 36 | return obs_dict, rewards, dones, extras 37 | 38 | @torch.no_grad() 39 | def evaluate_policy(self): 40 | return NotImplementedError 41 | 42 | @torch.no_grad() 43 | def evaluate_policy_steps(self, Nsteps:int): 44 | return NotImplementedError 45 | 46 | def save(self, path=None, name="last.ckpt"): 47 | raise NotImplementedError 48 | -------------------------------------------------------------------------------- /humanoidverse/agents/callbacks/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/humanoidverse/agents/callbacks/__init__.py -------------------------------------------------------------------------------- /humanoidverse/agents/callbacks/analysis_plot_template.html: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 38 | 39 | 40 |
41 |

Robot State Plots

42 |
43 |

Loading plots...

44 |
45 |
46 | 52 | 53 | -------------------------------------------------------------------------------- /humanoidverse/agents/callbacks/base_callback.py: -------------------------------------------------------------------------------- 1 | import torch 2 | from torch.nn import Module 3 | 4 | class RL_EvalCallback(Module): 5 | def __init__(self, config, training_loop): 6 | super().__init__() 7 | self.config = config 8 | self.training_loop = training_loop 9 | self.device = self.training_loop.device 10 | 11 | def on_pre_evaluate_policy(self): 12 | pass 13 | 14 | def on_pre_eval_env_step(self, actor_state): 15 | return actor_state 16 | 17 | def on_post_eval_env_step(self, actor_state): 18 | return actor_state 19 | 20 | def on_post_evaluate_policy(self): 21 | pass 22 | -------------------------------------------------------------------------------- /humanoidverse/agents/modules/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/humanoidverse/agents/modules/__init__.py -------------------------------------------------------------------------------- /humanoidverse/agents/modules/encoder_modules.py: -------------------------------------------------------------------------------- 1 | from __future__ import annotations 2 | 3 | import torch 4 | import torch.nn as nn 5 | from .modules import BaseModule 6 | 7 | 8 | class Estimator(nn.Module): 9 | def __init__(self, obs_dim_dict, module_config_dict): 10 | super(Estimator, self).__init__() 11 | self.module = BaseModule(obs_dim_dict, module_config_dict) 12 | 13 | # def estimate(self, obs_history): 14 | # return self.module(obs_history) 15 | 16 | def forward(self, obs_history): 17 | return self.module(obs_history) -------------------------------------------------------------------------------- /humanoidverse/agents/modules/modules.py: -------------------------------------------------------------------------------- 1 | import torch 2 | import torch.nn as nn 3 | import inspect 4 | 5 | class BaseModule(nn.Module): 6 | def __init__(self, obs_dim_dict, module_config_dict): 7 | super(BaseModule, self).__init__() 8 | self.obs_dim_dict = obs_dim_dict 9 | self.module_config_dict = module_config_dict 10 | 11 | self._calculate_input_dim() 12 | self._calculate_output_dim() 13 | self._build_network_layer(self.module_config_dict.layer_config) 14 | 15 | def _calculate_input_dim(self): 16 | # calculate input dimension based on the input specifications 17 | input_dim = 0 18 | for each_input in self.module_config_dict['input_dim']: 19 | if each_input in self.obs_dim_dict: 20 | # atomic observation type 21 | input_dim += self.obs_dim_dict[each_input] 22 | elif isinstance(each_input, (int, float)): 23 | # direct numeric input 24 | input_dim += each_input 25 | else: 26 | current_function_name = inspect.currentframe().f_code.co_name 27 | raise ValueError(f"{current_function_name} - Unknown input type: {each_input}") 28 | 29 | self.input_dim = input_dim 30 | 31 | def _calculate_output_dim(self): 32 | output_dim = 0 33 | for each_output in self.module_config_dict['output_dim']: 34 | if isinstance(each_output, (int, float)): 35 | output_dim += each_output 36 | else: 37 | current_function_name = inspect.currentframe().f_code.co_name 38 | raise ValueError(f"{current_function_name} - Unknown output type: {each_output}") 39 | self.output_dim = output_dim 40 | 41 | def _build_network_layer(self, layer_config): 42 | if layer_config['type'] == 'MLP': 43 | self._build_mlp_layer(layer_config) 44 | else: 45 | raise NotImplementedError(f"Unsupported layer type: {layer_config['type']}") 46 | 47 | def _build_mlp_layer(self, layer_config): 48 | layers = [] 49 | hidden_dims = layer_config['hidden_dims'] 50 | output_dim = self.output_dim 51 | activation = getattr(nn, layer_config['activation'])() 52 | 53 | layers.append(nn.Linear(self.input_dim, hidden_dims[0])) 54 | layers.append(activation) 55 | 56 | for l in range(len(hidden_dims)): 57 | if l == len(hidden_dims) - 1: 58 | layers.append(nn.Linear(hidden_dims[l], output_dim)) 59 | else: 60 | layers.append(nn.Linear(hidden_dims[l], hidden_dims[l + 1])) 61 | layers.append(activation) 62 | 63 | self.module = nn.Sequential(*layers) 64 | 65 | def forward(self, input): 66 | return self.module(input) -------------------------------------------------------------------------------- /humanoidverse/config/algo/mh_ppo.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | algo: 4 | _target_: humanoidverse.agents.mh_ppo.mh_ppo.MHPPO 5 | _recursive_: False 6 | config: 7 | num_learning_epochs: 5 8 | num_mini_batches: 4 9 | clip_param: 0.2 10 | gamma: 0.99 11 | lam: 0.95 12 | value_loss_coef: 1.0 13 | entropy_coef: 0.01 14 | l2c2: 15 | enable: False 16 | lambda_value: 1.0 17 | lambda_policy: 0.1 18 | 19 | actor_learning_rate: 1.e-3 # 5e-4 # 1.e-3 20 | critic_learning_rate: 1.e-3 # 5e-4 # 1.e-3 21 | max_grad_norm: 1.0 22 | use_clipped_value_loss: True 23 | schedule: "adaptive" 24 | desired_kl: 0.01 25 | 26 | num_steps_per_env: 24 27 | save_interval: 1000 28 | logging_interval: 25 29 | 30 | load_optimizer: True 31 | 32 | init_noise_std: 0.8 33 | 34 | num_learning_iterations: 1000000 35 | init_at_random_ep_len: True 36 | eval_callbacks: null 37 | 38 | phase_embed: 39 | type: Original # Original, Sinusoidal, SinusoidalV2, Learnable. See 'humanoidverse/agents/modules/ppo_modules.py' for details 40 | dim: 16 # if type==Original, not used 41 | # Other Idea: additive, concat, multiplicative, etc. 42 | 43 | module_dict: 44 | actor: 45 | input_dim: [actor_obs] 46 | output_dim: [robot_action_dim] 47 | layer_config: 48 | type: MLP 49 | hidden_dims: [512, 256, 128] 50 | activation: ELU 51 | critic: 52 | type: MLP 53 | input_dim: [critic_obs] 54 | # output_dim: [1] 55 | output_dim: [num_rew_fn] 56 | layer_config: 57 | type: MLP 58 | hidden_dims: [768, 512, 128] 59 | activation: ELU -------------------------------------------------------------------------------- /humanoidverse/config/base.yaml: -------------------------------------------------------------------------------- 1 | # First we define the global structures that will be used by all the configs. 2 | defaults: 3 | # - base/fabric 4 | - _self_ 5 | - base/hydra 6 | - base/structure 7 | 8 | # These are global variables that all levels of the config can access. 9 | ## Experiment setup 10 | seed: 0 11 | codebase_version: 1.0 # this is recorded to enable auto-conversion of models between different versions of the codebase 12 | headless: True 13 | num_envs: 4096 14 | 15 | ### Checkpoint logic 16 | auto_load_latest: False 17 | checkpoint: null 18 | 19 | ### Naming and dir structure 20 | project_name: TEST 21 | experiment_name: TEST 22 | 23 | base_dir: logs 24 | timestamp: ${now:%Y%m%d_%H%M%S} 25 | experiment_dir: ${base_dir}/${project_name}/${timestamp}-${experiment_name}-${log_task_name}-${robot.asset.robot_type} 26 | save_dir: ${experiment_dir}/.hydra 27 | 28 | force_flat_terrain: False 29 | 30 | use_wandb: False 31 | log_task_name: TEST 32 | 33 | ### Simulation 34 | sim_type: isaacgym 35 | env_spacing: 20 36 | output_dir: ${experiment_dir}/output 37 | 38 | eval_overrides: 39 | headless: False 40 | num_envs: 1 41 | auto_load_latest: False 42 | use_wandb: False 43 | -------------------------------------------------------------------------------- /humanoidverse/config/base/fabric.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | fabric: 4 | _target_: lightning.fabric.Fabric 5 | _convert_: all 6 | accelerator: gpu 7 | devices: ${ngpu} 8 | num_nodes: ${nodes} 9 | strategy: 10 | _target_: lightning.fabric.strategies.DDPStrategy 11 | precision: ${amp_precision} 12 | loggers: 13 | - _target_: lightning.fabric.loggers.TensorBoardLogger 14 | root_dir: ${save_dir} 15 | 16 | # These are global variables that all levels of the config can access. 17 | ## Compute setup 18 | ngpu: 1 19 | nodes: 1 20 | torch_deterministic: False 21 | amp_precision: 32-true 22 | force_flat_terrain: False 23 | 24 | eval_overrides: 25 | ngpu: 1 26 | nodes: 1 27 | 28 | fabric: 29 | loggers: null 30 | -------------------------------------------------------------------------------- /humanoidverse/config/base/hydra.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | hydra: # So hydra will put your config info in the same dir as your checkpoints 4 | run: 5 | dir: ${save_dir} 6 | sweep: 7 | dir: ${save_dir} 8 | job: 9 | chdir: False 10 | # job_logging: 11 | # disable_existing_loggers: true 12 | # hydra_logging: 13 | # disable_existing_loggers: true 14 | -------------------------------------------------------------------------------- /humanoidverse/config/base/structure.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | algo: ??? 4 | 5 | env: ??? 6 | 7 | robot: ??? 8 | 9 | terrain: ??? # This is defined in the terrain configs 10 | -------------------------------------------------------------------------------- /humanoidverse/config/base_eval.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | eval_timestamp: ${now:%Y%m%d_%H%M%S} 4 | eval_name: TEST 5 | eval_log_dir: logs_eval/${eval_name}/${eval_timestamp} 6 | hydra: 7 | run: 8 | dir: ${eval_log_dir} 9 | -------------------------------------------------------------------------------- /humanoidverse/config/deploy/multiple.yaml: -------------------------------------------------------------------------------- 1 | 2 | 3 | deploy: 4 | deploy_mode: "multiple" 5 | 6 | robot: 7 | motion: 8 | motion_file: "example/motion_data" 9 | log_task_name: motion_tracking # just for compatibility, don't think about it 10 | 11 | 12 | obs: 13 | obs_dict: 14 | actor_obs: 15 | - base_ang_vel 16 | - projected_gravity 17 | - dof_pos 18 | - dof_vel 19 | - actions 20 | - ref_motion_phase 21 | - history_actor 22 | critic_obs: 23 | - base_lin_vel 24 | - base_ang_vel 25 | - projected_gravity 26 | - dof_pos 27 | - dof_vel 28 | - actions 29 | - ref_motion_phase 30 | - dif_local_rigid_body_pos 31 | - local_ref_rigid_body_pos 32 | - history_critic 33 | obs_auxiliary: 34 | history_actor: 35 | base_ang_vel: 4 36 | projected_gravity: 4 37 | dof_pos: 4 38 | dof_vel: 4 39 | actions: 4 40 | ref_motion_phase: 4 41 | relyaw: 4 42 | history_critic: 43 | base_lin_vel: 4 44 | base_ang_vel: 4 45 | projected_gravity: 4 46 | dof_pos: 4 47 | dof_vel: 4 48 | actions: 4 49 | ref_motion_phase: 4 50 | relyaw: 4 51 | short_history: 52 | base_ang_vel: 5 53 | projected_gravity: 5 54 | dof_pos: 5 55 | dof_vel: 5 56 | actions: 5 57 | command_lin_vel: 5 58 | command_ang_vel: 5 59 | obs_scales: 60 | base_lin_vel: 2.0 61 | base_ang_vel: 0.25 62 | projected_gravity: 1.0 63 | command_lin_vel: 1.0 64 | command_ang_vel: 1.0 65 | dof_pos: 1.0 66 | dof_vel: 0.05 67 | base_ang_vel_noise: 0.25 68 | projected_gravity_noise: 1.0 69 | dof_pos_noise: 1.0 70 | dof_vel_noise: 0.05 71 | history: 1.0 72 | short_history: 1.0 73 | actions: 1.0 74 | dif_joint_angles: 1.0 75 | dif_joint_velocities: 1.0 76 | # dif_joint_velocities: 0.05 77 | local_ref_rigid_body_vel: 1.0 78 | global_ref_rigid_body_vel: 1.0 79 | dif_local_rigid_body_pos: 1.0 80 | local_ref_rigid_body_pos: 1.0 81 | local_ref_rigid_body_pos_relyaw: 1.0 82 | relyaw: 1.0 83 | ref_motion_phase: 1.0 84 | history_actor: 1.0 85 | history_critic: 1.0 86 | dr_base_com: 1.0 87 | dr_link_mass: 1.0 88 | dr_kp: 1.0 89 | dr_kd: 1.0 90 | dr_friction: 1.0 91 | dr_ctrl_delay: 1.0 92 | noise_scales: 93 | base_lin_vel: 0.0 94 | base_ang_vel: 0.0 95 | projected_gravity: 0.0 96 | base_ang_vel_noise: 0.3 97 | projected_gravity_noise: 0.2 98 | dof_pos_noise: 0.01 99 | dof_vel_noise: 1.0 100 | command_lin_vel: 0.0 101 | command_ang_vel: 0.0 102 | dof_pos: 0.0 103 | dof_vel: 0.0 104 | actions: 0.0 105 | dif_joint_angles: 0.0 106 | dif_joint_velocities: 0.0 107 | local_ref_rigid_body_vel: 0.0 108 | global_ref_rigid_body_vel: 0.0 109 | dif_local_rigid_body_pos: 0.0 110 | local_ref_rigid_body_pos: 0.0 111 | local_ref_rigid_body_pos_relyaw: 0.0 112 | relyaw: 0.0 113 | ref_motion_phase: 0.0 114 | history_actor: 0.0 115 | history_critic: 0.0 116 | history: 0.0 117 | short_history: 0.0 118 | dr_base_com: 0.0 119 | dr_link_mass: 0.0 120 | dr_kp: 0.0 121 | dr_kd: 0.0 122 | dr_friction: 0.0 123 | dr_ctrl_delay: 0.0 124 | add_noise_currculum: false 125 | obs_dims: 126 | - base_lin_vel: 3 127 | - base_ang_vel: 3 128 | - projected_gravity: 3 129 | - command_lin_vel: 2 130 | - command_ang_vel: 1 131 | - dof_pos: ${robot.dof_obs_size} 132 | - dof_vel: ${robot.dof_obs_size} 133 | - base_ang_vel_noise: 3 134 | - projected_gravity_noise: 3 135 | - dof_pos_noise: ${robot.dof_obs_size} 136 | - dof_vel_noise: ${robot.dof_obs_size} 137 | - actions: ${robot.dof_obs_size} 138 | - dif_joint_angles: ${robot.dof_obs_size} 139 | - dif_joint_velocities: ${robot.dof_obs_size} 140 | - dif_local_rigid_body_pos: ${eval:'3 * ${robot.num_bodies} + 9'} 141 | - local_ref_rigid_body_pos: ${eval:'3 * ${robot.num_bodies} + 9'} 142 | - local_ref_rigid_body_vel: ${eval:'3 * ${robot.num_bodies} + 9'} 143 | - global_ref_rigid_body_vel: ${eval:'3 * ${robot.num_bodies} + 9'} 144 | - local_ref_rigid_body_pos_relyaw: ${eval:'3 * ${robot.num_bodies} + 9'} 145 | - relyaw: 1 146 | - ref_motion_phase: 1 147 | - dr_base_com: 3 148 | - dr_link_mass: 22 149 | - dr_kp: ${robot.dof_obs_size} 150 | - dr_kd: ${robot.dof_obs_size} 151 | - dr_friction: 1 152 | - dr_ctrl_delay: 1 153 | post_compute_config: {} 154 | -------------------------------------------------------------------------------- /humanoidverse/config/deploy/single.yaml: -------------------------------------------------------------------------------- 1 | 2 | deploy: 3 | deploy_mode: "single" 4 | 5 | BYPASS_ACT: False 6 | SWITCH_EMA: True 7 | 8 | render: True 9 | defcmd: [0.0, 0.0, 0.0, 0.0] 10 | heading_cmd: True 11 | 12 | ctrl_dt: 0.02 13 | net: ??? 14 | msg_type: "hg" # "hg" or "go", currently only "hg" is supported 15 | imu_type: "pelvis" # "torso" or "pelvis", currently only "pelvis" is supported 16 | mode_machine: 5 17 | 18 | lowcmd_topic: "rt/lowcmd" 19 | lowstate_topic: "rt/lowstate" 20 | # lowcmd_topic: "low_cmd_topic" 21 | # lowstate_topic: "low_state_topic" 22 | 23 | dof_idx_23_to_29: [ 0, 1, 2, 3, 4, 5, 24 | 6, 7, 8, 9, 10, 11, 25 | 12,13,14, 26 | 15,16,17,18, 27 | 22,23,24,25,] 28 | locked_kp: 40 29 | locked_kd: 1 30 | 31 | # leg_joint2motor_idx: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11] 32 | # arm_waist_joint2motor_idx: [12, 13, 14, 33 | # 15, 16, 17, 18, 19, 20, 21, 34 | # 22, 23, 24, 25, 26, 27, 28] 35 | 36 | 37 | env: 38 | config: 39 | save_motion: True 40 | save_note: null 41 | -------------------------------------------------------------------------------- /humanoidverse/config/domain_rand/NO_domain_rand.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | 4 | defaults: 5 | - domain_rand_base 6 | 7 | domain_rand: 8 | push_robots : False 9 | push_interval_s : [5, 10] 10 | max_push_vel_xy : 1.0 11 | 12 | randomize_base_com : False 13 | base_com_range: #kg 14 | x : [-0.1, 0.1] 15 | y : [-0.1, 0.1] 16 | z : [-0.01, 0.01] 17 | 18 | randomize_link_mass : False 19 | link_mass_range : [0.7, 1.3] # *factor 20 | randomize_link_body_names : [ 21 | 'pelvis', 'left_hip_yaw_link', 'left_hip_roll_link', 'left_hip_pitch_link', 'left_knee_link', 22 | 'right_hip_yaw_link', 'right_hip_roll_link', 'right_hip_pitch_link', 'right_knee_link', 23 | ] 24 | 25 | randomize_pd_gain : False 26 | kp_range : [0.75, 1.25] 27 | kd_range : [0.75, 1.25] 28 | 29 | 30 | 31 | randomize_friction : False 32 | # randomize_friction : False 33 | friction_range : [-0.6, 1.2] 34 | 35 | randomize_base_mass : False # replaced by randomize_link_mass 36 | added_mass_range : [-5., 10.] 37 | 38 | randomize_torque_rfi : False 39 | rfi_lim : 0.1 40 | randomize_rfi_lim : False 41 | rfi_lim_range : [0.5, 1.5] 42 | 43 | randomize_ctrl_delay : False 44 | ctrl_delay_step_range : [0, 2] # integer max real delay is 90ms 45 | 46 | randomize_motion_ref_xyz: False # head only for now 47 | motion_ref_xyz_range : [[-0.02, 0.02],[-0.02, 0.02],[-0.1, 0.1]] 48 | 49 | motion_package_loss: False 50 | package_loss_range: [1, 10] # dt = 0.02s, delay for 0.02s - 0.2s 51 | package_loss_interval_s : 2 52 | 53 | 54 | born_offset : False 55 | born_offset_curriculum: False 56 | born_offset_level_down_threshold: 50 57 | born_offset_level_up_threshold: 120 58 | level_degree: 0.00005 59 | born_distance : 0.25 60 | born_offset_range: [0.0, 1] 61 | born_offset_possibility : 1.0 62 | 63 | born_heading_curriculum: False 64 | born_heading_randomization : False 65 | born_heading_level_down_threshold: 50 66 | born_heading_level_up_threshold: 120 67 | born_heading_degree: 10 68 | born_heading_range: [0, 180] 69 | born_heading_level_degree: 0.00005 70 | -------------------------------------------------------------------------------- /humanoidverse/config/domain_rand/domain_rand_base.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | domain_rand: 4 | # push robots 5 | push_robots : True 6 | push_interval_s : [5, 10] 7 | max_push_vel_xy : 1.0 8 | 9 | # base com 10 | randomize_base_com : True 11 | base_com_range: 12 | x : [-0.1., 0.1.] 13 | y : [-0.1., 0.1.] 14 | z : [-0.1, 0.1] 15 | 16 | # link mass 17 | randomize_link_mass : True 18 | link_mass_range : [0.8, 1.2] 19 | 20 | # pd gain 21 | randomize_pd_gain : True 22 | kp_range : [0.75, 1.25] 23 | kd_range : [0.75, 1.25] 24 | 25 | # friction 26 | randomize_friction : True 27 | friction_range : [0.5, 1.25] 28 | 29 | # base mass 30 | randomize_base_mass : False 31 | 32 | # rfi 33 | randomize_torque_rfi : True 34 | rfi_lim : 0.1 35 | randomize_rfi_lim : True 36 | rfi_lim_range : [0.5, 1.5] 37 | 38 | 39 | # control delay 40 | randomize_ctrl_delay : True 41 | ctrl_delay_step_range : [0, 2] # integer max real delay is 90ms -------------------------------------------------------------------------------- /humanoidverse/config/domain_rand/dr_nil.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | # Small version DR, modified from redr1l_inertia_all 3 | 4 | defaults: 5 | - domain_rand_base 6 | 7 | domain_rand: 8 | _push_fixed: true 9 | push_robots : True 10 | push_interval_s : [5, 10] 11 | max_push_vel_xy : 0.001 12 | 13 | randomize_base_com : True 14 | base_com_range: #kg 15 | x : [-0.001,0.001] 16 | y : [-0.001,0.001] 17 | z : [-0.001,0.001] 18 | 19 | randomize_link_mass : True 20 | link_mass_range : [0.999, 1.001] # *factor 21 | randomize_link_body_names : [ 22 | 'pelvis', 'torso_link', 23 | 'left_shoulder_pitch_link', 'left_shoulder_roll_link', 'left_shoulder_yaw_link', 'left_elbow_link', 24 | 'right_shoulder_pitch_link', 'right_shoulder_roll_link', 'right_shoulder_yaw_link', 'right_elbow_link', 25 | 'left_hip_yaw_link', 'left_hip_roll_link', 'left_hip_pitch_link', 'left_knee_link', 'left_ankle_pitch_link', 'left_ankle_roll_link', 26 | 'right_hip_yaw_link', 'right_hip_roll_link', 'right_hip_pitch_link', 'right_knee_link', 'right_ankle_pitch_link', 'right_ankle_roll_link' 27 | ] 28 | 29 | randomize_pd_gain : True 30 | kp_range : [0.999, 1.001] 31 | kd_range : [0.999, 1.001] 32 | 33 | 34 | 35 | randomize_friction : True 36 | # randomize_friction : False 37 | friction_range : [0.4, 1.2] 38 | 39 | randomize_base_mass : False # replaced by randomize_link_mass 40 | added_mass_range : [-5., 10.] 41 | 42 | randomize_torque_rfi : True 43 | use_rao : True 44 | rao_lim : 0.001 45 | rfi_lim : 0.001 46 | randomize_rfi_lim : True 47 | rfi_lim_range : [0.0005, 0.0015] 48 | 49 | randomize_ctrl_delay : True 50 | ctrl_delay_step_range : [0, 0] # integer max real delay is 90ms 51 | 52 | randomize_motion_ref_xyz: False # head only for now 53 | motion_ref_xyz_range : [[-0.02, 0.02],[-0.02, 0.02],[-0.1, 0.1]] 54 | 55 | motion_package_loss: False 56 | package_loss_range: [1, 10] # dt = 0.02s, delay for 0.02s - 0.2s 57 | package_loss_interval_s : 2 58 | 59 | 60 | born_offset : False 61 | born_offset_curriculum: False 62 | born_offset_level_down_threshold: 50 63 | born_offset_level_up_threshold: 120 64 | level_degree: 0.00005 65 | born_distance : 0.25 66 | born_offset_range: [0.0, 1] 67 | born_offset_possibility : 1.0 68 | 69 | born_heading_curriculum: False 70 | born_heading_randomization : False 71 | born_heading_level_down_threshold: 50 72 | born_heading_level_up_threshold: 120 73 | born_heading_degree: 10 74 | born_heading_range: [0, 180] 75 | born_heading_level_degree: 0.00005 76 | -------------------------------------------------------------------------------- /humanoidverse/config/domain_rand/main.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | # Modified from `redr1l.yaml` 3 | 4 | defaults: 5 | - domain_rand_base 6 | 7 | domain_rand: 8 | reinit_epis_rand: -1 9 | 10 | _push_fixed: true 11 | push_robots : True 12 | push_interval_s : [5, 10] 13 | max_push_vel_xy : 0.1 14 | 15 | randomize_base_com : True 16 | base_com_range: #kg 17 | x : [-0.05, 0.05] 18 | y : [-0.05, 0.05] 19 | z : [-0.01, 0.01] 20 | 21 | randomize_link_mass : True 22 | link_mass_range : [0.9, 1.1] # *factor 23 | randomize_link_com : False 24 | link_com_range: #m 25 | x : [-0.01, 0.01] 26 | y : [-0.01, 0.01] 27 | z : [-0.01, 0.01] 28 | randomize_link_body_names : [ 29 | 'pelvis', 'torso_link', 30 | 'left_shoulder_pitch_link', 'left_shoulder_roll_link', 'left_shoulder_yaw_link', 'left_elbow_link', 31 | 'right_shoulder_pitch_link', 'right_shoulder_roll_link', 'right_shoulder_yaw_link', 'right_elbow_link', 32 | 'left_hip_yaw_link', 'left_hip_roll_link', 'left_hip_pitch_link', 'left_knee_link', 'left_ankle_pitch_link', 'left_ankle_roll_link', 33 | 'right_hip_yaw_link', 'right_hip_roll_link', 'right_hip_pitch_link', 'right_knee_link', 'right_ankle_pitch_link', 'right_ankle_roll_link' 34 | ] 35 | 36 | randomize_link_inertia : True 37 | link_inertia_range : [0.9,1.1] # *factor 38 | randomize_link_inertia_names : [ 39 | 'pelvis', 'torso_link', 40 | 'left_shoulder_pitch_link', 'left_shoulder_roll_link', 'left_shoulder_yaw_link', 'left_elbow_link', 41 | 'right_shoulder_pitch_link', 'right_shoulder_roll_link', 'right_shoulder_yaw_link', 'right_elbow_link', 42 | 'left_hip_yaw_link', 'left_hip_roll_link', 'left_hip_pitch_link', 'left_knee_link', 'left_ankle_pitch_link', 'left_ankle_roll_link', 43 | 'right_hip_yaw_link', 'right_hip_roll_link', 'right_hip_pitch_link', 'right_knee_link', 'right_ankle_pitch_link', 'right_ankle_roll_link' 44 | ] 45 | 46 | heavy_upper: 47 | enable: False 48 | ratio: 1.1 # all mass and inertia are scaled by this ratio 49 | body_names: [ 50 | 'torso_link', 51 | 'left_shoulder_pitch_link', 'left_shoulder_roll_link', 'left_shoulder_yaw_link', 'left_elbow_link', 52 | 'right_shoulder_pitch_link', 'right_shoulder_roll_link', 'right_shoulder_yaw_link', 'right_elbow_link', 53 | ] 54 | randomize_pd_gain : True 55 | kp_range : [0.9, 1.1] 56 | kd_range : [0.9, 1.1] 57 | 58 | parallel_serial_pd: 59 | # multiplicative composition with randomize_pd_gain, not replacing it 60 | enable: False 61 | ratio: [0.8, 1.2] 62 | joint_idx: [4,5, 10,11, 13,14] 63 | 64 | 65 | randomize_friction : True 66 | # randomize_friction : False 67 | friction_range : [0.2, 1.2] 68 | randomize_restitution : False 69 | restitution_range : [0.0, 1.0] 70 | 71 | randomize_base_mass : False # replaced by randomize_link_mass 72 | added_mass_range : [-5., 10.] 73 | 74 | randomize_torque_rfi : True 75 | use_rao : True 76 | rao_lim : 0.05 77 | rfi_lim : 0.05 78 | randomize_rfi_lim : True 79 | rfi_lim_range : [0.5, 1.5] 80 | 81 | randomize_ctrl_delay : True 82 | ctrl_delay_step_range : [0, 2] # integer max real delay is 90ms 83 | 84 | randomize_motion_ref_xyz: False # head only for now 85 | motion_ref_xyz_range : [[-0.02, 0.02],[-0.02, 0.02],[-0.1, 0.1]] 86 | 87 | motion_package_loss: False 88 | package_loss_range: [1, 10] # dt = 0.02s, delay for 0.02s - 0.2s 89 | package_loss_interval_s : 2 90 | 91 | 92 | born_offset : False 93 | born_offset_curriculum: False 94 | born_offset_level_down_threshold: 50 95 | born_offset_level_up_threshold: 120 96 | level_degree: 0.00005 97 | born_distance : 0.25 98 | born_offset_range: [0.0, 1] 99 | born_offset_possibility : 1.0 100 | 101 | born_heading_curriculum: False 102 | born_heading_randomization : False 103 | born_heading_level_down_threshold: 50 104 | born_heading_level_up_threshold: 120 105 | born_heading_degree: 10 106 | born_heading_range: [0, 180] 107 | born_heading_level_degree: 0.00005 108 | -------------------------------------------------------------------------------- /humanoidverse/config/env/base_task.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | env: 4 | _target_: humanoidverse.envs.base_task.base_task.BaseTask 5 | _recursive_: False 6 | config: 7 | experiment_name: ${experiment_name} 8 | num_envs: ${num_envs} 9 | headless: ${headless} 10 | simulator: ${simulator} 11 | save_rendering_dir: null 12 | ckpt_dir: null 13 | 14 | # Globally accessible parameters 15 | log_task_name: base_task -------------------------------------------------------------------------------- /humanoidverse/config/env/legged_base.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | defaults: 4 | - base_task 5 | 6 | # Env structure 7 | env: 8 | _target_: humanoidverse.envs.legged_base_task.legged_robot_base.LeggedRobotBase 9 | _recursive_: False 10 | config: 11 | robot: ${robot} 12 | domain_rand: ${domain_rand} 13 | rewards: ${rewards} 14 | terrain: ${terrain} 15 | obs: ${obs} 16 | env_spacing: 5.0 17 | max_episode_length_s: 100000 18 | use_vec_reward: False 19 | 20 | normalization: 21 | clip_actions: 100.0 22 | clip_observations: 100.0 23 | 24 | # simulator: 25 | # sim: 26 | # fps: 200 # 1/dt , dt = 0.005 27 | # control_decimation: 4 # decimation 28 | # substeps: 1 29 | termination: 30 | terminate_when_close_to_dof_pos_limit: False 31 | terminate_when_close_to_dof_vel_limit: False 32 | terminate_when_close_to_torque_limit: False 33 | termination_scales: 34 | termination_close_to_dof_pos_limit : 0.98 35 | termination_close_to_dof_vel_limit : 0.98 36 | termination_close_to_torque_limit : 0.98 37 | 38 | # Globally accessible parameters 39 | log_task_name: legged_base -------------------------------------------------------------------------------- /humanoidverse/config/env/motion_tracking.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | defaults: 4 | - legged_base 5 | 6 | # Env structure 7 | env: 8 | _target_: humanoidverse.envs.motion_tracking.motion_tracking.LeggedRobotMotionTracking 9 | config: 10 | use_vec_reward: True 11 | max_episode_length_s: 100000 12 | termination: 13 | terminate_by_contact: False 14 | terminate_by_gravity: True 15 | terminate_by_low_height: False 16 | terminate_when_motion_end: True 17 | terminate_when_motion_far: True 18 | terminate_when_dof_far: False 19 | terminate_when_close_to_dof_pos_limit: False 20 | terminate_when_close_to_dof_vel_limit: False 21 | terminate_when_close_to_torque_limit: False 22 | termination_scales: 23 | termination_min_base_height : 0.2 24 | termination_gravity_x : 0.8 25 | termination_gravity_y : 0.8 26 | termination_gravity : 0.75 27 | termination_motion_far_threshold : 1.5 28 | termination_close_to_dof_pos_limit : 0.98 29 | termination_close_to_dof_vel_limit : 0.98 30 | termination_close_to_torque_limit : 0.98 31 | termination_probality: 32 | terminate_when_close_to_dof_pos_limit: 0.25 33 | terminate_when_close_to_dof_vel_limit: 0.25 34 | terminate_when_close_to_torque_limit: 0.25 35 | termination_curriculum: 36 | terminate_when_motion_far_curriculum: True 37 | terminate_when_motion_far_initial_threshold: 1.5 38 | terminate_when_motion_far_threshold_max: 2.0 39 | terminate_when_motion_far_threshold_min: 0.3 40 | terminate_when_motion_far_curriculum_degree: 0.000025 41 | terminate_when_motion_far_curriculum_level_down_threshold: 40 42 | terminate_when_motion_far_curriculum_level_up_threshold: 42 43 | 44 | terminate_when_dof_far_curriculum: 45 | enable: True 46 | init: 3.0 47 | max: 3.0 48 | min: 1.0 49 | degree: 0.000002 50 | level_down_threshold: 40 51 | level_up_threshold: 42 52 | 53 | resample_motion_when_training: False 54 | resample_time_interval_s: 2000 55 | 56 | noise_to_initial_level: 0 57 | init_noise_scale: 58 | dof_pos: 0.1 59 | dof_vel: 0.15 60 | root_pos: 0.05 61 | root_rot: 10 # degree, * 3.14 / 180 62 | root_vel: 0.01 63 | root_ang_vel: 0.01 64 | 65 | use_teleop_control: False 66 | enforce_randomize_motion_start_eval: False 67 | 68 | 69 | soft_dynamic_correction: 70 | enable: False 71 | alpha: 0.1 72 | type: "deter" # "prob" or "deter" 73 | curriculum: 74 | enable: True 75 | max_alpha: 0.9 76 | min_alpha: 0.0 77 | degree: 0.00001 78 | level_down_threshold: 40 79 | level_up_threshold: 42 80 | 81 | 82 | 83 | # Globally accessible parameters 84 | log_task_name: motion_tracking 85 | 86 | eval_overrides: 87 | env: 88 | config: 89 | max_episode_length_s: 100000 90 | -------------------------------------------------------------------------------- /humanoidverse/config/exp/base_exp.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | defaults: 4 | - /algo: base_algo 5 | - /env: base_task 6 | 7 | experiment_name: TESTBaseExp -------------------------------------------------------------------------------- /humanoidverse/config/exp/legged_base.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | defaults: 4 | - /algo: ppo 5 | - /env: legged_base 6 | 7 | experiment_name: TESTLeggedBase -------------------------------------------------------------------------------- /humanoidverse/config/exp/motion_tracking.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | defaults: 4 | # - /algo: ppo 5 | - /algo: mh_ppo 6 | - /env: motion_tracking 7 | 8 | experiment_name: TEST_Motion_Tracking -------------------------------------------------------------------------------- /humanoidverse/config/obs/legged_obs.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | # Be careful when using _raw, history 4 | obs: 5 | obs_dict: 6 | actor_obs: ??? 7 | critic_obs: ??? 8 | 9 | # define those coumpounds in obs_dict, for example, you can define different long/short history with different length 10 | obs_auxiliary: ??? 11 | 12 | obs_scales: ??? 13 | 14 | noise_scales: ??? 15 | 16 | obs_dims: ??? 17 | # Will store values after preprocessing, don't put anything here 18 | post_compute_config: {} 19 | motion_len: -1 -------------------------------------------------------------------------------- /humanoidverse/config/obs/motion_tracking/benchmark.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | # Modified from 'deepmimic_a2c_nolinvel_LARGEnoise_history.yaml', add priv information to critic 4 | 5 | # Be careful when using _raw, history 6 | obs: 7 | obs_dict: 8 | actor_obs: [ 9 | base_lin_vel, 10 | base_ang_vel, 11 | projected_gravity, 12 | dof_pos, 13 | dof_vel, 14 | actions, 15 | ref_motion_phase, 16 | dif_local_rigid_body_pos, 17 | local_ref_rigid_body_pos, 18 | history_actor 19 | ] 20 | 21 | critic_obs: [ 22 | base_lin_vel, 23 | base_ang_vel, 24 | projected_gravity, 25 | dof_pos, 26 | dof_vel, 27 | actions, 28 | ref_motion_phase, 29 | dif_local_rigid_body_pos, 30 | local_ref_rigid_body_pos, 31 | dr_base_com, 32 | dr_link_mass, 33 | dr_kp, 34 | dr_kd, 35 | dr_friction, 36 | dr_ctrl_delay, 37 | history_critic 38 | ] 39 | 40 | obs_auxiliary: # define those coumpounds in obs_dict, for example, you can define different long/short history with different length 41 | 42 | history_actor: { 43 | base_lin_vel: 4, 44 | base_ang_vel: 4, 45 | projected_gravity: 4, 46 | dof_pos: 4, 47 | dof_vel: 4, 48 | actions: 4, 49 | ref_motion_phase: 4 50 | } 51 | 52 | history_critic: { 53 | base_lin_vel: 4, 54 | base_ang_vel: 4, 55 | projected_gravity: 4, 56 | dof_pos: 4, 57 | dof_vel: 4, 58 | actions: 4, 59 | ref_motion_phase: 4 60 | } 61 | 62 | obs_scales: { 63 | base_lin_vel: 2.0, 64 | base_ang_vel: 0.25, 65 | projected_gravity: 1.0, 66 | dof_pos: 1.0, 67 | dof_vel: 0.05, 68 | actions: 1.0, 69 | dif_local_rigid_body_pos: 1.0, 70 | local_ref_rigid_body_pos: 1.0, 71 | ref_motion_phase: 1.0, 72 | history_actor: 1.0, 73 | history_critic: 1.0, 74 | dr_base_com: 1.0, 75 | dr_link_mass: 1.0, 76 | dr_kp: 1.0, 77 | dr_kd: 1.0, 78 | dr_friction: 1.0, 79 | dr_ctrl_delay: 1.0, 80 | } 81 | 82 | noise_scales: { 83 | base_lin_vel: 0.0, 84 | base_ang_vel: 0.0, 85 | projected_gravity: 0.0, 86 | dof_pos: 0.00, 87 | dof_vel: 0.0, 88 | actions: 0.0, 89 | dif_local_rigid_body_pos: 0.0, 90 | local_ref_rigid_body_pos: 0.0, 91 | ref_motion_phase: 0.0, 92 | history_actor: 0.0, 93 | history_critic: 0.0, 94 | dr_base_com: 0.0, 95 | dr_link_mass: 0.0, 96 | dr_kp: 0.0, 97 | dr_kd: 0.0, 98 | dr_friction: 0.0, 99 | dr_ctrl_delay: 0.0, 100 | } 101 | 102 | add_noise_currculum: False 103 | noise_initial_value: 0.05 104 | noise_value_max: 1.0 105 | noise_value_min: 0.00001 106 | soft_dof_pos_curriculum_degree: 0.00001 107 | soft_dof_pos_curriculum_level_down_threshold: 40 108 | soft_dof_pos_curriculum_level_up_threshold: 42 109 | 110 | obs_dims: 111 | - base_lin_vel: 3 112 | - base_ang_vel: 3 113 | - projected_gravity: 3 114 | - dof_pos: ${robot.dof_obs_size} 115 | - dof_vel: ${robot.dof_obs_size} 116 | - actions: ${robot.dof_obs_size} 117 | - dif_local_rigid_body_pos: ${eval:'3 * ${robot.num_bodies} + 9'} # hardcoded for 3 extended body 118 | - local_ref_rigid_body_pos: ${eval:'3 * ${robot.num_bodies} + 9'} 119 | - ref_motion_phase: 1 120 | - dr_base_com: 3 121 | - dr_link_mass: 22 # = num of 'randomize_link_body_names' 122 | - dr_kp: ${robot.dof_obs_size} 123 | - dr_kd: ${robot.dof_obs_size} 124 | - dr_friction: 1 125 | - dr_ctrl_delay: 1 126 | 127 | post_compute_config: {} # Will store values after preprocessing, don't put anything here 128 | motion_len: -1 -------------------------------------------------------------------------------- /humanoidverse/config/obs/motion_tracking/main.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | # Modified from 'deepmimic_a2c_nolinvel_LARGEnoise_history.yaml', add priv information to critic 4 | 5 | # Be careful when using _raw, history 6 | obs: 7 | obs_dict: 8 | actor_obs: [ 9 | base_ang_vel, 10 | projected_gravity, 11 | dof_pos, 12 | dof_vel, 13 | actions, 14 | ref_motion_phase, 15 | # dif_local_rigid_body_pos, 16 | # local_ref_rigid_body_pos, 17 | # history, 18 | history_actor 19 | ] 20 | 21 | critic_obs: [ 22 | base_lin_vel, 23 | base_ang_vel, 24 | projected_gravity, 25 | dof_pos, 26 | dof_vel, 27 | actions, 28 | ref_motion_phase, 29 | dif_local_rigid_body_pos, 30 | local_ref_rigid_body_pos, 31 | dr_base_com, 32 | dr_link_mass, 33 | dr_kp, 34 | dr_kd, 35 | dr_friction, 36 | dr_ctrl_delay, 37 | history_critic 38 | ] 39 | 40 | obs_auxiliary: # define those coumpounds in obs_dict, for example, you can define different long/short history with different length 41 | 42 | history_actor: { 43 | base_ang_vel: 4, 44 | projected_gravity: 4, 45 | dof_pos: 4, 46 | dof_vel: 4, 47 | actions: 4, 48 | ref_motion_phase: 4 49 | } 50 | 51 | history_critic: { 52 | base_lin_vel: 4, 53 | base_ang_vel: 4, 54 | projected_gravity: 4, 55 | dof_pos: 4, 56 | dof_vel: 4, 57 | actions: 4, 58 | ref_motion_phase: 4 59 | } 60 | 61 | obs_scales: { 62 | base_lin_vel: 2.0, 63 | base_ang_vel: 0.25, 64 | projected_gravity: 1.0, 65 | dof_pos: 1.0, 66 | dof_vel: 0.05, 67 | actions: 1.0, 68 | dif_local_rigid_body_pos: 1.0, 69 | local_ref_rigid_body_pos: 1.0, 70 | ref_motion_phase: 1.0, 71 | history_actor: 1.0, 72 | history_critic: 1.0, 73 | dr_base_com: 1.0, 74 | dr_link_mass: 1.0, 75 | dr_kp: 1.0, 76 | dr_kd: 1.0, 77 | dr_friction: 1.0, 78 | dr_ctrl_delay: 1.0, 79 | } 80 | 81 | noise_scales: { 82 | base_lin_vel: 0.0, 83 | base_ang_vel: 0.3, 84 | projected_gravity: 0.2, 85 | dof_pos: 0.01, 86 | dof_vel: 1.0, 87 | actions: 0.0, 88 | dif_local_rigid_body_pos: 0.0, 89 | local_ref_rigid_body_pos: 0.0, 90 | ref_motion_phase: 0.0, 91 | history_actor: 0.0, 92 | history_critic: 0.0, 93 | dr_base_com: 0.0, 94 | dr_link_mass: 0.0, 95 | dr_kp: 0.0, 96 | dr_kd: 0.0, 97 | dr_friction: 0.0, 98 | dr_ctrl_delay: 0.0, 99 | } 100 | 101 | add_noise_currculum: False 102 | noise_initial_value: 0.05 103 | noise_value_max: 1.0 104 | noise_value_min: 0.00001 105 | soft_dof_pos_curriculum_degree: 0.00001 106 | soft_dof_pos_curriculum_level_down_threshold: 40 107 | soft_dof_pos_curriculum_level_up_threshold: 42 108 | 109 | obs_dims: 110 | - base_lin_vel: 3 111 | - base_ang_vel: 3 112 | - projected_gravity: 3 113 | - dof_pos: ${robot.dof_obs_size} 114 | - dof_vel: ${robot.dof_obs_size} 115 | - actions: ${robot.dof_obs_size} 116 | - dif_local_rigid_body_pos: ${eval:'3 * ${robot.num_bodies} + 9'} # hardcoded for 3 extended body 117 | - local_ref_rigid_body_pos: ${eval:'3 * ${robot.num_bodies} + 9'} 118 | - ref_motion_phase: 1 119 | - dr_base_com: 3 120 | - dr_link_mass: 22 # = num of 'randomize_link_body_names' 121 | - dr_kp: ${robot.dof_obs_size} 122 | - dr_kd: ${robot.dof_obs_size} 123 | - dr_friction: 1 124 | - dr_ctrl_delay: 1 125 | 126 | post_compute_config: {} # Will store values after preprocessing, don't put anything here 127 | motion_len: -1 -------------------------------------------------------------------------------- /humanoidverse/config/opt/record.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | env: 4 | config: 5 | save_motion: True 6 | eval_timestamp: ${eval_timestamp} 7 | save_note: null 8 | ckpt_dir: null 9 | save_total_steps: 10000 -------------------------------------------------------------------------------- /humanoidverse/config/opt/wandb.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | use_wandb: True 4 | 5 | wandb: 6 | wandb_project: null 7 | wandb_tags: 'online' # 'online' or 'offline' or 'disabled' 8 | wandb_group: null 9 | wandb_id: null 10 | wandb_entity: agileh2o 11 | wandb_dir: ${experiment_dir}/.wandb -------------------------------------------------------------------------------- /humanoidverse/config/rewards/motion_tracking/main.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | # defaults: 4 | # - reward_base 5 | 6 | 7 | rewards: 8 | set_reward: Weiji 9 | set_reward_date: 20250417 10 | only_positive_rewards: False 11 | reward_scales: 12 | teleop_contact_mask: 0.5 13 | # teleop_contact_mask: 0.5 14 | teleop_max_joint_position: 1.0 15 | 16 | teleop_body_position_extend: 1.0 17 | teleop_vr_3point: 1.6 18 | teleop_body_position_feet: 1.5 19 | teleop_body_rotation_extend: 0.5 20 | teleop_body_ang_velocity_extend: 0.5 21 | teleop_body_velocity_extend: 0.5 22 | teleop_joint_position: 1.0 23 | teleop_joint_velocity: 1.0 24 | penalty_torques: -0.000001 25 | # penalty_dof_vel: -0.0001 26 | # penalty_dof_acc: -2.5e-8 27 | penalty_action_rate: -0.5 # need to be tuned up to -0.5 for real robot 28 | 29 | feet_air_time: 1.0 30 | penalty_feet_contact_forces: -0.01 31 | penalty_stumble: -2.0 32 | penalty_slippage: -1.0 33 | 34 | limits_dof_pos: -10.0 35 | limits_dof_vel: -5.0 36 | limits_torque: -5.0 37 | termination: -200.0 38 | collision: -30.0 39 | 40 | 41 | teleop_body_pos_lowerbody_weight : 1.0 42 | teleop_body_pos_upperbody_weight : 1.0 43 | 44 | desired_feet_max_height_for_this_air: 0.2 45 | 46 | reward_tracking_sigma: 47 | teleop_max_joint_pos: 1.0 48 | 49 | teleop_upper_body_pos: 0.015 50 | teleop_lower_body_pos: 0.015 51 | teleop_vr_3point_pos: 0.015 52 | teleop_feet_pos: 0.01 53 | 54 | teleop_body_rot: 0.1 55 | teleop_body_vel: 1.0 56 | teleop_body_ang_vel: 15.0 57 | teleop_joint_pos: 0.3 58 | teleop_joint_vel: 30.0 59 | 60 | adaptive_tracking_sigma: 61 | enable: True 62 | alpha: 1e-3 63 | type: origin # origin, mean, scale 64 | scale: 1.0 65 | 66 | 67 | locomotion_max_contact_force: 400.0 68 | desired_feet_air_time: 0.3 69 | 70 | reward_penalty_curriculum: True 71 | reward_initial_penalty_scale : 0.10 72 | reward_min_penalty_scale: 0.0 73 | reward_max_penalty_scale: 1.0 74 | reward_penalty_level_down_threshold: 40 75 | reward_penalty_level_up_threshold: 42 76 | reward_penalty_degree: 0.00001 # 0.00001 77 | num_compute_average_epl: 10000 78 | 79 | reward_limit: 80 | soft_dof_pos_limit: 0.9 81 | soft_dof_vel_limit: 0.9 82 | soft_torque_limit: 0.825 83 | 84 | reward_limits_curriculum: 85 | soft_dof_pos_curriculum: True 86 | soft_dof_pos_initial_limit: 0.95 87 | soft_dof_pos_max_limit: 0.95 88 | soft_dof_pos_min_limit: 0.95 89 | soft_dof_pos_curriculum_degree: 0.00000025 90 | soft_dof_pos_curriculum_level_down_threshold: 40 91 | soft_dof_pos_curriculum_level_up_threshold: 42 92 | 93 | soft_dof_vel_curriculum: True 94 | soft_dof_vel_initial_limit: 0.95 95 | soft_dof_vel_max_limit: 0.95 96 | soft_dof_vel_min_limit: 0.95 97 | soft_dof_vel_curriculum_degree: 0.00000025 98 | soft_dof_vel_curriculum_level_down_threshold: 40 99 | soft_dof_vel_curriculum_level_up_threshold: 42 100 | 101 | soft_torque_curriculum: True 102 | soft_torque_initial_limit: 0.85 103 | soft_torque_max_limit: 0.85 104 | soft_torque_min_limit: 0.85 105 | soft_torque_curriculum_degree: 0.00000025 106 | soft_torque_curriculum_level_down_threshold: 40 107 | soft_torque_curriculum_level_up_threshold: 42 108 | 109 | reward_penalty_reward_names : [ 110 | "penalty_torques", 111 | "penalty_dof_acc", 112 | "penalty_dof_vel", 113 | "penalty_action_rate", 114 | "limits_dof_pos", 115 | "limits_dof_vel", 116 | "limits_torque", 117 | "feet_heading_alignment", 118 | "penalty_feet_ori", 119 | "penalty_slippage", 120 | "collision" 121 | ] -------------------------------------------------------------------------------- /humanoidverse/config/robot/robot_base.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | robot: 4 | only_track_leg: ??? 5 | # Observation parameters 6 | dof_obs_size: ??? 7 | number_of_actions: ??? 8 | self_obs_max_coords: ??? 9 | num_bodies: ??? 10 | 11 | # Control parameters 12 | # dof_body_ids: ??? 13 | # dof_offsets: ??? 14 | 15 | algo_obs_dim_dict: ??? 16 | 17 | key_bodies: null 18 | contact_bodies: null 19 | 20 | foot_name: null 21 | 22 | init_state: null 23 | 24 | contact_pairs_multiplier: 16 25 | 26 | num_key_bodies: ${len:${robot.key_bodies}} 27 | mimic_small_marker_bodies: null 28 | 29 | 30 | control: 31 | # Can be "isaac_pd" or "P"/"V"/"T" for Proportional, Velocity, Torque control 32 | control_type: isaac_pd 33 | # PD Drive parameters: 34 | stiffness: null 35 | damping: null 36 | # action scale: target angle = actionScale * action + defaultAngle 37 | # only used in manual PD control 38 | action_scale: 1.0 39 | # Used with isaac pd controller 40 | isaac_pd_scale: False # This is needed for the SMPL model due to weight variations 41 | clamp_actions: 1.0 42 | clip_torques: ??? 43 | 44 | asset: 45 | collapse_fixed_joints: null 46 | replace_cylinder_with_capsule: null 47 | flip_visual_attachments: null 48 | armature: null 49 | thickness: null 50 | max_angular_velocity: null 51 | max_linear_velocity: null 52 | density: null 53 | angular_damping: null 54 | linear_damping: null 55 | disable_gravity: null 56 | fix_base_link: null 57 | default_dof_drive_mode: 1 # see GymDofDriveModeFlags (0 is none, 1 is pos tgt, 2 is vel tgt, 3 effort) 58 | 59 | robot_type: ??? 60 | urdf_file: null 61 | usd_file: null 62 | xml_file: null 63 | asset_root: "description/robots" 64 | self_collisions: 0 65 | filter_ints: null 66 | 67 | # motion_lib: 68 | # _target_: phys_anim.utils.motion_lib.MotionLib 69 | # motion_file: ${motion_file} 70 | # ref_height_adjust: 0. 71 | # fix_motion_heights: True 72 | -------------------------------------------------------------------------------- /humanoidverse/config/simulator/genesis.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | # Env structure 4 | simulator: 5 | _target_: humanoidverse.simulator.genesis.genesis.Genesis 6 | _recursive_: False 7 | config: 8 | name: "genesis" 9 | terrain: ${terrain} 10 | sim: 11 | fps: 200 12 | control_decimation: 4 13 | substeps: 1 14 | render_mode: "human" # [None, "human", "rgb_array"] 15 | render_interval: 1 16 | 17 | scene: 18 | num_envs: ${num_envs} 19 | env_spacing: ${env.config.env_spacing} 20 | replicate_physics: True -------------------------------------------------------------------------------- /humanoidverse/config/simulator/isaacgym.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | # Env structure 4 | simulator: 5 | _target_: humanoidverse.simulator.isaacgym.isaacgym.IsaacGym 6 | _recursive_: False 7 | config: 8 | name: "isaacgym" 9 | terrain: ${terrain} 10 | sim: 11 | fps: 200 12 | control_decimation: 4 13 | substeps: 1 14 | physx: 15 | num_threads: 4 16 | solver_type: 1 # 0: pgs, 1: tgs 17 | num_position_iterations: 4 18 | num_velocity_iterations: 0 19 | contact_offset: 0.01 20 | rest_offset: 0.0 21 | bounce_threshold_velocity: 0.5 22 | max_depenetration_velocity: 1.0 23 | default_buffer_size_multiplier: 5 24 | contact_collection: 2 -------------------------------------------------------------------------------- /humanoidverse/config/simulator/isaacsim.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | # Env structure 4 | simulator: 5 | _target_: humanoidverse.simulator.isaacsim.isaacsim.IsaacSim 6 | _recursive_: False 7 | config: 8 | name: "isaacsim" 9 | terrain: ${terrain} 10 | sim: 11 | fps: 200 12 | control_decimation: 4 13 | substeps: 1 14 | physx: 15 | num_threads: 10 16 | solver_type: 1 # 0: pgs, 1: tgs 17 | num_position_iterations: 4 18 | num_velocity_iterations: 0 19 | contact_offset: 0.01 20 | rest_offset: 0.0 21 | bounce_threshold_velocity: 0.5 22 | max_depenetration_velocity: 1.0 23 | default_buffer_size_multiplier: 5 24 | contact_collection: 2 25 | render_mode: "human" # [None, "human", "rgb_array"] 26 | render_interval: 4 27 | 28 | scene: 29 | num_envs: ${num_envs} 30 | env_spacing: ${env.config.env_spacing} 31 | replicate_physics: True -------------------------------------------------------------------------------- /humanoidverse/config/simulator/mujoco.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | # Env structure 4 | simulator: 5 | _target_: humanoidverse.simulator.mujoco.mujoco.MuJoCo 6 | _recursive_: False 7 | config: 8 | name: "mujoco" 9 | terrain: ${terrain} 10 | sim: 11 | fps: 500 12 | control_decimation: 10 13 | # fps: 200 14 | # control_decimation: 4 15 | render_mode: "human" # [None, "human", "rgb_array"] 16 | scene: 17 | num_envs: ${num_envs} 18 | env_spacing: ${env.config.env_spacing} -------------------------------------------------------------------------------- /humanoidverse/config/terrain/terrain_base.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | terrain: 4 | static_friction: 1.0 5 | dynamic_friction: 1.0 6 | restitution: 0.0 -------------------------------------------------------------------------------- /humanoidverse/config/terrain/terrain_locomotion.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | defaults: 4 | - terrain_base 5 | 6 | terrain: 7 | mesh_type : 'trimesh' # "heightfield" # none, plane, heightfield or trimesh 8 | horizontal_scale : 0.1 # [m] 9 | vertical_scale : 0.005 # [m] 10 | border_size : 40 # [m] 11 | curriculum : False 12 | # curriculum : False 13 | static_friction: 1.0 14 | dynamic_friction: 1.0 15 | restitution: 0.0 16 | # rough terrain only: 17 | measure_heights : False # keep it False 18 | measured_points_x : [ 0.] # 1mx1.6m rectangle (without center line) 19 | measured_points_y : [ 0.] 20 | selected : False # select a unique terrain type and pass all arguments 21 | terrain_kwargs : null # Dict of arguments for selected terrain 22 | max_init_terrain_level : 9 # starting curriculum state 23 | terrain_length : 8. 24 | terrain_width : 8. 25 | num_rows: 10 # number of terrain rows (levels) 26 | num_cols : 20 # number of terrain cols (types) 27 | terrain_types : ['flat', 'rough', 'low_obst', 'smooth_slope', 'rough_slope'] # do not duplicate! 28 | terrain_proportions : [0.2, 0.6, 0.2, 0.0, 0.0] 29 | # trimesh only: 30 | slope_treshold : 0.75 # slopes above this threshold will be corrected to vertical surfaces -------------------------------------------------------------------------------- /humanoidverse/config/terrain/terrain_locomotion_plane.yaml: -------------------------------------------------------------------------------- 1 | # @package _global_ 2 | 3 | defaults: 4 | - terrain_base 5 | 6 | terrain: 7 | mesh_type : 'plane' # "heightfield" # none, plane, heightfield or trimesh 8 | horizontal_scale : 0.1 # [m] 9 | vertical_scale : 0.005 # [m] 10 | border_size : 40 # [m] 11 | curriculum : False 12 | # curriculum : False 13 | static_friction: 1.0 14 | dynamic_friction: 1.0 15 | restitution: 0.0 16 | # rough terrain only: 17 | measure_heights : False # keep it False 18 | measured_points_x : [ 0.] # 1mx1.6m rectangle (without center line) 19 | measured_points_y : [ 0.] 20 | selected : False # select a unique terrain type and pass all arguments 21 | terrain_kwargs : null # Dict of arguments for selected terrain 22 | max_init_terrain_level : 9 # starting curriculum state 23 | terrain_length : 8. 24 | terrain_width : 8. 25 | num_rows: 10 # number of terrain rows (levels) 26 | num_cols : 20 # number of terrain cols (types) 27 | terrain_types : ['flat', 'rough', 'low_obst', 'smooth_slope', 'rough_slope'] # do not duplicate! 28 | terrain_proportions : [0.2, 0.6, 0.2, 0.0, 0.0] 29 | # trimesh only: 30 | slope_treshold : 0.75 # slopes above this threshold will be corrected to vertical surfaces -------------------------------------------------------------------------------- /humanoidverse/deploy/__init__.py: -------------------------------------------------------------------------------- 1 | from .urcirobot import URCIRobot, ObsCfg, URCIPolicyObs, CfgType 2 | 3 | 4 | -------------------------------------------------------------------------------- /humanoidverse/deploy/external/__init__.py: -------------------------------------------------------------------------------- 1 | 2 | # Grammar: ckpt=_external_xxxx 3 | # Then, ckpt is sent to here `GetExternalPolicy(description)` and get parsed 4 | 5 | from humanoidverse.deploy import * 6 | from typing import Dict 7 | from .core import * 8 | 9 | 10 | -------------------------------------------------------------------------------- /humanoidverse/deploy/external/core.py: -------------------------------------------------------------------------------- 1 | 2 | # Grammar: ckpt=_external_xxxx 3 | # Then, ckpt is sent to here `GetExternalPolicy(description)` and get parsed 4 | 5 | from humanoidverse.deploy import * 6 | from typing import Dict 7 | 8 | def GetZeroPolicy(name: str)->URCIPolicyObs: 9 | # Usage: $EVALMJC=_external_zero 10 | import numpy as np 11 | 12 | def policy_fn(obs_dict: Dict[str, np.ndarray]) -> np.ndarray: 13 | action = np.zeros((1,23)) 14 | return action 15 | 16 | def obs_fn(robot: URCIRobot)->np.ndarray: 17 | # print(robot.timer) 18 | q_split = np.split(robot.q, [6, 12, 15, 15+4], axis=0) 19 | print(q_split) 20 | return np.zeros(1) 21 | 22 | return (obs_fn, policy_fn) 23 | 24 | 25 | def GetSinPolicy(name: str)->URCIPolicyObs: 26 | # Usage: $EVALMJC=_external_zero 27 | import numpy as np 28 | 29 | action = np.zeros((1,23)) 30 | 31 | def policy_fn(obs_dict: Dict[str, np.ndarray]) -> np.ndarray: 32 | timer = obs_dict['actor_obs'][0] 33 | 34 | # frequency = 35 | 36 | # action[0,3] = np.sin(timer) * 0.3 37 | # action[0,9] = np.cos(timer) * 0.3 38 | 39 | 40 | 41 | action[0,4] = np.sin(timer) * 0.5 *4 42 | action[0,5] = np.cos(timer) * 0.2 *4 43 | 44 | action[0,10] = np.cos(timer) * 0.5 *4 45 | action[0,11] = np.sin(timer) * 0.2 *4 46 | return action 47 | 48 | def obs_fn(robot: URCIRobot)->np.ndarray: 49 | # print(robot.timer) 50 | return np.array(robot.timer*robot.dt) 51 | 52 | return (obs_fn, policy_fn) 53 | 54 | def GetExternalPolicy(description: str, config: CfgType)->URCIPolicyObs: 55 | external_name = description[len('_external'):] 56 | 57 | if external_name.startswith('_zero'): 58 | return GetZeroPolicy(external_name[5:]) 59 | elif external_name.startswith('_sin'): 60 | return GetSinPolicy(external_name[4:]) 61 | else: 62 | raise ValueError(f"Unknown external policy: {external_name}") 63 | 64 | 65 | 66 | 67 | 68 | -------------------------------------------------------------------------------- /humanoidverse/envs/__init__.py: -------------------------------------------------------------------------------- 1 | from .base_task import * 2 | from .env_utils import * 3 | from .legged_base_task import * 4 | from .motion_tracking import * -------------------------------------------------------------------------------- /humanoidverse/envs/base_task/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/humanoidverse/envs/base_task/__init__.py -------------------------------------------------------------------------------- /humanoidverse/envs/env_utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/humanoidverse/envs/env_utils/__init__.py -------------------------------------------------------------------------------- /humanoidverse/envs/env_utils/history_handler.py: -------------------------------------------------------------------------------- 1 | import torch 2 | from torch import Tensor 3 | from termcolor import colored 4 | from loguru import logger 5 | 6 | 7 | 8 | # @time_prot_cls_dec_ 9 | # TimePortion: 1.57% 10 | class HistoryHandler: 11 | 12 | def __init__(self, num_envs, history_config, obs_dims, device): 13 | self.obs_dims = obs_dims 14 | self.device = device 15 | self.num_envs = num_envs 16 | self.history = {} 17 | 18 | self.buffer_config = {} 19 | for aux_key, aux_config in history_config.items(): 20 | for obs_key, obs_num in aux_config.items(): 21 | if obs_key in self.buffer_config: 22 | self.buffer_config[obs_key] = max(self.buffer_config[obs_key], obs_num) 23 | else: 24 | self.buffer_config[obs_key] = obs_num 25 | 26 | for key in self.buffer_config.keys(): 27 | self.history[key] = torch.zeros(num_envs, self.buffer_config[key], obs_dims[key], device=self.device) 28 | 29 | logger.info(colored("History Handler Initialized", "green")) 30 | for key, value in self.buffer_config.items(): 31 | logger.info(f"Key: {key}, Value: {value}") 32 | 33 | def reset(self, reset_ids): 34 | if len(reset_ids)==0: 35 | return 36 | assert set(self.buffer_config.keys()) == set(self.history.keys()), f"History keys mismatch\n{self.buffer_config.keys()}\n{self.history.keys()}" 37 | for key in self.history.keys(): 38 | self.history[key][reset_ids] *= 0. 39 | 40 | def add(self, key: str, value: Tensor): 41 | assert key in self.history.keys(), f"Key {key} not found in history" 42 | val = self.history[key].clone() 43 | self.history[key][:, 1:] = val[:, :-1] 44 | self.history[key][:, 0] = value.clone() 45 | 46 | def query(self, key: str): 47 | assert key in self.history.keys(), f"Key {key} not found in history" 48 | return self.history[key].clone() -------------------------------------------------------------------------------- /humanoidverse/envs/env_utils/visualization.py: -------------------------------------------------------------------------------- 1 | 2 | class Point: 3 | # this is for visualization 4 | def __init__(self, pt): 5 | self.x = pt[0] 6 | self.y = pt[1] 7 | self.z = pt[2] -------------------------------------------------------------------------------- /humanoidverse/envs/legged_base_task/__init__.py: -------------------------------------------------------------------------------- 1 | from .legged_robot_base import * -------------------------------------------------------------------------------- /humanoidverse/envs/motion_tracking/__init__.py: -------------------------------------------------------------------------------- 1 | from .motion_tracking import * -------------------------------------------------------------------------------- /humanoidverse/isaac_utils/isaac_utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/humanoidverse/isaac_utils/isaac_utils/__init__.py -------------------------------------------------------------------------------- /humanoidverse/isaac_utils/isaac_utils/maths.py: -------------------------------------------------------------------------------- 1 | import torch 2 | import numpy as np 3 | import random 4 | import os 5 | 6 | @torch.jit.script 7 | def normalize(x, eps: float = 1e-9): 8 | return x / x.norm(p=2, dim=-1).clamp(min=eps, max=None).unsqueeze(-1) 9 | 10 | @torch.jit.script 11 | def torch_rand_float(lower, upper, shape, device): 12 | # type: (float, float, Tuple[int, int], str) -> Tensor 13 | return (upper - lower) * torch.rand(*shape, device=device) + lower 14 | 15 | @torch.jit.script 16 | def copysign(a, b): 17 | # type: (float, Tensor) -> Tensor 18 | a = torch.tensor(a, device=b.device, dtype=torch.float).repeat(b.shape[0]) 19 | return torch.abs(a) * torch.sign(b) 20 | 21 | 22 | 23 | def set_seed(seed, torch_deterministic=False): 24 | """ set seed across modules """ 25 | if seed == -1 and torch_deterministic: 26 | seed = 42 27 | elif seed == -1: 28 | seed = np.random.randint(0, 10000) 29 | print("Setting seed: {}".format(seed)) 30 | 31 | random.seed(seed) 32 | np.random.seed(seed) 33 | torch.manual_seed(seed) 34 | os.environ["PYTHONHASHSEED"] = str(seed) 35 | torch.cuda.manual_seed(seed) 36 | torch.cuda.manual_seed_all(seed) 37 | 38 | if torch_deterministic: 39 | # refer to https://docs.nvidia.com/cuda/cublas/index.html#cublasApi_reproducibility 40 | os.environ["CUBLAS_WORKSPACE_CONFIG"] = ":4096:8" 41 | torch.backends.cudnn.benchmark = False 42 | torch.backends.cudnn.deterministic = True 43 | torch.use_deterministic_algorithms(True) 44 | else: 45 | torch.backends.cudnn.benchmark = True 46 | torch.backends.cudnn.deterministic = False 47 | 48 | return seed 49 | 50 | -------------------------------------------------------------------------------- /humanoidverse/isaac_utils/setup.py: -------------------------------------------------------------------------------- 1 | from setuptools import setup 2 | 3 | setup( 4 | name="isaac_utils", 5 | packages=["isaac_utils"], 6 | version="0.0.1", 7 | description="Unified torch env_utils for IsaacGym and IsaacSim", 8 | author="", 9 | classifiers=[], 10 | ) -------------------------------------------------------------------------------- /humanoidverse/simulator/genesis/genesis_viewer.py: -------------------------------------------------------------------------------- 1 | 2 | class Viewer: 3 | """This class handles controlling the camera associated with a viewport in the simulator. 4 | 5 | It can be used to set the viewpoint camera to track different origin types: 6 | 7 | - **world**: the center of the world (static) 8 | - **env**: the center of an environment (static) 9 | - **asset_root**: the root of an asset in the scene (e.g. tracking a robot moving in the scene) 10 | 11 | On creation, the camera is set to track the origin type specified in the configuration. 12 | 13 | For the :attr:`asset_root` origin type, the camera is updated at each rendering step to track the asset's 14 | root position. For this, it registers a callback to the post update event stream from the simulation app. 15 | """ 16 | 17 | def __init__(self): 18 | """Initialize the ViewportCameraController. 19 | 20 | Args: 21 | env: The environment. 22 | cfg: The configuration for the viewport camera controller. 23 | 24 | Raises: 25 | ValueError: If origin type is configured to be "env" but :attr:`cfg.env_index` is out of bounds. 26 | ValueError: If origin type is configured to be "asset_root" but :attr:`cfg.asset_name` is unset. 27 | 28 | """ 29 | self._cfg = None 30 | return 31 | 32 | def __del__(self): 33 | """Unsubscribe from the callback.""" 34 | # use hasattr to handle case where __init__ has not completed before __del__ is called 35 | return 36 | 37 | """ 38 | Properties 39 | """ 40 | 41 | @property 42 | def cfg(self): 43 | """The configuration for the viewer.""" 44 | return self._cfg 45 | 46 | """ 47 | Public Functions 48 | """ 49 | 50 | def set_view_env_index(self, env_index: int): 51 | """Sets the environment index for the camera view. 52 | 53 | Args: 54 | env_index: The index of the environment to set the camera view to. 55 | 56 | Raises: 57 | ValueError: If the environment index is out of bounds. It should be between 0 and num_envs - 1. 58 | """ 59 | return 60 | 61 | def update_view_to_world(self): 62 | """Updates the viewer's origin to the origin of the world which is (0, 0, 0).""" 63 | return 64 | 65 | def update_view_to_env(self): 66 | """Updates the viewer's origin to the origin of the selected environment.""" 67 | return 68 | 69 | def update_view_to_asset_root(self, asset_name: str): 70 | """Updates the viewer's origin based upon the root of an asset in the scene. 71 | 72 | Args: 73 | asset_name: The name of the asset in the scene. The name should match the name of the 74 | asset in the scene. 75 | 76 | Raises: 77 | ValueError: If the asset is not in the scene. 78 | """ 79 | return 80 | 81 | def update_view_location(self, eye, lookat): 82 | """Updates the camera view pose based on the current viewer origin and the eye and lookat positions. 83 | 84 | Args: 85 | eye: The eye position of the camera. If None, the current eye position is used. 86 | lookat: The lookat position of the camera. If None, the current lookat position is used. 87 | """ 88 | return 89 | 90 | """ 91 | Private Functions 92 | """ 93 | 94 | def _update_tracking_callback(self, event): 95 | """Updates the camera view at each rendering step.""" 96 | return 97 | -------------------------------------------------------------------------------- /humanoidverse/simulator/genesis/tmp_gs_utils.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import torch 3 | 4 | 5 | def wrap_to_pi(angles): 6 | angles %= 2 * np.pi 7 | angles -= 2 * np.pi * (angles > np.pi) 8 | return angles 9 | 10 | 11 | def gs_rand_float(lower, upper, shape, device): 12 | return (upper - lower) * torch.rand(size=shape, device=device) + lower 13 | 14 | 15 | def gs_inv_quat(quat): 16 | qw, qx, qy, qz = quat.unbind(-1) 17 | inv_quat = torch.stack([1.0 * qw, -qx, -qy, -qz], dim=-1) 18 | return inv_quat 19 | 20 | 21 | def gs_transform_by_quat(pos, quat): 22 | qw, qx, qy, qz = quat.unbind(-1) 23 | 24 | rot_matrix = torch.stack( 25 | [ 26 | 1.0 - 2 * qy**2 - 2 * qz**2, 27 | 2 * qx * qy - 2 * qz * qw, 28 | 2 * qx * qz + 2 * qy * qw, 29 | 2 * qx * qy + 2 * qz * qw, 30 | 1 - 2 * qx**2 - 2 * qz**2, 31 | 2 * qy * qz - 2 * qx * qw, 32 | 2 * qx * qz - 2 * qy * qw, 33 | 2 * qy * qz + 2 * qx * qw, 34 | 1 - 2 * qx**2 - 2 * qy**2, 35 | ], 36 | dim=-1, 37 | ).reshape(*quat.shape[:-1], 3, 3) 38 | rotated_pos = torch.matmul(rot_matrix, pos.unsqueeze(-1)).squeeze(-1) 39 | 40 | return rotated_pos 41 | 42 | 43 | def gs_quat2euler(quat): # xyz 44 | # Extract quaternion components 45 | qw, qx, qy, qz = quat.unbind(-1) 46 | 47 | # Roll (x-axis rotation) 48 | sinr_cosp = 2 * (qw * qx + qy * qz) 49 | cosr_cosp = 1 - 2 * (qx * qx + qy * qy) 50 | roll = torch.atan2(sinr_cosp, cosr_cosp) 51 | 52 | # Pitch (y-axis rotation) 53 | sinp = 2 * (qw * qy - qz * qx) 54 | pitch = torch.where( 55 | torch.abs(sinp) >= 1, 56 | torch.sign(sinp) * torch.tensor(torch.pi / 2), 57 | torch.asin(sinp), 58 | ) 59 | 60 | # Yaw (z-axis rotation) 61 | siny_cosp = 2 * (qw * qz + qx * qy) 62 | cosy_cosp = 1 - 2 * (qy * qy + qz * qz) 63 | yaw = torch.atan2(siny_cosp, cosy_cosp) 64 | 65 | return torch.stack([roll, pitch, yaw], dim=-1) 66 | 67 | 68 | def gs_euler2quat(xyz): # xyz 69 | 70 | roll, pitch, yaw = xyz.unbind(-1) 71 | 72 | cosr = (roll * 0.5).cos() 73 | sinr = (roll * 0.5).sin() 74 | cosp = (pitch * 0.5).cos() 75 | sinp = (pitch * 0.5).sin() 76 | cosy = (yaw * 0.5).cos() 77 | siny = (yaw * 0.5).sin() 78 | 79 | qw = cosr * cosp * cosy + sinr * sinp * siny 80 | qx = sinr * cosp * cosy - cosr * sinp * siny 81 | qy = cosr * sinp * cosy + sinr * cosp * siny 82 | qz = cosr * cosp * siny - sinr * sinp * cosy 83 | 84 | return torch.stack([qw, qx, qy, qz], dim=-1) 85 | 86 | 87 | def gs_quat_from_angle_axis(angle, axis): 88 | theta = (angle / 2).unsqueeze(-1) 89 | xyz = normalize(axis) * theta.sin() 90 | w = theta.cos() 91 | return normalize(torch.cat([w, xyz], dim=-1)) 92 | 93 | 94 | def normalize(x, eps: float = 1e-9): 95 | return x / x.norm(p=2, dim=-1).clamp(min=eps, max=None).unsqueeze(-1) 96 | 97 | 98 | def gs_quat_mul(a, b): 99 | assert a.shape == b.shape 100 | shape = a.shape 101 | a = a.reshape(-1, 4) 102 | b = b.reshape(-1, 4) 103 | 104 | w1, x1, y1, z1 = a[:, 0], a[:, 1], a[:, 2], a[:, 3] 105 | w2, x2, y2, z2 = b[:, 0], b[:, 1], b[:, 2], b[:, 3] 106 | ww = (z1 + x1) * (x2 + y2) 107 | yy = (w1 - y1) * (w2 + z2) 108 | zz = (w1 + y1) * (w2 - z2) 109 | xx = ww + yy + zz 110 | qq = 0.5 * (xx + (z1 - x1) * (x2 - y2)) 111 | w = qq - ww + (z1 - y1) * (y2 - z2) 112 | x = qq - xx + (x1 + w1) * (x2 + w2) 113 | y = qq - yy + (w1 - x1) * (y2 + z2) 114 | z = qq - zz + (z1 + y1) * (w2 - x2) 115 | 116 | quat = torch.stack([w, x, y, z], dim=-1).view(shape) 117 | 118 | return quat 119 | 120 | 121 | def gs_quat_apply(a, b): 122 | shape = b.shape 123 | a = a.reshape(-1, 4) 124 | b = b.reshape(-1, 3) 125 | xyz = a[:, 1:] 126 | t = xyz.cross(b, dim=-1) * 2 127 | return (b + a[:, :1] * t + xyz.cross(t, dim=-1)).view(shape) 128 | 129 | 130 | def gs_quat_apply_yaw(quat, vec): 131 | quat_yaw = quat.clone().view(-1, 4) 132 | quat_yaw[:, 1:3] = 0. 133 | quat_yaw = normalize(quat_yaw) 134 | return gs_quat_apply(quat_yaw, vec) 135 | 136 | 137 | def gs_quat_conjugate(a): 138 | shape = a.shape 139 | a = a.reshape(-1, 4) 140 | return torch.cat((a[:, :1], -a[:, 1:], ), dim=-1).view(shape) 141 | -------------------------------------------------------------------------------- /humanoidverse/simulator/isaacsim/event_cfg.py: -------------------------------------------------------------------------------- 1 | 2 | from omni.isaac.lab.managers import EventTermCfg as EventTerm 3 | 4 | from omni.isaac.lab.managers import SceneEntityCfg 5 | from omni.isaac.lab.utils import configclass 6 | 7 | 8 | import omni.isaac.lab.envs.mdp as mdp 9 | 10 | # @configclass 11 | # class EventCfg: 12 | # """Configuration for events.""" 13 | 14 | # scale_body_mass = EventTerm( 15 | # func=mdp.randomize_rigid_body_mass, 16 | # mode="startup", 17 | # params={ 18 | # "asset_cfg": SceneEntityCfg("robot", body_names=".*"), 19 | # "mass_distribution_params": (0.8, 1.2), 20 | # "operation": "scale", 21 | # }, 22 | # ) 23 | 24 | # random_joint_friction = EventTerm( 25 | # func=mdp.randomize_joint_parameters, 26 | # mode="startup", 27 | # params={ 28 | # "asset_cfg": SceneEntityCfg("robot", joint_names=".*"), 29 | # "friction_distribution_params": (0.5, 1.25), 30 | # "operation": "scale", 31 | # }, 32 | # ) 33 | 34 | @configclass 35 | class EventCfg: 36 | """Configuration for events.""" 37 | scale_body_mass = None 38 | random_joint_friction = None 39 | 40 | 41 | -------------------------------------------------------------------------------- /humanoidverse/simulator/isaacsim/events.py: -------------------------------------------------------------------------------- 1 | from __future__ import annotations 2 | 3 | import numpy as np 4 | import torch 5 | from typing import TYPE_CHECKING, Literal 6 | 7 | import omni.isaac.lab.envs.mdp as mdp 8 | import omni.isaac.lab.utils.math as math_utils 9 | from omni.isaac.lab.assets import Articulation, RigidObject 10 | from omni.isaac.lab.managers import SceneEntityCfg 11 | 12 | from omni.isaac.lab.envs import ManagerBasedEnv 13 | 14 | def resolve_dist_fn( 15 | distribution: Literal["uniform", "log_uniform", "gaussian"] = "uniform", 16 | ): 17 | dist_fn = math_utils.sample_uniform 18 | 19 | if distribution == "uniform": 20 | dist_fn = math_utils.sample_uniform 21 | elif distribution == "log_uniform": 22 | dist_fn = math_utils.sample_log_uniform 23 | elif distribution == "gaussian": 24 | dist_fn = math_utils.sample_gaussian 25 | else: 26 | raise ValueError(f"Unrecognized distribution {distribution}") 27 | 28 | return dist_fn 29 | 30 | def randomize_body_com( 31 | env: ManagerBasedEnv, 32 | env_ids: torch.Tensor | None, 33 | asset_cfg: SceneEntityCfg, 34 | distribution_params: tuple[float, float] | tuple[torch.Tensor, torch.Tensor], 35 | operation: Literal["add", "abs", "scale"], 36 | distribution: Literal["uniform", "log_uniform", "gaussian"] = "uniform", 37 | num_envs: int = 1, # number of environments 38 | ): 39 | """Randomize the com of the bodies by adding, scaling or setting random values. 40 | 41 | This function allows randomizing the center of mass of the bodies of the asset. The function samples random values from the 42 | given distribution parameters and adds, scales or sets the values into the physics simulation based on the operation. 43 | 44 | .. tip:: 45 | This function uses CPU tensors to assign the body masses. It is recommended to use this function 46 | only during the initialization of the environment. 47 | """ 48 | # extract the used quantities (to enable type-hinting) 49 | asset: RigidObject | Articulation = env.scene[asset_cfg.name] 50 | 51 | # resolve environment ids 52 | if env_ids is None: 53 | env_ids = torch.arange(num_envs, device="cpu") 54 | else: 55 | env_ids = env_ids.cpu() 56 | 57 | # resolve body indices 58 | if asset_cfg.body_ids == slice(None): 59 | body_ids = torch.arange(asset.num_bodies, dtype=torch.int, device="cpu") 60 | else: 61 | body_ids = torch.tensor(asset_cfg.body_ids, dtype=torch.int, device="cpu") 62 | 63 | # get the current masses of the bodies (num_assets, num_bodies) 64 | coms = asset.root_physx_view.get_coms() 65 | # apply randomization on default values 66 | # import ipdb; ipdb.set_trace() 67 | coms[env_ids[:, None], body_ids] = env.default_coms[env_ids[:, None], body_ids].clone() 68 | 69 | dist_fn = resolve_dist_fn(distribution) 70 | 71 | if isinstance(distribution_params[0], torch.Tensor): 72 | distribution_params = (distribution_params[0].to(coms.device), distribution_params[1].to(coms.device)) 73 | 74 | env.base_com_bias[env_ids, :] = dist_fn( 75 | *distribution_params, (env_ids.shape[0], env.base_com_bias.shape[1]), device=coms.device 76 | ) 77 | 78 | # sample from the given range 79 | if operation == "add": 80 | coms[env_ids[:, None], body_ids, :3] += env.base_com_bias[env_ids[:, None], :] 81 | elif operation == "abs": 82 | coms[env_ids[:, None], body_ids, :3] = env.base_com_bias[env_ids[:, None], :] 83 | elif operation == "scale": 84 | coms[env_ids[:, None], body_ids, :3] *= env.base_com_bias[env_ids[:, None], :] 85 | else: 86 | raise ValueError( 87 | f"Unknown operation: '{operation}' for property randomization. Please use 'add', 'abs' or 'scale'." 88 | ) 89 | # set the mass into the physics simulation 90 | asset.root_physx_view.set_coms(coms, env_ids) -------------------------------------------------------------------------------- /humanoidverse/simulator/isaacsim/isaacsim_articulation_cfg.py: -------------------------------------------------------------------------------- 1 | 2 | import omni.isaac.lab.sim as sim_utils 3 | from omni.isaac.lab.actuators import ActuatorNetMLPCfg, DCMotorCfg, ImplicitActuatorCfg 4 | from omni.isaac.lab.assets.articulation import ArticulationCfg 5 | from omni.isaac.lab.utils.assets import ISAACLAB_NUCLEUS_DIR 6 | 7 | ARTICULATION_CFG = ArticulationCfg( 8 | spawn=sim_utils.UsdFileCfg( 9 | # usd_path=f"{ISAACLAB_NUCLEUS_DIR}/Robots/Unitree/H1/h1.usd", 10 | usd_path="description/robots/h1/h1.usd", 11 | activate_contact_sensors=True, 12 | rigid_props=sim_utils.RigidBodyPropertiesCfg( 13 | disable_gravity=False, 14 | retain_accelerations=False, 15 | linear_damping=0.0, 16 | angular_damping=0.0, 17 | max_linear_velocity=1000.0, 18 | max_angular_velocity=1000.0, 19 | max_depenetration_velocity=1.0, 20 | ), 21 | articulation_props=sim_utils.ArticulationRootPropertiesCfg( 22 | enabled_self_collisions=False, solver_position_iteration_count=4, solver_velocity_iteration_count=4 23 | ), 24 | ), 25 | init_state=ArticulationCfg.InitialStateCfg( 26 | pos=(0.0, 0.0, 1.05), 27 | joint_pos={ 28 | ".*_hip_yaw_joint": 0.0, 29 | ".*_hip_roll_joint": 0.0, 30 | ".*_hip_pitch_joint": -0.28, # -16 degrees 31 | ".*_knee_joint": 0.79, # 45 degrees 32 | ".*_ankle_joint": -0.52, # -30 degrees 33 | "torso_joint": 0.0, 34 | ".*_shoulder_pitch_joint": 0.28, 35 | ".*_shoulder_roll_joint": 0.0, 36 | ".*_shoulder_yaw_joint": 0.0, 37 | ".*_elbow_joint": 0.52, 38 | }, 39 | joint_vel={".*": 0.0}, 40 | ), 41 | soft_joint_pos_limit_factor=0.9, 42 | actuators={ 43 | "legs": ImplicitActuatorCfg( 44 | joint_names_expr=[".*_hip_yaw_joint", ".*_hip_roll_joint", ".*_hip_pitch_joint", ".*_knee_joint", "torso_joint"], 45 | effort_limit=300, 46 | velocity_limit=100.0, 47 | stiffness={ 48 | ".*_hip_yaw_joint": 150.0, 49 | ".*_hip_roll_joint": 150.0, 50 | ".*_hip_pitch_joint": 200.0, 51 | ".*_knee_joint": 200.0, 52 | "torso_joint": 200.0, 53 | }, 54 | damping={ 55 | ".*_hip_yaw_joint": 5.0, 56 | ".*_hip_roll_joint": 5.0, 57 | ".*_hip_pitch_joint": 5.0, 58 | ".*_knee_joint": 5.0, 59 | "torso_joint": 5.0, 60 | }, 61 | ), 62 | "feet": ImplicitActuatorCfg( 63 | joint_names_expr=[".*_ankle_joint"], 64 | effort_limit=100, 65 | velocity_limit=100.0, 66 | stiffness={".*_ankle_joint": 20.0}, 67 | damping={".*_ankle_joint": 4.0}, 68 | ), 69 | "arms": ImplicitActuatorCfg( 70 | joint_names_expr=[".*_shoulder_pitch_joint", ".*_shoulder_roll_joint", ".*_shoulder_yaw_joint", ".*_elbow_joint"], 71 | effort_limit=300, 72 | velocity_limit=100.0, 73 | stiffness={ 74 | ".*_shoulder_pitch_joint": 40.0, 75 | ".*_shoulder_roll_joint": 40.0, 76 | ".*_shoulder_yaw_joint": 40.0, 77 | ".*_elbow_joint": 40.0, 78 | }, 79 | damping={ 80 | ".*_shoulder_pitch_joint": 10.0, 81 | ".*_shoulder_roll_joint": 10.0, 82 | ".*_shoulder_yaw_joint": 10.0, 83 | ".*_elbow_joint": 10.0, 84 | }, 85 | ), 86 | }, 87 | ) -------------------------------------------------------------------------------- /humanoidverse/train_agent.py: -------------------------------------------------------------------------------- 1 | import os 2 | import sys 3 | from pathlib import Path 4 | 5 | import hydra 6 | from hydra.core.hydra_config import HydraConfig 7 | from hydra.core.config_store import ConfigStore 8 | from hydra.utils import instantiate 9 | from omegaconf import OmegaConf 10 | 11 | import logging 12 | from loguru import logger 13 | 14 | from utils.devtool import * 15 | 16 | 17 | from utils.config_utils import * # noqa: E402, F403 18 | @hydra.main(config_path="config", config_name="base", version_base="1.1") 19 | def main(config: OmegaConf): 20 | # import ipdb; ipdb.set_trace() 21 | simulator_type = config.simulator['_target_'].split('.')[-1] 22 | # import ipdb; ipdb.set_trace() 23 | if simulator_type == 'IsaacSim': 24 | from omni.isaac.lab.app import AppLauncher 25 | import argparse 26 | parser = argparse.ArgumentParser(description="Train an RL agent with RSL-RL.") 27 | AppLauncher.add_app_launcher_args(parser) 28 | 29 | args_cli, hydra_args = parser.parse_known_args() 30 | sys.argv = [sys.argv[0]] + hydra_args 31 | args_cli.num_envs = config.num_envs 32 | args_cli.seed = config.seed 33 | args_cli.env_spacing = config.env.config.env_spacing # config.env_spacing 34 | args_cli.output_dir = config.output_dir 35 | args_cli.headless = config.headless 36 | 37 | app_launcher = AppLauncher(args_cli) 38 | simulation_app = app_launcher.app 39 | 40 | # import ipdb; ipdb.set_trace() 41 | if simulator_type == 'IsaacGym': 42 | import isaacgym # noqa: F401 43 | 44 | 45 | # have to import torch after isaacgym 46 | import torch # noqa: E402 47 | from humanoidverse.envs.base_task.base_task import BaseTask # noqa: E402 48 | from humanoidverse.agents.base_algo.base_algo import BaseAlgo # noqa: E402 49 | from humanoidverse.utils.helpers import pre_process_config 50 | from humanoidverse.utils.logging import HydraLoggerBridge 51 | 52 | # resolve=False is important otherwise overrides 53 | # at inference time won't work properly 54 | # also, I believe this must be done before instantiation 55 | 56 | # logging to hydra log file 57 | hydra_log_path = os.path.join(HydraConfig.get().runtime.output_dir, "train.log") 58 | logger.remove() 59 | logger.add(hydra_log_path, level="DEBUG") 60 | 61 | # Get log level from LOGURU_LEVEL environment variable or use INFO as default 62 | console_log_level = os.environ.get("LOGURU_LEVEL", "INFO").upper() 63 | logger.add(sys.stdout, level=console_log_level, colorize=True) 64 | 65 | logging.basicConfig(level=logging.DEBUG) 66 | logging.getLogger().addHandler(HydraLoggerBridge()) 67 | 68 | unresolved_conf = OmegaConf.to_container(config, resolve=False) 69 | os.chdir(hydra.utils.get_original_cwd()) 70 | 71 | if config.use_wandb: 72 | import wandb 73 | project_name = f"{config.project_name}" 74 | run_name = f"{config.timestamp}_{config.experiment_name}_{config.log_task_name}_{config.robot.asset.robot_type}" 75 | wandb_dir = Path(config.wandb.wandb_dir) 76 | wandb_dir.mkdir(exist_ok=True, parents=True) 77 | logger.info(f"Saving wandb logs to {wandb_dir}") 78 | wandb.init(project=project_name, 79 | entity=config.wandb.wandb_entity, 80 | name=run_name, 81 | sync_tensorboard=True, 82 | config=unresolved_conf, 83 | dir=wandb_dir) 84 | 85 | if hasattr(config, 'device'): 86 | if config.device is not None: 87 | device = config.device 88 | else: 89 | device = "cuda:0" if torch.cuda.is_available() else "cpu" 90 | else: 91 | device = "cuda:0" if torch.cuda.is_available() else "cpu" 92 | 93 | pre_process_config(config) 94 | 95 | config.env.config.save_rendering_dir = str(Path(config.experiment_dir) / "renderings_training") 96 | env: BaseEnv = instantiate(config=config.env, device=device) 97 | 98 | 99 | experiment_save_dir = Path(config.experiment_dir) 100 | experiment_save_dir.mkdir(exist_ok=True, parents=True) 101 | 102 | logger.info(f"Saving config file to {experiment_save_dir}") 103 | with open(experiment_save_dir / "config.yaml", "w") as file: 104 | OmegaConf.save(unresolved_conf, file) 105 | 106 | algo: BaseAlgo = instantiate(device=device, env=env, config=config.algo, log_dir=experiment_save_dir) 107 | algo.setup() 108 | if config.checkpoint is not None: 109 | algo.load(config.checkpoint) 110 | 111 | # handle saving config 112 | algo.learn() 113 | 114 | if simulator_type == 'IsaacSim': 115 | simulation_app.close() 116 | 117 | if __name__ == "__main__": 118 | main() 119 | -------------------------------------------------------------------------------- /humanoidverse/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/humanoidverse/utils/__init__.py -------------------------------------------------------------------------------- /humanoidverse/utils/average_meters.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import torch 3 | import torch.nn as nn 4 | 5 | 6 | class AverageMeter(nn.Module): 7 | def __init__(self, in_shape, max_size): 8 | super(AverageMeter, self).__init__() 9 | self.max_size = max_size 10 | self.current_size = 0 11 | self.register_buffer("mean", torch.zeros(in_shape, dtype=torch.float32)) 12 | 13 | def update(self, values): 14 | size = values.size()[0] 15 | if size == 0: 16 | return 17 | new_mean = torch.mean(values.float(), dim=0) 18 | size = np.clip(size, 0, self.max_size) 19 | old_size = min(self.max_size - size, self.current_size) 20 | size_sum = old_size + size 21 | self.current_size = size_sum 22 | self.mean = (self.mean * old_size + new_mean * size) / size_sum 23 | 24 | def clear(self): 25 | self.current_size = 0 26 | self.mean.fill_(0) 27 | 28 | def __len__(self): 29 | return self.current_size 30 | 31 | def get_mean(self): 32 | return self.mean.squeeze(0).cpu().numpy() 33 | 34 | 35 | class TensorAverageMeter: 36 | def __init__(self): 37 | self.tensors = [] 38 | 39 | def add(self, x): 40 | if len(x.shape) == 0: 41 | x = x.unsqueeze(0) 42 | self.tensors.append(x) 43 | 44 | def mean(self): 45 | if len(self.tensors) == 0: 46 | return 0 47 | cat = torch.cat(self.tensors, dim=0) 48 | if cat.numel() == 0: 49 | return 0 50 | else: 51 | return cat.mean() 52 | 53 | def clear(self): 54 | self.tensors = [] 55 | 56 | def mean_and_clear(self): 57 | mean = self.mean() 58 | self.clear() 59 | return mean 60 | 61 | 62 | class TensorAverageMeterDict: 63 | def __init__(self): 64 | self.data = {} 65 | 66 | def add(self, data_dict): 67 | for k, v in data_dict.items(): 68 | # Originally used a defaultdict, this had lambda 69 | # pickling issues with DDP. 70 | if k not in self.data: 71 | self.data[k] = TensorAverageMeter() 72 | self.data[k].add(v) 73 | 74 | def mean(self): 75 | mean_dict = {k: v.mean() for k, v in self.data.items()} 76 | return mean_dict 77 | 78 | def clear(self): 79 | self.data = {} 80 | 81 | def mean_and_clear(self): 82 | mean = self.mean() 83 | self.clear() 84 | return mean -------------------------------------------------------------------------------- /humanoidverse/utils/common.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2018-2022, NVIDIA Corporation 2 | # All rights reserved. 3 | # 4 | # Redistribution and use in source and binary forms, with or without 5 | # modification, are permitted provided that the following conditions are met: 6 | # 7 | # 1. Redistributions of source code must retain the above copyright notice, this 8 | # list of conditions and the following disclaimer. 9 | # 10 | # 2. Redistributions in binary form must reproduce the above copyright notice, 11 | # this list of conditions and the following disclaimer in the documentation 12 | # and/or other materials provided with the distribution. 13 | # 14 | # 3. Neither the name of the copyright holder nor the names of its 15 | # contributors may be used to endorse or promote products derived from 16 | # this software without specific prior written permission. 17 | # 18 | # THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" 19 | # AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 20 | # IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE 21 | # DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE 22 | # FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 23 | # DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR 24 | # SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER 25 | # CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, 26 | # OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE 27 | # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 28 | 29 | import argparse 30 | import os 31 | import random 32 | import sys 33 | from datetime import datetime 34 | 35 | import numpy as np 36 | import torch 37 | 38 | 39 | def seeding(seed=0, torch_deterministic=False): 40 | print("Setting seed: {}".format(seed)) 41 | 42 | random.seed(seed) 43 | np.random.seed(seed) 44 | torch.manual_seed(seed) 45 | os.environ["PYTHONHASHSEED"] = str(seed) 46 | torch.cuda.manual_seed(seed) 47 | torch.cuda.manual_seed_all(seed) 48 | 49 | if torch_deterministic: 50 | # refer to https://docs.nvidia.com/cuda/cublas/index.html#cublasApi_reproducibility 51 | os.environ["CUBLAS_WORKSPACE_CONFIG"] = ":4096:8" 52 | torch.backends.cudnn.benchmark = False 53 | torch.backends.cudnn.deterministic = True 54 | torch.use_deterministic_algorithms(True) 55 | else: 56 | torch.backends.cudnn.benchmark = True 57 | torch.backends.cudnn.deterministic = False 58 | 59 | return seed 60 | -------------------------------------------------------------------------------- /humanoidverse/utils/config_utils.py: -------------------------------------------------------------------------------- 1 | import math 2 | from loguru import logger 3 | from omegaconf import OmegaConf 4 | 5 | try: 6 | OmegaConf.register_new_resolver("eval", eval) 7 | OmegaConf.register_new_resolver("if", lambda pred, a, b: a if pred else b) 8 | OmegaConf.register_new_resolver("eq", lambda x, y: x.lower() == y.lower()) 9 | OmegaConf.register_new_resolver("sqrt", lambda x: math.sqrt(float(x))) 10 | OmegaConf.register_new_resolver("sum", lambda x: sum(x)) 11 | OmegaConf.register_new_resolver("ceil", lambda x: math.ceil(x)) 12 | OmegaConf.register_new_resolver("int", lambda x: int(x)) 13 | OmegaConf.register_new_resolver("len", lambda x: len(x)) 14 | OmegaConf.register_new_resolver("sum_list", lambda lst: sum(lst)) 15 | except Exception as e: 16 | logger.warning(f"Warning: Some resolvers already registered: {e}") 17 | 18 | # OmegaConf.register_new_resolver("eval", eval) 19 | # OmegaConf.register_new_resolver("if", lambda pred, a, b: a if pred else b) 20 | # OmegaConf.register_new_resolver("eq", lambda x, y: x.lower() == y.lower()) 21 | # OmegaConf.register_new_resolver("sqrt", lambda x: math.sqrt(float(x))) 22 | # OmegaConf.register_new_resolver("sum", lambda x: sum(x)) 23 | # OmegaConf.register_new_resolver("ceil", lambda x: math.ceil(x)) 24 | # OmegaConf.register_new_resolver("int", lambda x: int(x)) 25 | # OmegaConf.register_new_resolver("len", lambda x: len(x)) 26 | # OmegaConf.register_new_resolver("sum_list", lambda lst: sum(lst)) 27 | -------------------------------------------------------------------------------- /humanoidverse/utils/devtool.py: -------------------------------------------------------------------------------- 1 | 2 | 3 | from typing import Callable, Any 4 | from functools import wraps 5 | def pdb_decorator(fn: Callable) -> Callable: 6 | @wraps(fn) 7 | def wrapper(*args: Any, **kwargs: Any) -> Any: 8 | return fn(*args, **kwargs) 9 | return wrapper 10 | 11 | 12 | -------------------------------------------------------------------------------- /humanoidverse/utils/inference_helpers.py: -------------------------------------------------------------------------------- 1 | import torch 2 | from torch import nn 3 | import os 4 | import copy 5 | 6 | def export_policy_as_jit(actor_critic, path, exported_policy_name): 7 | os.makedirs(path, exist_ok=True) 8 | path = os.path.join(path, exported_policy_name) 9 | model = copy.deepcopy(actor_critic.actor).to('cpu') 10 | traced_script_module = torch.jit.script(model) 11 | traced_script_module.save(path) 12 | 13 | def export_policy_as_onnx(inference_model, path, exported_policy_name, example_obs_dict): 14 | os.makedirs(path, exist_ok=True) 15 | path = os.path.join(path, exported_policy_name) 16 | 17 | actor = copy.deepcopy(inference_model['actor']).to('cpu') 18 | 19 | class PPOWrapper(nn.Module): 20 | def __init__(self, actor): 21 | """ 22 | model: The original PyTorch model. 23 | input_keys: List of input names as keys for the input dictionary. 24 | """ 25 | super(PPOWrapper, self).__init__() 26 | self.actor = actor 27 | 28 | def forward(self, actor_obs): 29 | """ 30 | Dynamically creates a dictionary from the input keys and args. 31 | """ 32 | return self.actor.act_inference(actor_obs) 33 | 34 | wrapper = PPOWrapper(actor) 35 | example_input_list = example_obs_dict["actor_obs"] 36 | 37 | # actor.double() 38 | # wrapper.double() 39 | # example_input_list.double() 40 | 41 | torch.onnx.export( 42 | wrapper, 43 | example_input_list, # Pass x1 and x2 as separate inputs 44 | path, 45 | verbose=True, 46 | input_names=["actor_obs"], # Specify the input names 47 | output_names=["action"], # Name the output 48 | # opset_version=17 # Specify the opset version, if needed 49 | export_params=True, 50 | opset_version=13, 51 | do_constant_folding=True, 52 | ) 53 | 54 | def export_policy_and_estimator_as_onnx(inference_model, path, exported_policy_name, example_obs_dict): 55 | os.makedirs(path, exist_ok=True) 56 | path = os.path.join(path, exported_policy_name) 57 | 58 | actor = copy.deepcopy(inference_model['actor']).to('cpu') 59 | left_hand_force_estimator = copy.deepcopy(inference_model['left_hand_force_estimator']).to('cpu') 60 | right_hand_force_estimator = copy.deepcopy(inference_model['right_hand_force_estimator']).to('cpu') 61 | 62 | class PPOForceEstimatorWrapper(nn.Module): 63 | def __init__(self, actor, left_hand_force_estimator, right_hand_force_estimator): 64 | """ 65 | model: The original PyTorch model. 66 | input_keys: List of input names as keys for the input dictionary. 67 | """ 68 | super(PPOForceEstimatorWrapper, self).__init__() 69 | self.actor = actor 70 | self.left_hand_force_estimator = left_hand_force_estimator 71 | self.right_hand_force_estimator = right_hand_force_estimator 72 | 73 | def forward(self, inputs): 74 | """ 75 | Dynamically creates a dictionary from the input keys and args. 76 | """ 77 | actor_obs, history_for_estimator = inputs 78 | left_hand_force_estimator_output = self.left_hand_force_estimator(history_for_estimator) 79 | right_hand_force_estimator_output = self.right_hand_force_estimator(history_for_estimator) 80 | input_for_actor = torch.cat([actor_obs, left_hand_force_estimator_output, right_hand_force_estimator_output], dim=-1) 81 | return self.actor.act_inference(input_for_actor), left_hand_force_estimator_output, right_hand_force_estimator_output 82 | 83 | wrapper = PPOForceEstimatorWrapper(actor, left_hand_force_estimator, right_hand_force_estimator) 84 | example_input_list = [example_obs_dict["actor_obs"], example_obs_dict["long_history_for_estimator"]] 85 | torch.onnx.export( 86 | wrapper, 87 | example_input_list, # Pass x1 and x2 as separate inputs 88 | path, 89 | verbose=True, 90 | input_names=["actor_obs", "long_history_for_estimator"], # Specify the input names 91 | output_names=["action", "left_hand_force_estimator_output", "right_hand_force_estimator_output"], # Name the output 92 | opset_version=13 # Specify the opset version, if needed 93 | ) -------------------------------------------------------------------------------- /humanoidverse/utils/logging.py: -------------------------------------------------------------------------------- 1 | import logging 2 | from loguru import logger 3 | from contextlib import contextmanager 4 | import sys 5 | import os 6 | 7 | class HydraLoggerBridge(logging.Handler): 8 | def emit(self, record): 9 | # Get corresponding loguru level 10 | try: 11 | level = logger.level(record.levelname).name 12 | except ValueError: 13 | level = record.levelno 14 | 15 | # Find caller from where the logged message originated 16 | frame, depth = logging.currentframe(), 2 17 | while frame and frame.f_code.co_filename == logging.__file__: 18 | frame = frame.f_back 19 | depth += 1 20 | 21 | logger.opt(depth=depth, exception=record.exc_info).log(level, record.getMessage()) 22 | 23 | class LoguruStream: 24 | def write(self, message): 25 | if message.strip(): # Only log non-empty messages 26 | logger.info(message.strip()) # Changed to debug level 27 | 28 | def flush(self): 29 | pass 30 | 31 | @contextmanager 32 | def capture_stdout_to_loguru(): 33 | logger.remove() 34 | logger.add(sys.stdout, level="INFO") 35 | loguru_stream = LoguruStream() 36 | old_stdout = sys.stdout 37 | sys.stdout = loguru_stream 38 | try: 39 | yield 40 | finally: 41 | sys.stdout = old_stdout 42 | logger.remove() 43 | console_log_level = os.environ.get("LOGURU_LEVEL", "INFO").upper() 44 | logger.add(sys.stdout, level=console_log_level, colorize=True) -------------------------------------------------------------------------------- /humanoidverse/utils/math.py: -------------------------------------------------------------------------------- 1 | # SPDX-FileCopyrightText: Copyright (c) 2021 NVIDIA CORPORATION & AFFILIATES. All rights reserved. 2 | # SPDX-License-Identifier: BSD-3-Clause 3 | # 4 | # Redistribution and use in source and binary forms, with or without 5 | # modification, are permitted provided that the following conditions are met: 6 | # 7 | # 1. Redistributions of source code must retain the above copyright notice, this 8 | # list of conditions and the following disclaimer. 9 | # 10 | # 2. Redistributions in binary form must reproduce the above copyright notice, 11 | # this list of conditions and the following disclaimer in the documentation 12 | # and/or other materials provided with the distribution. 13 | # 14 | # 3. Neither the name of the copyright holder nor the names of its 15 | # contributors may be used to endorse or promote products derived from 16 | # this software without specific prior written permission. 17 | # 18 | # THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" 19 | # AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 20 | # IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE 21 | # DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE 22 | # FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 23 | # DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR 24 | # SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER 25 | # CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, 26 | # OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE 27 | # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 28 | # 29 | # Copyright (c) 2021 ETH Zurich, Nikita Rudin 30 | 31 | import torch 32 | from torch import Tensor 33 | import numpy as np 34 | from humanoidverse.utils.torch_utils import quat_apply, normalize 35 | from typing import Tuple 36 | 37 | # xyzw 38 | # @ torch.jit.script 39 | def quat_apply_yaw(quat, vec): 40 | quat_yaw = quat.clone().view(-1, 4) 41 | quat_yaw[:, :2] = 0. 42 | quat_yaw = normalize(quat_yaw) 43 | return quat_apply(quat_yaw, vec) 44 | 45 | # @ torch.jit.script 46 | def wrap_to_pi(angles): 47 | angles %= 2*np.pi 48 | angles -= 2*np.pi * (angles > np.pi) 49 | return angles 50 | 51 | # @ torch.jit.script 52 | def torch_rand_sqrt_float(lower, upper, shape, device): 53 | # type: (float, float, Tuple[int, int], str) -> Tensor 54 | r = 2*torch.rand(*shape, device=device) - 1 55 | r = torch.where(r<0., -torch.sqrt(-r), torch.sqrt(r)) 56 | r = (r + 1.) / 2. 57 | return (upper - lower) * r + lower -------------------------------------------------------------------------------- /humanoidverse/utils/motion_lib/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/humanoidverse/utils/motion_lib/__init__.py -------------------------------------------------------------------------------- /humanoidverse/utils/motion_lib/dof_axis.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/humanoidverse/utils/motion_lib/dof_axis.npy -------------------------------------------------------------------------------- /humanoidverse/utils/motion_lib/motion_lib_robot.py: -------------------------------------------------------------------------------- 1 | from humanoidverse.utils.motion_lib.motion_lib_base import MotionLibBase 2 | from humanoidverse.utils.motion_lib.torch_humanoid_batch import Humanoid_Batch 3 | class MotionLibRobot(MotionLibBase): 4 | def __init__(self, motion_lib_cfg, num_envs, device): 5 | super().__init__(motion_lib_cfg = motion_lib_cfg, num_envs = num_envs, device = device) 6 | self.mesh_parsers = Humanoid_Batch(motion_lib_cfg) 7 | return -------------------------------------------------------------------------------- /humanoidverse/utils/motion_lib/motion_utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/humanoidverse/utils/motion_lib/motion_utils/__init__.py -------------------------------------------------------------------------------- /humanoidverse/utils/motion_lib/motion_utils/flags.py: -------------------------------------------------------------------------------- 1 | __all__ = ['flags', 'summation'] 2 | 3 | class Flags(object): 4 | def __init__(self, items): 5 | for key, val in items.items(): 6 | setattr(self,key,val) 7 | 8 | flags = Flags({ 9 | 'test': False, 10 | 'debug': False, 11 | "real_traj": False, 12 | "im_eval": False, 13 | }) 14 | -------------------------------------------------------------------------------- /humanoidverse/utils/real/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/humanoidverse/utils/real/__init__.py -------------------------------------------------------------------------------- /humanoidverse/utils/real/rotation_helper.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | from scipy.spatial.transform import Rotation as R 3 | 4 | 5 | def get_gravity_orientation(quaternion): 6 | # WXYZ 7 | qw = quaternion[0] 8 | qx = quaternion[1] 9 | qy = quaternion[2] 10 | qz = quaternion[3] 11 | 12 | gravity_orientation = np.zeros(3) 13 | 14 | gravity_orientation[0] = 2 * (-qz * qx + qw * qy) 15 | gravity_orientation[1] = -2 * (qz * qy + qw * qx) 16 | gravity_orientation[2] = 1 - 2 * (qw * qw + qz * qz) 17 | 18 | return gravity_orientation 19 | 20 | 21 | def transform_imu_data(waist_yaw, waist_yaw_omega, imu_quat, imu_omega): 22 | # imu_quat: WXYZ 23 | RzWaist = R.from_euler("z", waist_yaw).as_matrix() 24 | R_torso = R.from_quat([imu_quat[1], imu_quat[2], imu_quat[3], imu_quat[0]]).as_matrix() 25 | R_pelvis = np.dot(R_torso, RzWaist.T) 26 | w = np.dot(RzWaist, imu_omega[0]) - np.array([0, 0, waist_yaw_omega]) 27 | return R.from_matrix(R_pelvis).as_quat()[[3, 0, 1, 2]], w 28 | 29 | 30 | def quaternion_to_euler_array(quat:np.ndarray)->np.ndarray: 31 | # Ensure quaternion is in the correct format [x, y, z, w] 32 | x, y, z, w = quat 33 | 34 | # Roll (x-axis rotation) 35 | t0 = +2.0 * (w * x + y * z) 36 | t1 = +1.0 - 2.0 * (x * x + y * y) 37 | roll_x = np.arctan2(t0, t1) 38 | 39 | # Pitch (y-axis rotation) 40 | t2 = +2.0 * (w * y - z * x) 41 | t2 = np.clip(t2, -1.0, 1.0) 42 | pitch_y = np.arcsin(t2) 43 | 44 | # Yaw (z-axis rotation) 45 | t3 = +2.0 * (w * z + x * y) 46 | t4 = +1.0 - 2.0 * (y * y + z * z) 47 | yaw_z = np.arctan2(t3, t4) 48 | 49 | # Returns roll, pitch, yaw in a NumPy array in radians 50 | return np.array([roll_x, pitch_y, yaw_z]) 51 | 52 | 53 | 54 | def rpy_to_quaternion_array(rpy: np.ndarray) -> np.ndarray: 55 | """Convert roll-pitch-yaw (radians) to quaternion [x, y, z, w]""" 56 | roll, pitch, yaw = rpy 57 | 58 | cy = np.cos(yaw * 0.5) 59 | sy = np.sin(yaw * 0.5) 60 | cp = np.cos(pitch * 0.5) 61 | sp = np.sin(pitch * 0.5) 62 | cr = np.cos(roll * 0.5) 63 | sr = np.sin(roll * 0.5) 64 | 65 | w = cr * cp * cy + sr * sp * sy 66 | x = sr * cp * cy - cr * sp * sy 67 | y = cr * sp * cy + sr * cp * sy 68 | z = cr * cp * sy - sr * sp * cy 69 | 70 | return np.array([x, y, z, w]) 71 | 72 | 73 | 74 | def quat_mul_np(q1:np.ndarray, q2:np.ndarray)->np.ndarray: 75 | # q1: XYZW # q2: XYZW 76 | # q1: XYZW # q2: XYZW 77 | x1, y1, z1, w1 = q1 78 | x2, y2, z2, w2 = q2 79 | 80 | w = w1 * w2 - x1 * x2 - y1 * y2 - z1 * z2 81 | x = w1 * x2 + x1 * w2 + y1 * z2 - z1 * y2 82 | y = w1 * y2 - x1 * z2 + y1 * w2 + z1 * x2 83 | z = w1 * z2 + x1 * y2 - y1 * x2 + z1 * w2 84 | 85 | return np.array([x, y, z, w]) # XYZW 86 | 87 | def random_perturb_quaternion(quat:np.ndarray, max_angle:float)->np.ndarray: 88 | """ 89 | Randomly perturb a quaternion by a random axis and angle 90 | """ 91 | # quat: XYZW 92 | 93 | # Generate random axis and angle for perturbation 94 | axis = np.random.randn(3) 95 | axis = axis / np.linalg.norm(axis) # Normalize axis 96 | angle = max_angle * np.random.rand() # Random angle within max_angle 97 | 98 | # Convert angle-axis to quaternion 99 | sin_half_angle = np.sin(angle / 2) 100 | cos_half_angle = np.cos(angle / 2) 101 | perturb_quat = np.array([sin_half_angle * axis[0], 102 | sin_half_angle * axis[1], 103 | sin_half_angle * axis[2], 104 | cos_half_angle]) # XYZW format 105 | 106 | # Multiply quaternions to apply perturbation 107 | return quat_mul_np(perturb_quat, quat) -------------------------------------------------------------------------------- /motion_source/README.md: -------------------------------------------------------------------------------- 1 | # Motion Source 2 | 3 | This folder describes how we collect human motion from various sources and process them to the SMPL format. 4 | 5 | ## SMPL Format 6 | 7 | SMPL format data consists of a dictionary. In details: 8 | ``` 9 | 'betas': (10,) - this is the shape paramater of SMPL model, not important 10 | 'gender': str - the gender of SMPL model, default is neutral 11 | 'poses': (frames, 66) - joint rotations, 66 represents 22 joints with 3 angles of rotation each. 12 | 'trans': (frames, 3) - root translation (global translation) 13 | 'mocap_framerate': int - motion fps, amass and video data is 30, mocap data is 120. 14 | ``` 15 | 16 | 17 | ## AMASS 18 | 19 | AMASS motions do not need further processing in this step. 20 | 21 | ## Video 22 | 23 | We use [GVHMR](https://github.com/zju3dv/GVHMR) to extract motions from videos. 24 | 25 | ## LAFAN (Unitree) 26 | 27 | > Note that the Unitree Retargeted `LAFAN` dataset is not available on the public. It's originally released in [here](https://huggingface.co/datasets/unitreerobotics/LAFAN1_Retargeting_Dataset). 28 | 29 | The format of **Unitree LAFAN** dataset is `csv`. Two steps to process these data: 30 | 31 | 1. convert csv to pkl: 32 | ``` 33 | python convert_lafan_pkl.py --filepath --start 0 --end 100 34 | ``` 35 | - ` ` is the path of csv file and does not contain `.csv`. 36 | - `start` is the start frame index and `end` is the end frame index. We select the `[start:end]` frames from the csv file. 37 | 38 | 2. count pkl contact mask: 39 | 40 | ``` 41 | python count_pkl_contact_mask.py robot=unitree_g1_29dof_anneal_23dof +input_folder= +target_folder= 42 | ``` 43 | - It computes the contact mask of the motion data by a thresholding method. -------------------------------------------------------------------------------- /motion_source/convert_lafan_pkl.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | from scipy.spatial.transform import Rotation as sRot 3 | import joblib 4 | import argparse 5 | import os 6 | 7 | 8 | def convert(file_name, data): 9 | data = data.astype(np.float32) 10 | print(data.shape) 11 | root_trans = data[:, :3] 12 | root_qua = data[:, 3:7] 13 | dof = data[:, 7:] 14 | 15 | dof_new = np.concatenate((dof[:, :19], dof[:, 22:26]), axis=1) 16 | root_aa = sRot.from_quat(root_qua).as_rotvec() 17 | 18 | dof_axis = np.load('../description/robots/g1/dof_axis.npy', allow_pickle=True) 19 | dof_axis = dof_axis.astype(np.float32) 20 | 21 | pose_aa = np.concatenate( 22 | (np.expand_dims(root_aa, axis=1), dof_axis * np.expand_dims(dof_new, axis=2), np.zeros((data.shape[0], 3, 3))), 23 | axis=1).astype(np.float32) 24 | 25 | data_dump = { 26 | "root_trans_offset": root_trans, 27 | "pose_aa": pose_aa, 28 | "dof": dof_new, 29 | "root_rot": root_qua, 30 | "smpl_joints": pose_aa, 31 | "fps": 30 32 | } 33 | 34 | all_data = {} 35 | all_data[file_name] = data_dump 36 | 37 | os.makedirs('./lafan_pkl', exist_ok=True) 38 | joblib.dump(all_data, f'./lafan_pkl/{file_name}.pkl') 39 | 40 | 41 | if __name__ == "__main__": 42 | parser = argparse.ArgumentParser() 43 | parser.add_argument('--filepath', type=str, help="File name", required=True) 44 | parser.add_argument('--start', type=int, help="Start frame", default=0) 45 | parser.add_argument('--end', type=int, help="End frame", default=100) 46 | args = parser.parse_args() 47 | 48 | start_frame = args.start 49 | end_frame = args.end 50 | robot_type = 'g1' 51 | 52 | filepath = args.filepath 53 | csv_file = filepath + '.csv' 54 | data = np.genfromtxt(csv_file, delimiter=',')[start_frame:end_frame, :] 55 | 56 | output_name = filepath.split('/')[-1] + '_' + f'{start_frame}' + '_' + f'{end_frame}' 57 | 58 | convert(output_name, data) -------------------------------------------------------------------------------- /motion_source/count_pkl_contact_mask.py: -------------------------------------------------------------------------------- 1 | import glob 2 | import os 3 | import sys 4 | import pdb 5 | import os.path as osp 6 | import numpy as np 7 | 8 | sys.path.append(os.getcwd()) 9 | 10 | from utils.torch_humanoid_batch import Humanoid_Batch 11 | import torch 12 | import joblib 13 | import hydra 14 | from omegaconf import DictConfig, OmegaConf 15 | 16 | from scipy.spatial.transform import Rotation as sRot 17 | 18 | def foot_detect(positions, thres=0.002): 19 | fid_r, fid_l = 12,6 20 | positions = positions.numpy() 21 | velfactor, heightfactor = np.array([thres]), np.array([0.12]) 22 | feet_l_x = (positions[1:, fid_l, 0] - positions[:-1, fid_l, 0]) ** 2 23 | feet_l_y = (positions[1:, fid_l, 1] - positions[:-1, fid_l, 1]) ** 2 24 | feet_l_z = (positions[1:, fid_l, 2] - positions[:-1, fid_l, 2]) ** 2 25 | feet_l_h = positions[1:,fid_l,2] 26 | feet_l = (((feet_l_x + feet_l_y + feet_l_z) < velfactor).astype(int) & (feet_l_h < heightfactor).astype(int)).astype(np.float32) 27 | feet_l = np.expand_dims(feet_l,axis=1) 28 | feet_l = np.concatenate([np.array([[1.]]),feet_l],axis=0) 29 | 30 | feet_r_x = (positions[1:, fid_r, 0] - positions[:-1, fid_r, 0]) ** 2 31 | feet_r_y = (positions[1:, fid_r, 1] - positions[:-1, fid_r, 1]) ** 2 32 | feet_r_z = (positions[1:, fid_r, 2] - positions[:-1, fid_r, 2]) ** 2 33 | feet_r_h = positions[1:,fid_r,2] 34 | feet_r = (((feet_r_x + feet_r_y + feet_r_z) < velfactor).astype(int) & (feet_r_h < heightfactor).astype(int)).astype(np.float32) 35 | feet_r = np.expand_dims(feet_r,axis=1) 36 | feet_r = np.concatenate([np.array([[1.]]),feet_r],axis=0) 37 | return feet_l, feet_r 38 | 39 | def count_pose_aa(motion): 40 | dof = motion['dof'] 41 | root_qua = motion['root_rot'] 42 | dof_new = np.concatenate((dof[:, :19], dof[:, 22:26]), axis=1) 43 | root_aa = sRot.from_quat(root_qua).as_rotvec() 44 | 45 | dof_axis = np.load('../description/robots/g1/dof_axis.npy', allow_pickle=True) 46 | dof_axis = dof_axis.astype(np.float32) 47 | 48 | pose_aa = np.concatenate( 49 | (np.expand_dims(root_aa, axis=1), dof_axis * np.expand_dims(dof_new, axis=2), np.zeros((dof_new.shape[0], 3, 3))), 50 | axis=1).astype(np.float32) 51 | 52 | return pose_aa,dof_new 53 | 54 | def process_motion(motion, cfg): 55 | device = torch.device("cpu") 56 | humanoid_fk = Humanoid_Batch(cfg.robot) # load forward kinematics model 57 | 58 | # breakpoint() 59 | 60 | if 'pose_aa' not in motion.keys(): 61 | pose_aa,dof = count_pose_aa(motion=motion) 62 | motion['pose_aa'] = pose_aa 63 | motion['dof'] = dof 64 | pose_aa = torch.from_numpy(pose_aa).unsqueeze(0) 65 | else: 66 | pose_aa = torch.from_numpy(motion['pose_aa']).unsqueeze(0) 67 | root_trans = torch.from_numpy(motion['root_trans_offset']).unsqueeze(0) 68 | 69 | fk_return = humanoid_fk.fk_batch(pose_aa, root_trans) 70 | 71 | feet_l, feet_r = foot_detect(fk_return.global_translation_extend[0]) 72 | 73 | motion['contact_mask'] = np.concatenate([feet_l,feet_r],axis=-1) 74 | motion['smpl_joints'] = fk_return.global_translation_extend[0].detach().numpy() 75 | 76 | return motion 77 | 78 | 79 | @hydra.main(version_base=None, config_path="../description/robots/cfg", config_name="config") 80 | def main(cfg: DictConfig) -> None: 81 | folder_path = cfg.input_folder 82 | if folder_path[-1]=='/': 83 | target_folder_path = folder_path[:-1] + '_contact_mask' 84 | else: 85 | target_folder_path = folder_path+'_contact_mask' 86 | os.makedirs(target_folder_path, exist_ok=True) 87 | target_folder_list = os.listdir(target_folder_path) 88 | for filename in os.listdir(folder_path): 89 | if filename.split('.')[0] + '_cont_mask.pkl' in target_folder_list: 90 | continue 91 | filename = filename.split('.')[0] 92 | motion_file = folder_path + '/' + f'{filename}.pkl' 93 | motion_data = joblib.load(motion_file) 94 | motion_data_keys = list(motion_data.keys()) 95 | motion = process_motion(motion_data[motion_data_keys[0]], cfg) 96 | save_data={} 97 | save_data[motion_data_keys[0]] = motion 98 | dumped_file = f'{target_folder_path}/{filename}_cont_mask.pkl' 99 | joblib.dump(save_data, dumped_file) 100 | 101 | if __name__ == "__main__": 102 | main() -------------------------------------------------------------------------------- /motion_source/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/motion_source/utils/__init__.py -------------------------------------------------------------------------------- /overview.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/overview.jpg -------------------------------------------------------------------------------- /robot_motion_process/README.md: -------------------------------------------------------------------------------- 1 | 2 | # Robot Motion Processing 3 | 4 | In this folder, we provide tools for visualizing and analyzing the processed motion and the motion trajectory from simulation. 5 | 6 | All of the retargeted reference motion and motion trajectory collected from simulation take the same data format. `motion_readpkl.py` provide an example code to load them. 7 | 8 | To collect the motion trajectory in IsaacGym, use `humanoidverse/sample_eps.py`. For MuJoCo, use `humanoidverse/urci.py`, the program will automatically save the motion trajectory to a pickle file when each episode ends. 9 | 10 | 11 | 12 | ## Visualization 13 | 14 | `vis_q_mj.py` and `vis_rr.py` are both used to visualize the motion data. These two scripts load the same data format, but have a different GUI. `vis_q_mj.py` is based on `mujoco_py`, provide reference motion & torque visualization and support manually correct the contact mask. `vis_rr.py` is based on `rerun`, provide more interactive visualization. 15 | 16 | 17 | 18 | Usage: 19 | 20 | ```bash 21 | python robot_motion_process/vis_q_mj.py +motion_file=... 22 | ``` 23 | See `def key_call_back(keycode)` in `vis_q_mj.py` for the key mapping. 24 | 25 | ```bash 26 | python robot_motion_process/vis_rr.py --filepath=... 27 | ``` 28 | 29 | Note that you need to install the additional dependencies for `vis_rr.py`. 30 | 31 | ```bash 32 | # (Optional) Install additional dependencies for visualization 33 | pip install rerun-sdk==0.22.0 trimesh 34 | ``` 35 | 36 | 37 | 38 | ## Interpolation 39 | 40 | `motion_interpolation_pkl.py` is used to preprocess the motion data. Given a default robot pose, this script will add a linear interpolation between the default pose and the motion frames at both the beginning and the end. So that the processed motion data all begin and end with the default pose. 41 | 42 | This is useful for deployment since it's inconvenient to initialize the real-world robot to the beginning pose of each motion. Some pose may be strange so that the robot cannot stand stably. 43 | 44 | Usage: 45 | ```bash 46 | python robot_motion_process/motion_interpolation_pkl.py +origin_file_name=... +start=... +end=... +start_inter_frame=... +end_inter_frame=... 47 | ``` 48 | - `origin_file_name` is the path to the original motion data, it should be a pickle file. 49 | - `start` and `end` are the start and end frame index of the motion data, we select the `[start:end]` frames from the original motion data. 50 | - `start_inter_frame` and `end_inter_frame` are the number of the added interpolation frames at the beginning and the end of the motion data. 51 | 52 | 53 | ## Trajectory Analysis 54 | 55 | `traj_vis.ipynb` is used to visualize the motion trajectory. For motion data collected from policy rollout, you can use it to compare the motion trajectory with the reference motion. 56 | 57 | Usage: Change the path to motion data in the notebook according to the example cells and run all cells. 58 | 59 | 60 | 61 | -------------------------------------------------------------------------------- /robot_motion_process/motion_readpkl.py: -------------------------------------------------------------------------------- 1 | import joblib 2 | import numpy as np 3 | import argparse 4 | 5 | def main(): 6 | parser = argparse.ArgumentParser(description="Load a motion file using joblib.") 7 | parser.add_argument('motion_file', type=str, help='Path to the motion file to load.') 8 | args = parser.parse_args() 9 | 10 | out_motion_data = joblib.load(args.motion_file) 11 | motion_data = next(iter(out_motion_data.values())) 12 | 13 | print(out_motion_data.keys()) 14 | print(motion_data.keys()) 15 | # print(motion_data) 16 | breakpoint() 17 | 18 | if __name__ == "__main__": 19 | main() 20 | -------------------------------------------------------------------------------- /setup.py: -------------------------------------------------------------------------------- 1 | from setuptools import find_packages, setup 2 | 3 | setup( 4 | name='pbhc', 5 | version='0.0.1', 6 | license="MIT", 7 | packages=find_packages(), 8 | description='Physics-Based Humanoid motion Control (PBHC)', 9 | python_requires=">=3.8", 10 | install_requires=[ 11 | "hydra-core>=1.2.0", 12 | "numpy==1.24.4", 13 | "mujoco==3.2.3", 14 | "mujoco-python-viewer", 15 | "rich", 16 | "ipdb", 17 | "matplotlib", 18 | "termcolor", 19 | "wandb", 20 | "plotly", 21 | "tqdm", 22 | "loguru", 23 | "meshcat", 24 | "pynput", 25 | "scipy", 26 | "tensorboard", 27 | "onnx", 28 | "onnxruntime", 29 | "opencv-python", 30 | "joblib", 31 | "easydict", 32 | "lxml", 33 | "numpy-stl", 34 | "open3d", 35 | "toolz", 36 | # 'rerun-sdk==0.22.0', 37 | # 'trimesh', 38 | # 'git+https://github.com/ZhengyiLuo/SMPLSim.git@master', 39 | 'typer', 40 | 'dm_control', 41 | 'loop_rate_limiters', 42 | 'mink', 43 | 'quadprog', 44 | 'einops' 45 | ] 46 | ) -------------------------------------------------------------------------------- /smpl_retarget/README.md: -------------------------------------------------------------------------------- 1 | # SMPL Motion Retarget 2 | 3 | Given the SMPL format motion data, this folder describes how we retarget them to the robot, take Unitree G1 as an example. 4 | 5 | Our code incorporates the retargeting pipelines from [MaskedMimic](https://github.com/NVlabs/ProtoMotions) and [PHC](https://github.com/ZhengyiLuo/PHC) - the former is built upon the differential inverse kinematics framework [Mink](https://github.com/kevinzakka/mink), while the latter employs gradient-based optimization. 6 | 7 | Both methods can be used to retarget human motion to the robot with slightly different results. We use Mink pipeline in our experiments. 8 | 9 | ## Motion Filter (Optional) 10 | 11 | Due to reconstruction inaccuracies and out-of-distribution issues in HMR models, motions extracted from videos may violate physical and biomechanical constraints. Thus, we try to filter out these motions via physics-based principles in our motion pipeline. You can choose it as an optional step. 12 | 13 | The motion filter implementation is adapted from the official [IPMAN](https://github.com/sha2nkt/ipman-r) codebase, so please refer to its installation process. 14 | 15 | Usage: 16 | ``` 17 | cd motion_filter 18 | python utils/motion_filter.py --folder --convert_rot True 19 | ``` 20 | - If motions are y-up, set `--convert_rot False`. If motions are z-up, set `--convert_rot True`. 21 | 22 | ## Mink Retarget 23 | 24 | First install `poselib`: 25 | ``` 26 | cd poselib 27 | pip install -e . 28 | ``` 29 | 30 | Retarget command: 31 | ``` 32 | python mink_retarget/convert_fit_motion.py 33 | ``` 34 | 35 | Add `--correct` will use motion correction. 36 | 37 | `` is the root folder of motion data. The motion data folder should be like this: 38 | 39 | ``` 40 | motion_data/ 41 | ├── video_motion/ 42 | | └── video1.npz 43 | | └── video2.npz 44 | | └── video3.npz 45 | | └── ... 46 | └── amass/ 47 | └── reverse_spin.npz 48 | └── wushu_form.npz 49 | └── ... 50 | ``` 51 | In above case, `` is `motion_data/` 52 | 53 | ## PHC Retarget 54 | 55 | Download the [SMPL](https://smpl.is.tue.mpg.de/) v1.1.0 parameters and place them in the `smpl_model/smpl/` folder. Rename the files `basicmodel_neutral_lbs_10_207_0_v1.1.0.pkl`, `basicmodel_m_lbs_10_207_0_v1.1.0.pkl`, `basicmodel_f_lbs_10_207_0_v1.1.0.pkl` to `SMPL_NEUTRAL.pkl`, `SMPL_MALE.pkl` and `SMPL_FEMALE.pkl` respectively. 56 | 57 | The folder structure should be organized like this: 58 | ``` 59 | smpl_model/smpl/ 60 | ├── SMPL_FEMALE.pkl 61 | ├── SMPL_MALE.pkl 62 | └── SMPL_NEUTRAL.pkl 63 | ``` 64 | 65 | Retarget command: 66 | ``` 67 | python phc_retarget/fit_smpl_motion.py robot=unitree_g1_29dof_anneal_23dof +motion= 68 | ``` -------------------------------------------------------------------------------- /smpl_retarget/mink_retarget/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/smpl_retarget/mink_retarget/__init__.py -------------------------------------------------------------------------------- /smpl_retarget/mink_retarget/retargeting/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/smpl_retarget/mink_retarget/retargeting/__init__.py -------------------------------------------------------------------------------- /smpl_retarget/mink_retarget/retargeting/config.py: -------------------------------------------------------------------------------- 1 | from easydict import EasyDict 2 | 3 | def g1_mapping(): 4 | #### Config for extension 5 | extend_config = [ 6 | { 7 | "joint_name": "left_rubber_hand_2", 8 | "parent_name": "left_wrist_yaw_link", 9 | "pos": [0.12, 0, 0.0], 10 | "rot": [1.0, 0.0, 0.0, 0.0], 11 | }, 12 | { 13 | "joint_name": "right_rubber_hand_2", 14 | "parent_name": "right_wrist_yaw_link", 15 | "pos": [0.12, 0, 0.0], 16 | "rot": [1.0, 0.0, 0.0, 0.0], 17 | }, 18 | { 19 | "joint_name": "head", 20 | "parent_name": "pelvis", 21 | "pos": [0.0, 0.0, 0.4], 22 | "rot": [1.0, 0.0, 0.0, 0.0], 23 | }, 24 | { 25 | "joint_name": "left_toe_link", 26 | "parent_name": "left_ankle_roll_link", 27 | "pos": [0.08, 0.0, -0.01], 28 | "rot": [1.0, 0.0, 0.0, 0.0], 29 | }, 30 | { 31 | "joint_name": "right_toe_link", 32 | "parent_name": "right_ankle_roll_link", 33 | "pos": [0.08, 0.0, -0.01], 34 | "rot": [1.0, 0.0, 0.0, 0.0], 35 | }, 36 | 37 | ] 38 | 39 | base_link = "torso_link" 40 | joint_matches = [ 41 | ["pelvis", "Pelvis"], 42 | ["left_hip_yaw_link", "L_Hip"], 43 | ["left_knee_link", "L_Knee"], 44 | ["left_ankle_roll_link", "L_Ankle"], 45 | ["right_hip_yaw_link", "R_Hip"], 46 | ["right_knee_link", "R_Knee"], 47 | ["right_ankle_roll_link", "R_Ankle"], 48 | ["left_shoulder_pitch_link", "L_Shoulder"], 49 | ["left_elbow_link", "L_Elbow"], 50 | ["left_rubber_hand_2", "L_Hand"], 51 | ["right_shoulder_pitch_link", "R_Shoulder"], 52 | ["right_elbow_link", "R_Elbow"], 53 | ["right_rubber_hand_2", "R_Hand"], 54 | ["head", "Head"], 55 | ] 56 | 57 | smpl_pose_modifier = [ 58 | {"Pelvis": "[np.pi/2, 0, np.pi/2]"}, 59 | {"L_Shoulder": "[0, 0, -np.pi/2]"}, 60 | {"R_Shoulder": "[0, 0, np.pi/2]"}, 61 | {"L_Elbow": "[0, -np.pi/2, 0]"}, 62 | {"R_Elbow": "[0, np.pi/2, 0]"}, 63 | ] 64 | 65 | asset_file = "../description/robots/g1/g1_29dof_rev_1_0.xml" 66 | 67 | return EasyDict( 68 | extend_config=extend_config, 69 | base_link=base_link, 70 | joint_matches=joint_matches, 71 | smpl_pose_modifier=smpl_pose_modifier, 72 | asset_file=asset_file, 73 | ) 74 | 75 | def smplx_with_limits_mapping(): 76 | #### Config for extension 77 | extend_config = [] 78 | 79 | base_link = "Pelvis" 80 | 81 | smplx_joints = [ 82 | "Pelvis", 83 | "L_Hip", 84 | "L_Knee", 85 | "L_Ankle", 86 | "L_Toe", 87 | "R_Hip", 88 | "R_Knee", 89 | "R_Ankle", 90 | "R_Toe", 91 | "Torso", 92 | "Spine", 93 | "Chest", 94 | "Neck", 95 | "Head", 96 | "L_Thorax", 97 | "L_Shoulder", 98 | "L_Elbow", 99 | "L_Wrist", 100 | "L_Index1", 101 | "L_Index2", 102 | "L_Index3", 103 | "L_Middle1", 104 | "L_Middle2", 105 | "L_Middle3", 106 | "L_Pinky1", 107 | "L_Pinky2", 108 | "L_Pinky3", 109 | "L_Ring1", 110 | "L_Ring2", 111 | "L_Ring3", 112 | "L_Thumb1", 113 | "L_Thumb2", 114 | "L_Thumb3", 115 | "R_Thorax", 116 | "R_Shoulder", 117 | "R_Elbow", 118 | "R_Wrist", 119 | "R_Index1", 120 | "R_Index2", 121 | "R_Index3", 122 | "R_Middle1", 123 | "R_Middle2", 124 | "R_Middle3", 125 | "R_Pinky1", 126 | "R_Pinky2", 127 | "R_Pinky3", 128 | "R_Ring1", 129 | "R_Ring2", 130 | "R_Ring3", 131 | "R_Thumb1", 132 | "R_Thumb2", 133 | "R_Thumb3", 134 | ] 135 | joint_matches = [[joint, joint] for joint in smplx_joints] 136 | 137 | smpl_pose_modifier = [] 138 | 139 | asset_file = "data/assets/mjcf/smplx_humanoid_with_limits.xml" 140 | 141 | return EasyDict( 142 | extend_config=extend_config, 143 | base_link=base_link, 144 | joint_matches=joint_matches, 145 | smpl_pose_modifier=smpl_pose_modifier, 146 | asset_file=asset_file, 147 | ) 148 | 149 | 150 | def get_config(humanoid_type: str): 151 | if humanoid_type == "g1": 152 | return g1_mapping() 153 | elif humanoid_type == "smplx_humanoid_with_limits": 154 | return smplx_with_limits_mapping() 155 | else: 156 | raise NotImplementedError 157 | -------------------------------------------------------------------------------- /smpl_retarget/mink_retarget/shape_optimized_neutral.pkl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/smpl_retarget/mink_retarget/shape_optimized_neutral.pkl -------------------------------------------------------------------------------- /smpl_retarget/motion_filter/constants.py: -------------------------------------------------------------------------------- 1 | FOCAL_LENGTH = 5000. 2 | IMG_RES = 224 3 | CONTACT_THRESH = 0.1 # gmof_rho in overlay vis = contact thresh 4 | #IMG_RES = 256 5 | 6 | JOINT_REGRESSOR_TRAIN_EXTRA = 'data/J_regressor_extra.npy' 7 | 8 | # Mean and standard deviation for normalizing input image 9 | IMG_NORM_MEAN = [0.485, 0.456, 0.406] 10 | IMG_NORM_STD = [0.229, 0.224, 0.225] 11 | 12 | # Dictionary of ground plane offsets in datasets 13 | GROUND_OFFSETS = { 14 | 'h36m': -0.095, 15 | } 16 | 17 | """ 18 | We create a superset of joints containing the OpenPose joints together with the ones that each dataset provides. 19 | We keep a superset of 24 joints such that we include all joints from every dataset. 20 | If a dataset doesn't provide annotations for a specific joint, we simply ignore it. 21 | The joints used here are the following: 22 | """ 23 | JOINT_NAMES = [ 24 | # 25 OpenPose joints (in the order provided by OpenPose) 25 | 'OP Nose', 26 | 'OP Neck', 27 | 'OP RShoulder', 28 | 'OP RElbow', 29 | 'OP RWrist', 30 | 'OP LShoulder', 31 | 'OP LElbow', 32 | 'OP LWrist', 33 | 'OP MidHip', 34 | 'OP RHip', 35 | 'OP RKnee', 36 | 'OP RAnkle', 37 | 'OP LHip', 38 | 'OP LKnee', 39 | 'OP LAnkle', 40 | 'OP REye', 41 | 'OP LEye', 42 | 'OP REar', 43 | 'OP LEar', 44 | 'OP LBigToe', 45 | 'OP LSmallToe', 46 | 'OP LHeel', 47 | 'OP RBigToe', 48 | 'OP RSmallToe', 49 | 'OP RHeel', 50 | # 24 Ground Truth joints (superset of joints from different datasets) 51 | 'Right Ankle', 52 | 'Right Knee', 53 | 'Right Hip', 54 | 'Left Hip', 55 | 'Left Knee', 56 | 'Left Ankle', 57 | 'Right Wrist', 58 | 'Right Elbow', 59 | 'Right Shoulder', 60 | 'Left Shoulder', 61 | 'Left Elbow', 62 | 'Left Wrist', 63 | 'Neck (LSP)', 64 | 'Top of Head (LSP)', 65 | 'Pelvis (MPII)', 66 | 'Thorax (MPII)', 67 | 'Spine (H36M)', 68 | 'Jaw (H36M)', 69 | 'Head (H36M)', 70 | 'Nose', 71 | 'Left Eye', 72 | 'Right Eye', 73 | 'Left Ear', 74 | 'Right Ear' 75 | ] 76 | 77 | # Dict containing the joints in numerical order 78 | JOINT_IDS = {JOINT_NAMES[i]: i for i in range(len(JOINT_NAMES))} 79 | 80 | # Map joints to SMPL joints 81 | JOINT_MAP = { 82 | 'OP Nose': 24, 'OP Neck': 12, 'OP RShoulder': 17, 83 | 'OP RElbow': 19, 'OP RWrist': 21, 'OP LShoulder': 16, 84 | 'OP LElbow': 18, 'OP LWrist': 20, 'OP MidHip': 0, 85 | 'OP RHip': 2, 'OP RKnee': 5, 'OP RAnkle': 8, 86 | 'OP LHip': 1, 'OP LKnee': 4, 'OP LAnkle': 7, 87 | 'OP REye': 25, 'OP LEye': 26, 'OP REar': 27, 88 | 'OP LEar': 28, 'OP LBigToe': 29, 'OP LSmallToe': 30, 89 | 'OP LHeel': 31, 'OP RBigToe': 32, 'OP RSmallToe': 33, 'OP RHeel': 34, 90 | 'Right Ankle': 8, 'Right Knee': 5, 'Right Hip': 45, 91 | 'Left Hip': 46, 'Left Knee': 4, 'Left Ankle': 7, 92 | 'Right Wrist': 21, 'Right Elbow': 19, 'Right Shoulder': 17, 93 | 'Left Shoulder': 16, 'Left Elbow': 18, 'Left Wrist': 20, 94 | 'Neck (LSP)': 47, 'Top of Head (LSP)': 48, 95 | 'Pelvis (MPII)': 49, 'Thorax (MPII)': 50, 96 | 'Spine (H36M)': 51, 'Jaw (H36M)': 52, 97 | 'Head (H36M)': 53, 'Nose': 24, 'Left Eye': 26, 98 | 'Right Eye': 25, 'Left Ear': 28, 'Right Ear': 27 99 | } 100 | 101 | # Joint selectors 102 | # Indices to get the 14 LSP joints from the 17 H36M joints 103 | H36M_TO_J17 = [6, 5, 4, 1, 2, 3, 16, 15, 14, 11, 12, 13, 8, 10, 0, 7, 9] 104 | H36M_TO_J14 = H36M_TO_J17[:14] 105 | # Indices to get the 14 LSP joints from the ground truth joints 106 | J24_TO_J17 = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 18, 14, 16, 17] 107 | J24_TO_J14 = J24_TO_J17[:14] 108 | 109 | # Permutation of SMPL pose parameters when flipping the shape 110 | SMPL_JOINTS_FLIP_PERM = [0, 2, 1, 3, 5, 4, 6, 8, 7, 9, 11, 10, 12, 14, 13, 15, 17, 16, 19, 18, 21, 20, 23, 22] 111 | SMPL_POSE_FLIP_PERM = [] 112 | for i in SMPL_JOINTS_FLIP_PERM: 113 | SMPL_POSE_FLIP_PERM.append(3*i) 114 | SMPL_POSE_FLIP_PERM.append(3*i+1) 115 | SMPL_POSE_FLIP_PERM.append(3*i+2) 116 | # Permutation indices for the 24 ground truth joints 117 | J24_FLIP_PERM = [5, 4, 3, 2, 1, 0, 11, 10, 9, 8, 7, 6, 12, 13, 14, 15, 16, 17, 18, 19, 21, 20, 23, 22] 118 | # Permutation indices for the full set of 49 joints 119 | J49_FLIP_PERM = [0, 1, 5, 6, 7, 2, 3, 4, 8, 12, 13, 14, 9, 10, 11, 16, 15, 18, 17, 22, 23, 24, 19, 20, 21]\ 120 | + [25+i for i in J24_FLIP_PERM] 121 | -------------------------------------------------------------------------------- /smpl_retarget/motion_filter/model/__init__.py: -------------------------------------------------------------------------------- 1 | from .smpl import SMPL -------------------------------------------------------------------------------- /smpl_retarget/motion_filter/model/smpl.py: -------------------------------------------------------------------------------- 1 | import torch 2 | import numpy as np 3 | import smplx 4 | from smplx import SMPL as _SMPL 5 | # from smplx.body_models import ModelOutput 6 | from smplx.utils import SMPLOutput 7 | from smplx.lbs import vertices2joints 8 | import constants 9 | 10 | 11 | class SMPL(_SMPL): 12 | """ Extension of the official SMPL implementation to support more joints """ 13 | 14 | def __init__(self, *args, **kwargs): 15 | super(SMPL, self).__init__(*args, **kwargs) 16 | joints = [constants.JOINT_MAP[i] for i in constants.JOINT_NAMES] 17 | J_regressor_extra = np.load(constants.JOINT_REGRESSOR_TRAIN_EXTRA) 18 | self.register_buffer('J_regressor_extra', torch.tensor(J_regressor_extra, dtype=torch.float32)) 19 | self.joint_map = torch.tensor(joints, dtype=torch.long) 20 | 21 | def forward(self, *args, **kwargs): 22 | kwargs['get_skin'] = True 23 | smpl_output = super(SMPL, self).forward(*args, **kwargs) 24 | extra_joints = vertices2joints(self.J_regressor_extra, smpl_output.vertices) 25 | joints = torch.cat([smpl_output.joints, extra_joints], dim=1) 26 | joints = joints[:, self.joint_map, :] 27 | output = SMPLOutput(vertices=smpl_output.vertices, 28 | global_orient=smpl_output.global_orient, 29 | body_pose=smpl_output.body_pose, 30 | joints=joints, 31 | betas=smpl_output.betas, 32 | full_pose=smpl_output.full_pose) 33 | 34 | return output -------------------------------------------------------------------------------- /smpl_retarget/motion_filter/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/smpl_retarget/motion_filter/utils/__init__.py -------------------------------------------------------------------------------- /smpl_retarget/motion_filter/utils/mesh_utils.py: -------------------------------------------------------------------------------- 1 | import trimesh 2 | import os.path as osp 3 | import numpy as np 4 | import torch 5 | import os 6 | import sys 7 | import torch.nn as nn 8 | 9 | ESSENTIALS_DIR = 'data/essentials' 10 | 11 | 12 | def sparse_batch_mm(m1, m2): 13 | """ 14 | 15 | m1: sparse matrix of size N x M 16 | m2: dense matrix of size B x M x K 17 | returns m1@m2 matrix of size B x N x K 18 | """ 19 | 20 | batch_size = m2.shape[0] 21 | # stack m2 into columns: (B x N x K) -> (N, B, K) -> (N, B * K) 22 | m2_stack = m2.transpose(0, 1).reshape(m1.shape[1], -1) 23 | result = m1.mm(m2_stack).reshape(m1.shape[0], batch_size, -1) \ 24 | .transpose(1, 0) 25 | return result 26 | 27 | 28 | class HDfier(): 29 | def __init__(self, model_type='smpl', device='cuda'): 30 | hd_operator_path = osp.join(ESSENTIALS_DIR, 'hd_model', model_type, 31 | f'{model_type}_neutral_hd_vert_regressor_sparse.npz') 32 | hd_operator = np.load(hd_operator_path) 33 | self.hd_operator = torch.sparse.FloatTensor( 34 | torch.tensor(hd_operator['index_row_col']), 35 | torch.tensor(hd_operator['values']), 36 | torch.Size(hd_operator['size'])) 37 | self.model_type = model_type 38 | self.device = device 39 | 40 | def hdfy_mesh(self, vertices): 41 | """ 42 | Applies a regressor that maps SMPL vertices to uniformly distributed vertices 43 | """ 44 | # device = body.vertices.device 45 | # check if vertices ndim are 3, if not , add a new axis 46 | if vertices.dim() != 3: 47 | # batchify the vertices 48 | vertices = vertices[None, :, :] 49 | 50 | # check if vertices are an ndarry, if yes, make pytorch tensor 51 | if isinstance(vertices, np.ndarray): 52 | vertices = torch.from_numpy(vertices).to(self.device) 53 | 54 | vertices = vertices.to(torch.double) 55 | 56 | if self.hd_operator.device != vertices.device: 57 | self.hd_operator = self.hd_operator.to(vertices.device) 58 | hd_verts = sparse_batch_mm(self.hd_operator, vertices).to(torch.float) 59 | return hd_verts 60 | -------------------------------------------------------------------------------- /smpl_retarget/motion_filter/utils/smpl_paraser.py: -------------------------------------------------------------------------------- 1 | from smpl_sim.smpllib.smpl_parser import ( 2 | SMPL_Parser, 3 | SMPLH_Parser, 4 | SMPLX_Parser, 5 | ) 6 | import torch 7 | import argparse 8 | import numpy as np 9 | 10 | if __name__ == '__main__': 11 | parser = argparse.ArgumentParser() 12 | parser.add_argument('--filepath', type=str, required=True) 13 | args = parser.parse_args() 14 | filepath = args.filepath 15 | motion_file = np.load(filepath, allow_pickle=True) 16 | smpl_parser_n = SMPL_Parser(model_path="data/smpl", gender="neutral") 17 | 18 | root_trans = motion_file['trans'] 19 | pose_aa = np.concatenate([motion_file['poses'][:, :66], np.zeros((root_trans.shape[0], 6))], axis=-1) 20 | betas = motion_file['betas'] 21 | gender = motion_file['gender'] 22 | 23 | origin_verts, origin_global_trans = smpl_parser_n.get_joints_verts(torch.from_numpy(pose_aa), torch.from_numpy(betas).unsqueeze(0), torch.from_numpy(root_trans)) 24 | 25 | breakpoint() -------------------------------------------------------------------------------- /smpl_retarget/poselib/.gitignore: -------------------------------------------------------------------------------- 1 | # Byte-compiled / optimized / DLL files 2 | __pycache__/ 3 | *.py[cod] 4 | *$py.class 5 | 6 | # C extensions 7 | *.so 8 | 9 | # Distribution / packaging 10 | .Python 11 | build/ 12 | develop-eggs/ 13 | dist/ 14 | downloads/ 15 | eggs/ 16 | .eggs/ 17 | lib/ 18 | lib64/ 19 | parts/ 20 | sdist/ 21 | var/ 22 | wheels/ 23 | *.egg-info/ 24 | .installed.cfg 25 | *.egg 26 | MANIFEST 27 | 28 | # PyInstaller 29 | # Usually these files are written by a python script from a template 30 | # before PyInstaller builds the exe, so as to inject date/other infos into it. 31 | *.manifest 32 | *.spec 33 | 34 | # Installer logs 35 | pip-log.txt 36 | pip-delete-this-directory.txt 37 | 38 | # Unit test / coverage reports 39 | htmlcov/ 40 | .tox/ 41 | .coverage 42 | .coverage.* 43 | .cache 44 | nosetests.xml 45 | coverage.xml 46 | *.cover 47 | .hypothesis/ 48 | .pytest_cache/ 49 | 50 | # Translations 51 | *.mo 52 | *.pot 53 | 54 | # Django stuff: 55 | *.log 56 | local_settings.py 57 | db.sqlite3 58 | 59 | # Flask stuff: 60 | instance/ 61 | .webassets-cache 62 | 63 | # Scrapy stuff: 64 | .scrapy 65 | 66 | # Sphinx documentation 67 | docs/_build/ 68 | 69 | # PyBuilder 70 | target/ 71 | 72 | # Jupyter Notebook 73 | .ipynb_checkpoints 74 | 75 | # pyenv 76 | .python-version 77 | 78 | # celery beat schedule file 79 | celerybeat-schedule 80 | 81 | # SageMath parsed files 82 | *.sage.py 83 | 84 | # Environments 85 | .env 86 | .venv 87 | env/ 88 | venv/ 89 | ENV/ 90 | env.bak/ 91 | venv.bak/ 92 | 93 | # Spyder project settings 94 | .spyderproject 95 | .spyproject 96 | 97 | # Rope project settings 98 | .ropeproject 99 | 100 | # mkdocs documentation 101 | /site 102 | 103 | # mypy 104 | .mypy_cache/ 105 | 106 | # vscode 107 | .vscode/ 108 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/README.md: -------------------------------------------------------------------------------- 1 | # poselib 2 | 3 | `poselib` is a library for loading, manipulating, and retargeting skeleton poses and motions. It is separated into three modules: `poselib.core` for basic data loading and tensor operations, `poselib.skeleton` for higher-level skeleton operations, and `poselib.visualization` for displaying skeleton poses. 4 | 5 | ## poselib.core 6 | - `poselib.core.rotation3d`: A set of Torch JIT functions for dealing with quaternions, transforms, and rotation/transformation matrices. 7 | - `quat_*` manipulate and create quaternions in [x, y, z, w] format (where w is the real component). 8 | - `transform_*` handle 7D transforms in [quat, pos] format. 9 | - `rot_matrix_*` handle 3x3 rotation matrices. 10 | - `euclidean_*` handle 4x4 Euclidean transformation matrices. 11 | - `poselib.core.tensor_utils`: Provides loading and saving functions for PyTorch tensors. 12 | 13 | ## poselib.skeleton 14 | - `poselib.skeleton.skeleton3d`: Utilities for loading and manipulating skeleton poses, and retargeting poses to different skeletons. 15 | - `SkeletonTree` is a class that stores a skeleton as a tree structure. 16 | - `SkeletonState` describes the static state of a skeleton, and provides both global and local joint angles. 17 | - `SkeletonMotion` describes a time-series of skeleton states and provides utilities for computing joint velocities. 18 | 19 | ## poselib.visualization 20 | - `poselib.visualization.common`: Functions used for visualizing skeletons interactively in `matplotlib`. 21 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/__init__.py: -------------------------------------------------------------------------------- 1 | __version__ = "0.0.1" 2 | 3 | from .core import * 4 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/core/__init__.py: -------------------------------------------------------------------------------- 1 | from .tensor_utils import * 2 | from .rotation3d import * 3 | from .backend import Serializable, logger 4 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/core/backend/__init__.py: -------------------------------------------------------------------------------- 1 | from .abstract import Serializable 2 | 3 | from .logger import logger 4 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/core/backend/abstract.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # NVIDIA CORPORATION and its licensors retain all intellectual property 3 | # and proprietary rights in and to this software, related documentation 4 | # and any modifications thereto. Any use, reproduction, disclosure or 5 | # distribution of this software and related documentation without an express 6 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 7 | 8 | from abc import ABCMeta, abstractmethod, abstractclassmethod 9 | from collections import OrderedDict 10 | import json 11 | 12 | import numpy as np 13 | import os 14 | 15 | TENSOR_CLASS = {} 16 | 17 | 18 | def register(name): 19 | global TENSOR_CLASS 20 | 21 | def core(tensor_cls): 22 | TENSOR_CLASS[name] = tensor_cls 23 | return tensor_cls 24 | 25 | return core 26 | 27 | 28 | def _get_cls(name): 29 | global TENSOR_CLASS 30 | return TENSOR_CLASS[name] 31 | 32 | 33 | class NumpyEncoder(json.JSONEncoder): 34 | """ Special json encoder for numpy types """ 35 | 36 | def default(self, obj): 37 | if isinstance( 38 | obj, 39 | ( 40 | np.int_, 41 | np.intc, 42 | np.intp, 43 | np.int8, 44 | np.int16, 45 | np.int32, 46 | np.int64, 47 | np.uint8, 48 | np.uint16, 49 | np.uint32, 50 | np.uint64, 51 | ), 52 | ): 53 | return int(obj) 54 | elif isinstance(obj, (np.float_, np.float16, np.float32, np.float64)): 55 | return float(obj) 56 | elif isinstance(obj, (np.ndarray,)): 57 | return dict(__ndarray__=obj.tolist(), dtype=str(obj.dtype), shape=obj.shape) 58 | return json.JSONEncoder.default(self, obj) 59 | 60 | 61 | def json_numpy_obj_hook(dct): 62 | if isinstance(dct, dict) and "__ndarray__" in dct: 63 | data = np.asarray(dct["__ndarray__"], dtype=dct["dtype"]) 64 | return data.reshape(dct["shape"]) 65 | return dct 66 | 67 | 68 | class Serializable: 69 | """ Implementation to read/write to file. 70 | All class the is inherited from this class needs to implement to_dict() and 71 | from_dict() 72 | """ 73 | 74 | @abstractclassmethod 75 | def from_dict(cls, dict_repr, *args, **kwargs): 76 | """ Read the object from an ordered dictionary 77 | 78 | :param dict_repr: the ordered dictionary that is used to construct the object 79 | :type dict_repr: OrderedDict 80 | :param args, kwargs: the arguments that need to be passed into from_dict() 81 | :type args, kwargs: additional arguments 82 | """ 83 | pass 84 | 85 | @abstractmethod 86 | def to_dict(self): 87 | """ Construct an ordered dictionary from the object 88 | 89 | :rtype: OrderedDict 90 | """ 91 | pass 92 | 93 | @classmethod 94 | def from_file(cls, path, *args, **kwargs): 95 | """ Read the object from a file (either .npy or .json) 96 | 97 | :param path: path of the file 98 | :type path: string 99 | :param args, kwargs: the arguments that need to be passed into from_dict() 100 | :type args, kwargs: additional arguments 101 | """ 102 | if path.endswith(".json"): 103 | with open(path, "r") as f: 104 | d = json.load(f, object_hook=json_numpy_obj_hook) 105 | elif path.endswith(".npy"): 106 | d = np.load(path, allow_pickle=True).item() 107 | else: 108 | assert False, "failed to load {} from {}".format(cls.__name__, path) 109 | assert d["__name__"] == cls.__name__, "the file belongs to {}, not {}".format( 110 | d["__name__"], cls.__name__ 111 | ) 112 | return cls.from_dict(d, *args, **kwargs) 113 | 114 | def to_file(self, path: str) -> None: 115 | """ Write the object to a file (either .npy or .json) 116 | 117 | :param path: path of the file 118 | :type path: string 119 | """ 120 | if os.path.dirname(path) != "" and not os.path.exists(os.path.dirname(path)): 121 | os.makedirs(os.path.dirname(path)) 122 | d = self.to_dict() 123 | d["__name__"] = self.__class__.__name__ 124 | if path.endswith(".json"): 125 | with open(path, "w") as f: 126 | json.dump(d, f, cls=NumpyEncoder, indent=4) 127 | elif path.endswith(".npy"): 128 | np.save(path, d) 129 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/core/backend/logger.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # NVIDIA CORPORATION and its licensors retain all intellectual property 3 | # and proprietary rights in and to this software, related documentation 4 | # and any modifications thereto. Any use, reproduction, disclosure or 5 | # distribution of this software and related documentation without an express 6 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 7 | 8 | import logging 9 | 10 | logger = logging.getLogger("poselib") 11 | logger.setLevel(logging.INFO) 12 | 13 | if not len(logger.handlers): 14 | formatter = logging.Formatter( 15 | fmt="%(asctime)-15s - %(levelname)s - %(module)s - %(message)s" 16 | ) 17 | handler = logging.StreamHandler() 18 | handler.setFormatter(formatter) 19 | logger.addHandler(handler) 20 | logger.info("logger initialized") 21 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/core/tensor_utils.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | 3 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 4 | # NVIDIA CORPORATION and its licensors retain all intellectual property 5 | # and proprietary rights in and to this software, related documentation 6 | # and any modifications thereto. Any use, reproduction, disclosure or 7 | # distribution of this software and related documentation without an express 8 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 9 | 10 | from collections import OrderedDict 11 | from .backend import Serializable 12 | import torch 13 | 14 | 15 | class TensorUtils(Serializable): 16 | @classmethod 17 | def from_dict(cls, dict_repr, *args, **kwargs): 18 | """ Read the object from an ordered dictionary 19 | 20 | :param dict_repr: the ordered dictionary that is used to construct the object 21 | :type dict_repr: OrderedDict 22 | :param kwargs: the arguments that need to be passed into from_dict() 23 | :type kwargs: additional arguments 24 | """ 25 | return torch.from_numpy(dict_repr["arr"].astype(dict_repr["context"]["dtype"])) 26 | 27 | def to_dict(self): 28 | """ Construct an ordered dictionary from the object 29 | 30 | :rtype: OrderedDict 31 | """ 32 | return NotImplemented 33 | 34 | def tensor_to_dict(x): 35 | """ Construct an ordered dictionary from the object 36 | 37 | :rtype: OrderedDict 38 | """ 39 | x_np = x.numpy() 40 | return { 41 | "arr": x_np, 42 | "context": { 43 | "dtype": x_np.dtype.name 44 | } 45 | } 46 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/core/tests/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/smpl_retarget/poselib/poselib/core/tests/__init__.py -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/core/tests/test_rotation.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # NVIDIA CORPORATION and its licensors retain all intellectual property 3 | # and proprietary rights in and to this software, related documentation 4 | # and any modifications thereto. Any use, reproduction, disclosure or 5 | # distribution of this software and related documentation without an express 6 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 7 | 8 | from ..rotation3d import * 9 | import numpy as np 10 | import torch 11 | 12 | q = torch.from_numpy(np.array([[0, 1, 2, 3], [-2, 3, -1, 5]], dtype=np.float32)) 13 | print("q", q) 14 | r = quat_normalize(q) 15 | x = torch.from_numpy(np.array([[1, 0, 0], [0, -1, 0]], dtype=np.float32)) 16 | print(r) 17 | print(quat_rotate(r, x)) 18 | 19 | angle = torch.from_numpy(np.array(np.random.rand() * 10.0, dtype=np.float32)) 20 | axis = torch.from_numpy(np.array([1, np.random.rand() * 10.0, np.random.rand() * 10.0], dtype=np.float32),) 21 | 22 | print(repr(angle)) 23 | print(repr(axis)) 24 | 25 | rot = quat_from_angle_axis(angle, axis) 26 | x = torch.from_numpy(np.random.rand(5, 6, 3)) 27 | y = quat_rotate(quat_inverse(rot), quat_rotate(rot, x)) 28 | print(x.numpy()) 29 | print(y.numpy()) 30 | assert np.allclose(x.numpy(), y.numpy()) 31 | 32 | m = torch.from_numpy(np.array([[1, 0, 0], [0, 0, -1], [0, 1, 0]], dtype=np.float32)) 33 | r = quat_from_rotation_matrix(m) 34 | t = torch.from_numpy(np.array([0, 1, 0], dtype=np.float32)) 35 | se3 = transform_from_rotation_translation(r=r, t=t) 36 | print(se3) 37 | print(transform_apply(se3, t)) 38 | 39 | rot = quat_from_angle_axis( 40 | torch.from_numpy(np.array([45, -54], dtype=np.float32)), 41 | torch.from_numpy(np.array([[1, 0, 0], [0, 1, 0]], dtype=np.float32)), 42 | degree=True, 43 | ) 44 | trans = torch.from_numpy(np.array([[1, 1, 0], [1, 1, 0]], dtype=np.float32)) 45 | transform = transform_from_rotation_translation(r=rot, t=trans) 46 | 47 | t = transform_mul(transform, transform_inverse(transform)) 48 | gt = np.zeros((2, 7)) 49 | gt[:, 0] = 1.0 50 | print(t.numpy()) 51 | print(gt) 52 | # assert np.allclose(t.numpy(), gt) 53 | 54 | transform2 = torch.from_numpy(np.array([[1, 0, 0, 1], [0, 0, -1, 0], [0, 1, 0, 0], [0, 0, 0, 1]], dtype=np.float32),) 55 | transform2 = euclidean_to_transform(transform2) 56 | print(transform2) 57 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/skeleton/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/smpl_retarget/poselib/poselib/skeleton/__init__.py -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/skeleton/backend/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/smpl_retarget/poselib/poselib/skeleton/backend/__init__.py -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/skeleton/backend/fbx/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/smpl_retarget/poselib/poselib/skeleton/backend/fbx/__init__.py -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/skeleton/backend/fbx/fbx_read_wrapper.py: -------------------------------------------------------------------------------- 1 | """ 2 | Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 3 | 4 | NVIDIA CORPORATION and its licensors retain all intellectual property and proprietary 5 | rights in and to this software, related documentation and any modifications thereto. Any 6 | use, reproduction, disclosure or distribution of this software and related documentation 7 | without an express license agreement from NVIDIA CORPORATION is strictly prohibited. 8 | """ 9 | 10 | """ 11 | Script that reads in fbx files from python 2 12 | 13 | This requires a configs file, which contains the command necessary to switch conda 14 | environments to run the fbx reading script from python 2 15 | """ 16 | 17 | from ....core import logger 18 | 19 | import inspect 20 | import os 21 | 22 | import numpy as np 23 | 24 | # Get the current folder to import the config file 25 | current_folder = os.path.realpath( 26 | os.path.abspath(os.path.split(inspect.getfile(inspect.currentframe()))[0]) 27 | ) 28 | 29 | 30 | def fbx_to_array(fbx_file_path, fbx_configs, root_joint, fps): 31 | """ 32 | Reads an fbx file to an array. 33 | 34 | Currently reading of the frame time is not supported. 120 fps is hard coded TODO 35 | 36 | :param fbx_file_path: str, file path to fbx 37 | :return: tuple with joint_names, parents, transforms, frame time 38 | """ 39 | 40 | # Ensure the file path is valid 41 | fbx_file_path = os.path.abspath(fbx_file_path) 42 | assert os.path.exists(fbx_file_path) 43 | 44 | # Switch directories to the env_utils folder to ensure the reading works 45 | previous_cwd = os.getcwd() 46 | os.chdir(current_folder) 47 | 48 | # Call the python 2.7 script 49 | temp_file_path = os.path.join(current_folder, fbx_configs["tmp_path"]) 50 | python_path = fbx_configs["fbx_py27_path"] 51 | logger.info("executing python script to read fbx data using Autodesk FBX SDK...") 52 | command = '{} fbx_py27_backend.py "{}" "{}" "{}" "{}"'.format( 53 | python_path, fbx_file_path, temp_file_path, root_joint, fps 54 | ) 55 | logger.debug("executing command: {}".format(command)) 56 | os.system(command) 57 | logger.info( 58 | "executing python script to read fbx data using Autodesk FBX SDK... done" 59 | ) 60 | 61 | with open(temp_file_path, "rb") as f: 62 | data = np.load(f) 63 | output = ( 64 | data["names"], 65 | data["parents"], 66 | data["transforms"], 67 | data["fps"], 68 | ) 69 | 70 | # Remove the temporary file 71 | os.remove(temp_file_path) 72 | 73 | # Return the os to its previous cwd, otherwise reading multiple files might fail 74 | os.chdir(previous_cwd) 75 | return output 76 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/skeleton/tests/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/smpl_retarget/poselib/poselib/skeleton/tests/__init__.py -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/skeleton/tests/ant.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 72 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/skeleton/tests/test_skeleton.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # NVIDIA CORPORATION and its licensors retain all intellectual property 3 | # and proprietary rights in and to this software, related documentation 4 | # and any modifications thereto. Any use, reproduction, disclosure or 5 | # distribution of this software and related documentation without an express 6 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 7 | 8 | from ...core import * 9 | from ..skeleton3d import SkeletonTree, SkeletonState, SkeletonMotion 10 | 11 | import numpy as np 12 | import torch 13 | 14 | from ...visualization.common import ( 15 | plot_skeleton_state, 16 | plot_skeleton_motion_interactive, 17 | ) 18 | 19 | from ...visualization.plt_plotter import Matplotlib3DPlotter 20 | from ...visualization.skeleton_plotter_tasks import ( 21 | Draw3DSkeletonMotion, 22 | Draw3DSkeletonState, 23 | ) 24 | 25 | 26 | def test_skel_tree(): 27 | skel_tree = SkeletonTree.from_mjcf( 28 | "/home/serfcx/DL_Animation/rl_mimic/data/skeletons/humanoid_mimic_mod_2_noind.xml", 29 | backend="pytorch", 30 | ) 31 | skel_tree_rec = SkeletonTree.from_dict(skel_tree.to_dict(), backend="pytorch") 32 | # assert skel_tree.to_str() == skel_tree_rec.to_str() 33 | print(skel_tree.node_names) 34 | print(skel_tree.local_translation) 35 | print(skel_tree.parent_indices) 36 | skel_state = SkeletonState.zero_pose(skeleton_tree=skel_tree) 37 | plot_skeleton_state(task_name="draw_skeleton", skeleton_state=skel_state) 38 | skel_state = skel_state.drop_nodes_by_names(["right_hip", "left_hip"]) 39 | plot_skeleton_state(task_name="draw_skeleton", skeleton_state=skel_state) 40 | 41 | 42 | def test_skel_motion(): 43 | skel_motion = SkeletonMotion.from_file( 44 | "/tmp/tmp.npy", backend="pytorch", load_context=True 45 | ) 46 | 47 | plot_skeleton_motion_interactive(skel_motion) 48 | 49 | 50 | def test_grad(): 51 | source_motion = SkeletonMotion.from_file( 52 | "c:\\Users\\bmatusch\\carbmimic\\data\\motions\\JogFlatTerrain_01_ase.npy", 53 | backend="pytorch", 54 | device="cuda:0", 55 | ) 56 | source_tpose = SkeletonState.from_file( 57 | "c:\\Users\\bmatusch\\carbmimic\\data\\skeletons\\fox_tpose.npy", 58 | backend="pytorch", 59 | device="cuda:0", 60 | ) 61 | 62 | target_tpose = SkeletonState.from_file( 63 | "c:\\Users\\bmatusch\\carbmimic\\data\\skeletons\\flex_tpose.npy", 64 | backend="pytorch", 65 | device="cuda:0", 66 | ) 67 | target_skeleton_tree = target_tpose.skeleton_tree 68 | 69 | joint_mapping = { 70 | "upArm_r": "right_shoulder", 71 | "upArm_l": "left_shoulder", 72 | "loArm_r": "right_elbow", 73 | "loArm_l": "left_elbow", 74 | "upLeg_r": "right_hip", 75 | "upLeg_l": "left_hip", 76 | "loLeg_r": "right_knee", 77 | "loLeg_l": "left_knee", 78 | "foot_r": "right_ankle", 79 | "foot_l": "left_ankle", 80 | "hips": "pelvis", 81 | "neckA": "neck", 82 | "spineA": "abdomen", 83 | } 84 | 85 | rotation_to_target_skeleton = quat_from_angle_axis( 86 | angle=torch.tensor(90.0).float(), 87 | axis=torch.tensor([1, 0, 0]).float(), 88 | degree=True, 89 | ) 90 | 91 | target_motion = source_motion.retarget_to( 92 | joint_mapping=joint_mapping, 93 | source_tpose_local_rotation=source_tpose.local_rotation, 94 | source_tpose_root_translation=source_tpose.root_translation, 95 | target_skeleton_tree=target_skeleton_tree, 96 | target_tpose_local_rotation=target_tpose.local_rotation, 97 | target_tpose_root_translation=target_tpose.root_translation, 98 | rotation_to_target_skeleton=rotation_to_target_skeleton, 99 | scale_to_target_skeleton=0.01, 100 | ) 101 | 102 | target_state = SkeletonState( 103 | target_motion.tensor[800, :], 104 | target_motion.skeleton_tree, 105 | target_motion.is_local, 106 | ) 107 | 108 | skeleton_tree = target_state.skeleton_tree 109 | root_translation = target_state.root_translation 110 | global_translation = target_state.global_translation 111 | 112 | q = np.zeros((len(skeleton_tree), 4), dtype=np.float32) 113 | q[..., 3] = 1.0 114 | q = torch.from_numpy(q) 115 | max_its = 10000 116 | 117 | task = Draw3DSkeletonState(task_name="", skeleton_state=target_state) 118 | plotter = Matplotlib3DPlotter(task) 119 | 120 | for i in range(max_its): 121 | r = quat_normalize(q) 122 | s = SkeletonState.from_rotation_and_root_translation( 123 | skeleton_tree, r=r, t=root_translation, is_local=True 124 | ) 125 | print(" quat norm: {}".format(q.norm(p=2, dim=-1).mean().numpy())) 126 | 127 | task.update(s) 128 | plotter.update() 129 | plotter.show() 130 | 131 | 132 | test_grad() -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/skeleton/tests/transfer_npy.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # NVIDIA CORPORATION and its licensors retain all intellectual property 3 | # and proprietary rights in and to this software, related documentation 4 | # and any modifications thereto. Any use, reproduction, disclosure or 5 | # distribution of this software and related documentation without an express 6 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 7 | 8 | import numpy as np 9 | from ...core import Tensor, SO3, Quaternion, Vector3D 10 | from ..skeleton3d import SkeletonTree, SkeletonState, SkeletonMotion 11 | 12 | tpose = np.load( 13 | "/home/serfcx/DL_Animation/rl_mimic/data/skeletons/flex_tpose.npy" 14 | ).item() 15 | 16 | local_rotation = SO3.from_numpy(tpose["local_rotation"], dtype="float32") 17 | root_translation = Vector3D.from_numpy(tpose["root_translation"], dtype="float32") 18 | skeleton_tree = tpose["skeleton_tree"] 19 | parent_indices = Tensor.from_numpy(skeleton_tree["parent_indices"], dtype="int32") 20 | local_translation = Vector3D.from_numpy( 21 | skeleton_tree["local_translation"], dtype="float32" 22 | ) 23 | node_names = skeleton_tree["node_names"] 24 | skeleton_tree = SkeletonTree(node_names, parent_indices, local_translation) 25 | skeleton_state = SkeletonState.from_rotation_and_root_translation( 26 | skeleton_tree=skeleton_tree, r=local_rotation, t=root_translation, is_local=True 27 | ) 28 | 29 | skeleton_state.to_file( 30 | "/home/serfcx/DL_Animation/rl_mimic/data/skeletons/flex_tpose_new.npy" 31 | ) 32 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/visualization/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/smpl_retarget/poselib/poselib/visualization/__init__.py -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/visualization/core.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. 2 | # NVIDIA CORPORATION and its licensors retain all intellectual property 3 | # and proprietary rights in and to this software, related documentation 4 | # and any modifications thereto. Any use, reproduction, disclosure or 5 | # distribution of this software and related documentation without an express 6 | # license agreement from NVIDIA CORPORATION is strictly prohibited. 7 | 8 | """ 9 | The base abstract classes for plotter and the plotting tasks. It describes how the plotter 10 | deals with the tasks in the general cases 11 | """ 12 | from typing import List 13 | 14 | 15 | class BasePlotterTask(object): 16 | _task_name: str # unique name of the task 17 | _task_type: str # type of the task is used to identify which callable 18 | 19 | def __init__(self, task_name: str, task_type: str) -> None: 20 | self._task_name = task_name 21 | self._task_type = task_type 22 | 23 | @property 24 | def task_name(self): 25 | return self._task_name 26 | 27 | @property 28 | def task_type(self): 29 | return self._task_type 30 | 31 | def get_scoped_name(self, name): 32 | return self._task_name + "/" + name 33 | 34 | def __iter__(self): 35 | """Should override this function to return a list of task primitives 36 | """ 37 | raise NotImplementedError 38 | 39 | 40 | class BasePlotterTasks(object): 41 | def __init__(self, tasks) -> None: 42 | self._tasks = tasks 43 | 44 | def __iter__(self): 45 | for task in self._tasks: 46 | yield from task 47 | 48 | 49 | class BasePlotter(object): 50 | """An abstract plotter which deals with a plotting task. The children class needs to implement 51 | the functions to create/update the objects according to the task given 52 | """ 53 | 54 | _task_primitives: List[BasePlotterTask] 55 | 56 | def __init__(self, task: BasePlotterTask) -> None: 57 | self._task_primitives = [] 58 | self.create(task) 59 | 60 | @property 61 | def task_primitives(self): 62 | return self._task_primitives 63 | 64 | def create(self, task: BasePlotterTask) -> None: 65 | """Create more task primitives from a task for the plotter""" 66 | new_task_primitives = list(task) # get all task primitives 67 | self._task_primitives += new_task_primitives # append them 68 | self._create_impl(new_task_primitives) 69 | 70 | def update(self) -> None: 71 | """Update the plotter for any updates in the task primitives""" 72 | self._update_impl(self._task_primitives) 73 | 74 | def _update_impl(self, task_list: List[BasePlotterTask]) -> None: 75 | raise NotImplementedError 76 | 77 | def _create_impl(self, task_list: List[BasePlotterTask]) -> None: 78 | raise NotImplementedError 79 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/visualization/tests/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/smpl_retarget/poselib/poselib/visualization/tests/__init__.py -------------------------------------------------------------------------------- /smpl_retarget/poselib/poselib/visualization/tests/test_plotter.py: -------------------------------------------------------------------------------- 1 | from typing import cast 2 | 3 | import matplotlib.pyplot as plt 4 | import numpy as np 5 | 6 | from ..core import BasePlotterTask, BasePlotterTasks 7 | from ..plt_plotter import Matplotlib3DPlotter 8 | from ..simple_plotter_tasks import Draw3DDots, Draw3DLines 9 | 10 | task = Draw3DLines(task_name="test", 11 | lines=np.array([[[0, 0, 0], [0, 0, 1]], [[0, 1, 1], [0, 1, 0]]]), color="blue") 12 | task2 = Draw3DDots(task_name="test2", 13 | dots=np.array([[0, 0, 0], [0, 0, 1], [0, 1, 1], [0, 1, 0]]), color="red") 14 | task3 = BasePlotterTasks([task, task2]) 15 | plotter = Matplotlib3DPlotter(cast(BasePlotterTask, task3)) 16 | plt.show() 17 | -------------------------------------------------------------------------------- /smpl_retarget/poselib/setup.py: -------------------------------------------------------------------------------- 1 | from setuptools import setup 2 | 3 | setup( 4 | name="poselib", 5 | packages=["poselib"], 6 | version="0.0.42", 7 | description="Framework Agnostic Tensor Programming", 8 | author="Qiyang Li, Kelly Guo, Brendon Matusch", 9 | classifiers=[ 10 | "Programming Language :: Python", 11 | "Programming Language :: Python :: 3", 12 | "License :: OSI Approved :: GNU General Public License (GPL)", 13 | "Operating System :: OS Independent", 14 | "Development Status :: 1 - Planning", 15 | "Environment :: Console", 16 | "Intended Audience :: Science/Research", 17 | "Topic :: Scientific/Engineering :: GIS", 18 | ], 19 | ) 20 | -------------------------------------------------------------------------------- /smpl_retarget/retargeted_motion_data/phc/shape_optimized_v1.pkl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/smpl_retarget/retargeted_motion_data/phc/shape_optimized_v1.pkl -------------------------------------------------------------------------------- /smpl_vis/README.md: -------------------------------------------------------------------------------- 1 | # SMPL Motions Visualization 2 | 3 | This folder describes how to visualize the SMPL format motion data. We provide two methods: `Blender` and `PyTorch3D`. 4 | 5 | ## Blender Vis 6 | 7 | [Blender](https://www.blender.org/) is an open source software for 3D CG. It's also a powerful tool for motion visualization. 8 | 9 | Import `npz` motion data into blender for visualization. 10 | 11 | 1. Download version `2.9.0` of blender and the [SMPL add-on](https://smpl.is.tue.mpg.de/index.html). 12 | 13 | 2. Add the SMPL object to the blender scene and run the `import_motion_blender.py` script in the `scripting` bar to bind the motion data to the SMPL object. 14 | 15 | ## PyTorch3D Vis 16 | 17 | This implementation is adapted from the `GVHMR` code, so please refer to its [installation](https://github.com/zju3dv/GVHMR/blob/main/docs/INSTALL.md) process. 18 | 19 | `smpl_neutral_J_regressor.pt` and `smplx2smpl_sparse.pt` must be put in `./body_model` (These two files can be found in the [repo](https://github.com/zju3dv/GVHMR/tree/main/hmr4d/utils/body_model)). Download [SMPLX] parameters and place it in the `../smpl_retarget/smpl_model/smplx/` folder. Rename the file to `SMPLX_NEUTRAL.npz`. The folder structure of `../smpl_retarget/smpl_model` should be organized like this: 20 | ``` 21 | smpl_model/ 22 | ├── smpl/ 23 | | └── SMPL_FEMALE.pkl 24 | | └── SMPL_MALE.pkl 25 | | └── SMPL_NEUTRAL.pkl 26 | └── smplx/ 27 | └── SMPLX_NEUTRAL.npz 28 | ``` 29 | 30 | Run the following command to visualize motion: 31 | 32 | ``` 33 | python smpl_render.py --filepath 34 | ``` 35 | 36 | - `` contains `.npz` -------------------------------------------------------------------------------- /smpl_vis/import_motion_blender.py: -------------------------------------------------------------------------------- 1 | import bpy 2 | import numpy as np 3 | from pathlib import Path 4 | from scipy.spatial.transform import Rotation as sRot 5 | import numpy as np 6 | 7 | npz_path = "Path_TO_NPZ" 8 | body_model_type = "SMPL-male" 9 | gender = "neutral" 10 | fps = 30 11 | 12 | try: 13 | data = np.load(npz_path, allow_pickle=True) 14 | print("Loaded keys:", list(data.keys())) 15 | except Exception as e: 16 | raise Exception(f"can not import the file: {str(e)}") 17 | 18 | poses = data['poses'][:, :72] 19 | betas = data['betas'] 20 | trans = data['trans'] 21 | 22 | if body_model_type not in bpy.data.objects: 23 | raise Exception(f"please first import {body_model_type} to scence") 24 | 25 | body = bpy.data.objects[body_model_type] 26 | 27 | if 'betas' in data: 28 | if hasattr(body, 'shape_keys'): 29 | for i in range(10): 30 | key = body.shape_keys.key_blocks.get(f'Shape{i:03d}') 31 | if key: 32 | key.value = betas[i] 33 | 34 | bone_list = ['Pelvis', 'L_Hip', 'R_Hip', 'Spine1', 'L_Knee', 'R_Knee', 'Spine2', 'L_Ankle', 35 | 'R_Ankle', 'Spine3', 'L_Foot', 'R_Foot', 'Neck', 'L_Collar', 'R_Collar', 36 | 'Head', 'L_Shoulder', 'R_Shoulder', 'L_Elbow', 'R_Elbow', 'L_Wrist', 37 | 'R_Wrist', 'L_Hand', 'R_Hand'] 38 | 39 | bpy.context.scene.frame_end = len(poses) 40 | for frame_idx in range(len(poses)): 41 | pose = poses[frame_idx] 42 | trans_vec = trans[frame_idx] 43 | 44 | i = 0 45 | for bone in bone_list: 46 | bone = body.pose.bones[bone] 47 | bone.rotation_mode = 'XYZ' 48 | if bone=='Pelvis': 49 | bone.rotation_euler = (sRot.from_rotvec(pose[:3]) * sRot.from_euler('xyz', np.array([np.pi / 2, 0, np.pi]))).as_rotvec() 50 | else: 51 | bone.rotation_euler = (pose[i*3], pose[i*3+1], pose[i*3+2]) 52 | bone.keyframe_insert(data_path="rotation_euler", frame=frame_idx) 53 | i+=1 54 | 55 | root_bone = body.pose.bones['Pelvis'] 56 | root_bone.location = (trans_vec[0], trans_vec[1], trans_vec[2]) 57 | root_bone.keyframe_insert(data_path="location", frame=frame_idx) 58 | #root_bone.location = trans_vec 59 | 60 | body.keyframe_insert(data_path="location", frame=frame_idx) 61 | body.keyframe_insert(data_path="rotation_euler", frame=frame_idx) 62 | 63 | print(f"Processed frame {frame_idx}/{len(poses)}") 64 | 65 | bpy.context.scene.render.fps = fps 66 | bpy.context.scene.frame_preview_start = 0 67 | bpy.context.scene.frame_preview_end = len(poses) 68 | 69 | print("Import Success！") 70 | -------------------------------------------------------------------------------- /smpl_vis/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TeleHuman/PBHC/0cb674b452c0ddbe37b5701522b475fac3afafd4/smpl_vis/utils/__init__.py -------------------------------------------------------------------------------- /smpl_vis/utils/geo_transform.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | import torch 3 | import torch.nn.functional as F 4 | from einops import einsum 5 | import imageio.v3 as iio 6 | 7 | def get_writer(video_path, fps=30, crf=17): 8 | """remember to .close()""" 9 | writer = iio.imopen(video_path, "w", plugin="pyav") 10 | writer.init_video_stream("libx264", fps=fps) 11 | writer._video_stream.options = {"crf": str(crf)} 12 | return writer 13 | 14 | 15 | def get_video_lwh(video_path): 16 | L, H, W, _ = iio.improps(video_path, plugin="pyav").shape 17 | return L, W, H 18 | 19 | def create_camera_sensor(width=None, height=None, f_fullframe=None): 20 | if width is None or height is None: 21 | # The 4:3 aspect ratio is widely adopted by image sensors in mobile phones. 22 | if np.random.rand() < 0.5: 23 | width, height = 1200, 1600 24 | else: 25 | width, height = 1600, 1200 26 | 27 | # Sample FOV from common options: 28 | # 1. wide-angle lenses are common in mobile phones, 29 | # 2. telephoto lenses has less perspective effect, which should makes it easy to learn 30 | if f_fullframe is None: 31 | f_fullframe_options = [24, 26, 28, 30, 35, 40, 50, 60, 70] 32 | f_fullframe = np.random.choice(f_fullframe_options) 33 | 34 | # We use diag to map focal-length: https://www.nikonians.org/reviews/fov-tables 35 | diag_fullframe = (24**2 + 36**2) ** 0.5 36 | diag_img = (width**2 + height**2) ** 0.5 37 | focal_length = diag_img / diag_fullframe * f_fullframe 38 | 39 | K_fullimg = torch.eye(3) 40 | K_fullimg[0, 0] = focal_length 41 | K_fullimg[1, 1] = focal_length 42 | K_fullimg[0, 2] = width / 2 43 | K_fullimg[1, 2] = height / 2 44 | 45 | return width, height, K_fullimg 46 | 47 | def apply_T_on_points(points, T): 48 | """ 49 | Args: 50 | points: (..., N, 3) 51 | T: (..., 4, 4) 52 | Returns: (..., N, 3) 53 | """ 54 | points_T = torch.einsum("...ki,...ji->...jk", T[..., :3, :3], points) + T[..., None, :3, 3] 55 | return points_T 56 | 57 | def transform_mat(R, t): 58 | """ 59 | Args: 60 | R: Bx3x3 array of a batch of rotation matrices 61 | t: Bx3x(1) array of a batch of translation vectors 62 | Returns: 63 | T: Bx4x4 Transformation matrix 64 | """ 65 | # No padding left or right, only add an extra row 66 | if len(R.shape) > len(t.shape): 67 | t = t[..., None] 68 | return torch.cat([F.pad(R, [0, 0, 0, 1]), F.pad(t, [0, 0, 0, 1], value=1)], dim=-1) 69 | 70 | def compute_T_ayfz2ay(joints, inverse=False): 71 | """ 72 | Args: 73 | joints: (B, J, 3), in the start-frame, ay-coordinate 74 | Returns: 75 | if inverse == False: 76 | T_ayfz2ay: (B, 4, 4) 77 | else : 78 | T_ay2ayfz: (B, 4, 4) 79 | """ 80 | t_ayfz2ay = joints[:, 0, :].detach().clone() 81 | t_ayfz2ay[:, 1] = 0 # do not modify y 82 | 83 | RL_xz_h = joints[:, 1, [0, 2]] - joints[:, 2, [0, 2]] # (B, 2), hip point to left side 84 | RL_xz_s = joints[:, 16, [0, 2]] - joints[:, 17, [0, 2]] # (B, 2), shoulder point to left side 85 | RL_xz = RL_xz_h + RL_xz_s 86 | I_mask = RL_xz.pow(2).sum(-1) < 1e-4 # do not rotate, when can't decided the face direction 87 | # if I_mask.sum() > 0: 88 | # Log.warn("{} samples can't decide the face direction".format(I_mask.sum())) 89 | 90 | x_dir = torch.zeros_like(t_ayfz2ay) # (B, 3) 91 | x_dir[:, [0, 2]] = F.normalize(RL_xz, 2, -1) 92 | y_dir = torch.zeros_like(x_dir) 93 | y_dir[..., 1] = 1 # (B, 3) 94 | z_dir = torch.cross(x_dir, y_dir, dim=-1) 95 | R_ayfz2ay = torch.stack([x_dir, y_dir, z_dir], dim=-1) # (B, 3, 3) 96 | R_ayfz2ay[I_mask] = torch.eye(3).to(R_ayfz2ay) 97 | 98 | if inverse: 99 | R_ay2ayfz = R_ayfz2ay.transpose(1, 2) 100 | t_ay2ayfz = -einsum(R_ayfz2ay, t_ayfz2ay, "b i j , b i -> b j") 101 | return transform_mat(R_ay2ayfz, t_ay2ayfz) 102 | else: 103 | return transform_mat(R_ayfz2ay, t_ayfz2ay) -------------------------------------------------------------------------------- /smpl_vis/utils/smpl_utils.py: -------------------------------------------------------------------------------- 1 | from .body_model_smpl import BodyModelSMPLH, BodyModelSMPLX 2 | def make_smplx(type="neu_fullpose", **kwargs): 3 | if type == "smpl": 4 | bm_kwargs = { 5 | "model_path": "../smpl_retarget/smpl_model", 6 | "model_type": "smpl", 7 | "gender": "neutral", 8 | "num_betas": 10, 9 | "create_body_pose": False, 10 | "create_betas": False, 11 | "create_global_orient": False, 12 | "create_transl": False, 13 | } 14 | bm_kwargs.update(kwargs) 15 | # model = SMPL(**bm_kwargs) 16 | model = BodyModelSMPLH(**bm_kwargs) 17 | elif type == "smplh": 18 | bm_kwargs = { 19 | "model_type": "smplh", 20 | "gender": kwargs.get("gender", "male"), 21 | "use_pca": False, 22 | "flat_hand_mean": False, 23 | } 24 | model = BodyModelSMPLH(model_path="../body_model", **bm_kwargs) 25 | elif type == "supermotion": 26 | # SuperMotion is trained on BEDLAM dataset, the smplx config is the same except only 10 betas are used 27 | bm_kwargs = { 28 | "model_type": "smplx", 29 | "gender": "neutral", 30 | "num_pca_comps": 12, 31 | "flat_hand_mean": False, 32 | } 33 | bm_kwargs.update(kwargs) 34 | model = BodyModelSMPLX(model_path="../smpl_retarget/smpl_model", **bm_kwargs) 35 | else: 36 | raise NotImplementedError 37 | 38 | return model --------------------------------------------------------------------------------