mlsanityData to 7 Oct 2026

OSCAR: Data-Driven Operational Space Control for Adaptive and Robust Robot Manipulation

arXiv 2110.00704 ↗First posted 2 Oct 2021cs.RO cs.AI cs.LGcited by 7

Not on the map. Why?

Abstract (arXiv)

Learning performant robot manipulation policies can be challenging due to high-dimensional continuous actions and complex physics-based dynamics. This can be alleviated through intelligent choice of action space. Operational Space Control (OSC) has been used as an effective task-space controller for manipulation. Nonetheless, its strength depends on the underlying modeling fidelity, and is prone to failure when there are modeling errors. In this work, we propose OSC for Adaptation and Robustness (OSCAR), a data-driven variant of OSC that compensates for modeling errors by inferring relevant dynamics parameters from online trajectories. OSCAR decomposes dynamics learning into task-agnostic and task-specific phases, decoupling the dynamics dependencies of the robot and the extrinsics due to its environment. This structure enables robust zero-shot performance under out-of-distribution and rapid adaptation to significant domain shifts through additional finetuning. We evaluate our method on a variety of simulated manipulation problems, and find substantial improvements over an array of controller baselines. For more results and information, please visit https://cremebrule.github.io/oscar-web/.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Builds on 30 of its 53 references that are in the map. First posted on arXiv 2 Oct 2021.

Builds on

30 of 53 references.

References · 30

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,635
  3. 2018Neural Ordinary Differential Equationscited by 3,733
  4. 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,806
  5. 2021Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learningcited by 882
  6. 2020robosuite: A Modular Simulation Framework and Benchmark for Robot Learningcited by 484
  7. 2021RMA: Rapid Motor Adaptation for Legged Robotscited by 430
  8. 2021Brax -- A Differentiable Physics Engine for Large Scale Rigid Body Simulationcited by 375
  9. 2019Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learningcited by 205
  10. 2014Robots that can adapt like animalscited by 203
  11. 2018Closing the Sim-to-Real Loop: Adapting Simulation Randomization with Real World Experiencecited by 203
  12. 2018Distributed Distributional Deterministic Policy Gradientscited by 195
18 more
  1. 2018Making Sense of Vision and Touch: Self-Supervised Learning of Multimodal Representations for Contact-Rich Taskscited by 150
  2. 2020Simplifying Hamiltonian and Lagrangian Neural Networks via Explicit Constraintscited by 102
  3. 2019BayesSim: adaptive domain randomization via probabilistic inference for robotics simulatorscited by 78
  4. 2018Benchmarking Reinforcement Learning Algorithms on Real-World Robotscited by 64
  5. 2021DiSECt: A Differentiable Simulation Engine for Autonomous Robotic Cuttingcited by 62
  6. 2019Variable Impedance Control in End-Effector Space: An Action Space for Reinforcement Learning in Contact-Rich Taskscited by 56
  7. 2019Sim-to-Real Transfer for Biped Locomotioncited by 50
  8. 2019Learning Robotic Manipulation through Visual Planning and Actingcited by 39
  9. 2020ContactNets: Learning Discontinuous Contact Dynamics with Smooth, Implicit Representationscited by 38
  10. 2019Reinforcement Learning on Variable Impedance Controller for High-Precision Robotic Assemblycited by 33
  11. 2019Deep Lagrangian Networks for end-to-end learning of energy-based control for under-actuated systemscited by 32
  12. 2021LASER: Learning a Latent Action Space for Efficient Reinforcement Learningcited by 32
  13. 2021Sim-to-Real for Robotic Tactile Sensing via Physics-Based Simulation and Learned Latent Projectionscited by 27
  14. 2020Latent Space Roadmap for Visual Action Planning of Deformable and Rigid Object Manipulationcited by 25
  15. 2021Learning Visually Guided Latent Actions for Assistive Teleoperationcited by 12
  16. 2021Learning Contact Dynamics using Physically Structured Neural Networkscited by 10
  17. 2021Fundamental Challenges in Deep Learning for Stiff Contact Dynamicscited by 9
  18. 2020Learning User-Preferred Mappings for Intuitive Robot Controlcited by 4

Built on by

Cited by year (2026 to 7 Oct)

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

7 papers.

  1. 2024TRANSIC: Sim-to-Real Policy Transfer by Learning from Online Correctioncited by 50
  2. 2022Factory: Fast Contact for Robotic Assemblycited by 49
  3. 2023How to Learn and Generalize From Three Minutes of Data: Physics-Constrained and Uncertainty-Aware Neural Stochastic Differential Equationscited by 10
  4. 2022Just Round: Quantized Observation Spaces Enable Memory Efficient Learning of Dynamic Locomotioncited by 3
  5. 2023Decoupling Skill Learning from Robotic Control for Generalizable Object Manipulationcited by 2
  6. 2025Learning Tool-Aware Adaptive Compliant Control for Autonomous Regolith Excavationcited by 1
  7. 2026Feasibility-aware Imitation Learning from Observation with Multimodal Feedbackcited by 0