mlsanityData to 14 Oct 2026

Composable Energy Policies for Reactive Motion Generation and Reinforcement Learning

arXiv 2105.04962 ↗First posted 11 May 2021cs.RO cs.LGcited by 16

Not on the map. Why?

Abstract (arXiv)

Reactive motion generation problems are usually solved by computing actions as a sum of policies. However, these policies are independent of each other and thus, they can have conflicting behaviors when summing their contributions together. We introduce Composable Energy Policies (CEP), a novel framework for modular reactive motion generation. CEP computes the control action by optimization over the product of a set of stochastic policies. This product of policies will provide a high probability to those actions that satisfy all the components and low probability to the others. Optimizing over the product of the policies avoids the detrimental effect of conflicting behaviors between policies choosing an action that satisfies all the objectives. Besides, we show that CEP naturally adapts to the Reinforcement Learning problem allowing us to integrate, in a hierarchical fashion, any distribution as prior, from multimodal distributions to non-smooth distributions and learn a new policy given them.

Key facts. Cited by 16 papers in the map (data through 14 Oct 2026). Builds on 39 of its 95 references that are in the map. First posted on arXiv 11 May 2021.

Builds on

39 of 95 references.

References · 39

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  3. 2015Continuous control with deep reinforcement learningcited by 3,407
  4. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,689
  5. 2018Addressing Function Approximation Error in Actor-Critic Methodscited by 2,417
  6. 2017Reinforcement Learning with Deep Energy-Based Policiescited by 669
  7. 2018Reinforcement Learning and Control as Probabilistic Inference: Tutorial and Reviewcited by 482
  8. 2018Residual Reinforcement Learning for Robot Controlcited by 274
  9. 2020Accelerating Reinforcement Learning with Learned Skill Priorscited by 223
  10. 2018Residual Policy Learningcited by 164
  11. 2018Making Sense of Vision and Touch: Self-Supervised Learning of Multimodal Representations for Contact-Rich Taskscited by 154
  12. 2020Parrot: Data-Driven Behavioral Priors for Reinforcement Learningcited by 138
27 more
  1. 2017Stochastic Neural Networks for Hierarchical Reinforcement Learningcited by 124
  2. 2018A micro Lie theory for state estimation in roboticscited by 124
  3. 2019MCP: Learning Composable Hierarchical Control with Multiplicative Compositional Policiescited by 108
  4. 2018Composable Deep Reinforcement Learning for Robotic Manipulationcited by 72
  5. 2016Learning and Transfer of Modulated Locomotor Controllerscited by 59
  6. 2020Neural Dynamic Policies for End-to-End Sensorimotor Learningcited by 58
  7. 2020A Distributional View on Multi-Objective Policy Optimizationcited by 52
  8. 2020MushroomRL: Simplifying Reinforcement Learning Researchcited by 50
  9. 2017Continuous-Time Gaussian Process Motion Planning via Probabilistic Inferencecited by 49
  10. 2020A Boolean Task Algebra for Reinforcement Learningcited by 48
  11. 2021STORM: An Integrated Framework for Fast Joint-Space Model-Predictive Control for Reactive Manipulationcited by 47
  12. 2018Riemannian Motion Policiescited by 46
  13. 2020ImitationFlow: Learning Deep Stable Stochastic Dynamic Systems by Normalizing Flowscited by 33
  14. 2017Real-time Perception meets Reactive Motion Generationcited by 30
  15. 2020Stein Variational Model Predictive Controlcited by 30
  16. 2020Residual Policy Learning for Shared Autonomycited by 26
  17. 2020Compositional Visual Generation and Inference with Energy Based Modelscited by 26
  18. 2018RMPflow: A Computational Graph for Automatic Motion Policy Generationcited by 21
  19. 2021Value Iteration in Continuous Actions, States and Timecited by 18
  20. 2021Benchmarking Structured Policies and Policy Optimization for Real-World Dexterous Object Manipulationcited by 15
  21. 2019Stochastic Optimal Control as Approximate Input Inferencecited by 14
  22. 2020Learning Vision-based Reactive Policies for Obstacle Avoidancecited by 12
  23. 2020Generalized Nonlinear and Finsler Geometry for Roboticscited by 10
  24. 2021RMP2: A Structured Composable Policy Class for Robot Learningcited by 8
  25. 2020Geometric Fabrics for the Acceleration-based Design of Robotic Motioncited by 7
  26. 2020Optimization Fabricscited by 4
  27. 2021Composable Geometric Motion Policies using Multi-Task Pullback Bundle Dynamical Systemscited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 1; 2022: 6; 2023: 4; 2024: 2; 2025: 2; 2026: 1.

16 papers.

Sort:

By area:

  1. 2023Video Language Planningcited by 162
  2. 2022SE(3)-DiffusionFields: Learning smooth cost functions for joint grasp and motion optimization through diffusioncited by 127
  3. 2022SE(3)-Equivariant Relational Rearrangement with Neural Descriptor Fieldscited by 51
  4. 2023Compositional Diffusion-Based Continuous Constraint Solverscited by 40
  5. 2024Deep Generative Models in Robotics: A Survey on Learning from Multimodal Demonstrationscited by 35
  6. 2023Probabilistic Adaptation of Text-to-Video Modelscited by 29
  7. 2022Learning Implicit Priors for Motion Optimizationcited by 20
  8. 2024Compositional Generative Inverse Designcited by 20
  9. 2022Regularized Deep Signed Distance Fields for Reactive Motion Generationcited by 16
  10. 2025Multi-Modal Manipulation via Multi-Modal Policy Consensuscited by 13
  11. 2021Learning Stable Vector Fields on Lie Groupscited by 12
  12. 2022Hierarchical Policy Blending as Inference for Reactive Robot Controlcited by 8
4 more
  1. 2022Reactive Motion Generation on Learned Riemannian Manifoldscited by 6
  2. 2023Inferring Relational Potentials in Interacting Systemscited by 5
  3. 2026Latent Energy Action Planning with World Modelscited by 4
  4. 2025Trajectory Entropy Reinforcement Learning for Robust Robot Motor Skill Learningcited by 0