└── readme.md /readme.md: -------------------------------------------------------------------------------- 1 | A list of papers on model-based control that I have read so far. The ones that I particularly liked are marked with :star:. 2 | 3 | **Model Learning and Model-predictive Control (MPC)** 4 | - [Learning model-based planning from scratch](https://arxiv.org/abs/1707.06170), R. Pascanu and Y.Li et al., Arxiv 2017 5 | - [Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models](https://arxiv.org/abs/1805.12114), K. Chua et al., NIPS 2018 6 | - [SOLAR: Deep Structured Latent Representations for Model-Based Reinforcement Learning](https://arxiv.org/abs/1808.09105), M. Zhang et al., arXiv 2018 7 | - [Interaction Networks for Learning about Objects, Relations and Physics](https://arxiv.org/abs/1612.00222), P. Battaglia et al., NIPS 2016 :star: 8 | - [Learning Particle Dynamics for Manipulating Rigid Bodies, Deformable Objects, and Fluids](https://arxiv.org/abs/1810.01566), Y. Li et al., arXiv 2018 :star: 9 | - [Propagation Networks for Model-Based Control Under Partial Observation](https://arxiv.org/abs/1809.11169), Y. Li et al., arXiv 2018 10 | - [Efficient Model-Based Deep Reinforcement Learning with Variational State Tabulation](https://arxiv.org/abs/1802.04325), D. Corneil et al., ICML 2018 :star: 11 | - [A Compositional Object-Based Approach to Learning Physical Dynamics](https://arxiv.org/abs/1612.00341), M. Chang et al., ICLR 2017 :star: 12 | - [SPNets: Differentiable Fluid Dynamics for Deep Neural Networks](https://arxiv.org/abs/1806.06094), C. Schenck et al., CoRL 2018 13 | - [Augmenting Physical Simulators with Stochastic Neural Networks: Case Study of Planar Pushing and Bouncing](https://arxiv.org/abs/1808.03246), A. Ajay et al., IROS 2018 14 | - [Graph networks as learnable physics engines for inference and control](https://arxiv.org/abs/1806.01242), A. Sanchez-Gonzalez et al., arXiv 2018 :star: 15 | - [Learning Latent Dynamics for Planning from Pixels](https://arxiv.org/abs/1811.04551), D. Hafner et al., arXiv 2018 16 | 17 | 18 | **Pixel to Control** 19 | - [Embed to Control: A Locally Linear Latent Dynamics Model for Control from Raw Images](https://arxiv.org/abs/1506.07365), M. Watter and J. Springenberg et al., NIPS 2015 :star: 20 | - [Robust Locally-Linear Controllable Embedding](https://arxiv.org/abs/1710.05373), E. Banijamali et al., AISTATS 2018 21 | - [End-to-End Training of Deep Visuomotor Policies](https://arxiv.org/abs/1504.00702), S. Levine et al., JMLR 2016 22 | - [Unsupervised Learning for Physical Interaction through Video Prediction](https://arxiv.org/abs/1605.07157), C. Finn et al., NIPS 2016 23 | - [Deep Spatial Autoencoders for Visuomotor Learning](https://arxiv.org/abs/1509.06113), C. Finn et al., ICRA 2016 24 | - [Deep Visual Foresight for Planning Robot Motion](https://arxiv.org/abs/1610.00696), C. Finn et al., ICRA 2017 25 | - [Learning Plannable Representations with Causal InfoGAN](https://arxiv.org/abs/1807.09341), T. Kurutach et al., NIPS 2018 :star: 26 | - [Learning Latent Dynamics for Planning from Pixels](https://arxiv.org/abs/1811.04551), D. Hafner et al., arXiv 2018 27 | - [SE3-Pose-Nets: Structured Deep Dynamics Models for Visuomotor Planning and Control](https://arxiv.org/abs/1710.00489), A. Byravan et al., ICRA 2018 :star: 28 | 29 | **Model-based + Model-free** 30 | - [MBMF: Model-Based Priors for Model-Free Reinforcement Learning](https://arxiv.org/abs/1709.03153), S. Bansal et al., CoRL 2017 31 | - [Continuous deep q-learning with model-based acceleration](https://arxiv.org/abs/1603.00748), S. Gu et al., ICML 2016 32 | - [Recurrent World Models Facilitate Policy Evolution](https://arxiv.org/abs/1809.01999), D. Ha et al., NIPS 2018 33 | 34 | **Learned Optimal Control** 35 | - [Embed to Control: A Locally Linear Latent Dynamics Model for Control from Raw Images](https://arxiv.org/abs/1506.07365), M. Watter and J. Springenberg et al., NIPS 2015 :star: 36 | - [Robust Locally-Linear Controllable Embedding](https://arxiv.org/abs/1710.05373), E. Banijamali et al., AISTATS 2018 37 | - [Differentiable MPC for End-to-end Planning and Control](https://arxiv.org/abs/1810.13400), B. Amos et al., NIPS 2018 :star: 38 | - [Path Integral Networks: End-to-End Differentiable Optimal Control](https://arxiv.org/abs/1706.09597), Okada et al., NIPS 2017 39 | - [Learning to Control a Low-Cost Manipulator using Data-Efficient Reinforcement Learning](http://www.roboticsproceedings.org/rss07/p08.pdf), M. Deisenroth et al., RSS 2011 40 | - [SOLAR: Deep Structured Latent Representations for Model-Based Reinforcement Learning](https://arxiv.org/abs/1808.09105), M. Zhang et al., arXiv 2018 41 | 42 | **State Estimation** 43 | - [Deep Variational Bayes Filters: Unsupervised Learning of State Space Models from Raw Data](https://arxiv.org/abs/1605.06432), M. Carl et al., ICLR 2017 :star: 44 | - [Deep Kalman Filters](https://arxiv.org/abs/1511.05121) R. G. Krishnan et al., arXiv 2015 45 | - [Differentiable Particle Filters: End-to-End Learning with Algorithmic Priors](https://arxiv.org/abs/1805.11122), R. Jonschkowski et al., RSS 2018 :star: 46 | - [QMDP-Net: Deep Learning for Planning under Partial Observability](https://arxiv.org/abs/1703.06692), P. Karkus et al., NIPS 2017 47 | - [Generative Temporal Models with Spatial Memory for Partially Observed Environments](https://arxiv.org/abs/1804.09401), M. Fraccaro et al., ICML 2018 :star: 48 | 49 | **Survey** 50 | - [Learning Physical Dynamical Systems for Prediction and Control: A Survey](https://www.cs.princeton.edu/courses/archive/spring18/cos598B/public/projects/LiteratureReview/COS598B_spr2018_PhysicalDynamicalSystems.pdf), J. LaChance, 2018 51 | 52 | **Koopman Theory** 53 | - [Learning Koopman Invariant Subspaces for Dynamic Mode Decomposition](https://arxiv.org/abs/1710.04340), N Takeishi et al., NIPS 2017 :star: 54 | - [Deep Dynamical Modeling and Control of Unsteady Fluid Flows](https://arxiv.org/abs/1805.07472), J. Morton et al., NIPS 2018 55 | - [Deep learning for universal linear embeddings of nonlinear dynamics](https://arxiv.org/abs/1712.09707), B Lusch et al., Nature Communications 2018 56 | - [Data-driven discovery of Koopman eigenfunctions for control](https://arxiv.org/abs/1707.01146), E. Kaiser et al., arXiv 2017 57 | 58 | **Optimal Control** 59 | - [Control-Limited Differential Dynamic Programming](https://homes.cs.washington.edu/~todorov/papers/TassaICRA14.pdf), Y. Tassa et al., ICRA 2014 60 | 61 | **Other Resources** 62 | - [Learning Dynamical System Models from Data](http://rll.berkeley.edu/deeprlcoursesp17/docs/week_3_lecture_1_dynamics_learning.pdf), Sergey Levine, CS 294-112: Deep Reinforcement Learning 63 | - [EE263: Introduction to Linear Dynamical Systems](http://ee263.stanford.edu/lectures.html) 64 | - [CS 287: Advanced Robotics](https://people.eecs.berkeley.edu/~pabbeel/cs287-fa15/) 65 | - [Control Bootcamp](https://www.youtube.com/watch?v=Pi7l8mMjYVE&list=PLMrJAkhIeNNR20Mz-VpzgfQs5zrYi085m), Steve Brunton, 2017 66 | - [STUDYWOLF blog](https://studywolf.wordpress.com/), Travis DeWolf 67 | --------------------------------------------------------------------------------