paper-with-me

홈 › Papers

Efficiently Learning Robust Torque-based Locomotion Through Reinforcement with Model-Based Supervision

2026-01-22 · Yashuai Yan, Tobias Egle, Christian Ott, Dongheui Lee arxiv

We propose a control framework that integrates model-based bipedal locomotion with residual reinforcement learning (RL) to achieve robust and adaptive walking in the presence of real-world uncertainties. Our approach leverages a model-based controller, comprising a Divergent Component of Motion (DCM) trajectory planner and a whole-body controller, as a reliable base policy. To address the uncertainties of inaccurate dynamics modeling and sensor noise, we introduce a residual policy trained through RL with domain randomization. Crucially, we employ a model-based oracle policy, which has privileged access to ground-truth dynamics during training, to supervise the residual policy via a novel supervised loss. This supervision enables the policy to efficiently learn corrective behaviors that compensate for unmodeled effects without extensive reward shaping. Our method demonstrates improved robustness and generalization across a range of randomized conditions, offering a scalable solution for sim-to-real transfer in bipedal locomotion.

📄 PDF Abstract BibTeX arXiv:2601.16109

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

GainAdaptor: Learning Quadrupedal Locomotion with Dual Actors for Adaptable and Energy-Efficient Walking on Various Terrains

2024-12-12 · Mincheol Kim, Nahyun Kwon, Jung-Yup Kim

Deep reinforcement learning (DRL) has emerged as an innovative solution for controlling legged robots in challenging environments using minimalist architectures. Traditional control methods for legged robots, such as inv…

Deep Reinforcement Learning

Learning Torque Control for Quadrupedal Locomotion

2022-03-10 · Shuxiao Chen, Bike Zhang, Mark W. Mueller, Akshara Rai 외

Reinforcement learning (RL) has become a promising approach to developing controllers for quadrupedal robots. Conventionally, an RL design for locomotion follows a position-based paradigm, wherein an RL policy outputs ta…

PositionReinforcement Learning (RL)

Latent Action Priors for Locomotion with Deep Reinforcement Learning

2024-10-04 · Oliver Hausdörfer, Alexander von Rohr, Éric Lefort, Angela Schoellig

Deep Reinforcement Learning (DRL) enables robots to learn complex behaviors through interaction with the environment. However, due to the unrestricted nature of the learning algorithms, the resulting solutions are often …

Deep Reinforcement LearningEfficient ExplorationImitation LearningInductive Bias+2

Achieving Precise and Reliable Locomotion with Differentiable Simulation-Based System Identification

2025-08-06 · Vyacheslav Kovalev, Ekaterina Chaikovskaia, Egor Davydenko, Roman Gorbachev arxiv

Accurate system identification is crucial for reducing trajectory drift in bipedal locomotion, particularly in reinforcement learning and model-based control. In this paper, we present a novel control framework that inte…

Reinforcement Learning

HybridMimic: Hybrid RL-Centroidal Control for Humanoid Motion Mimicking

2026-03-06 · Ludwig Chee-Ying Tay, I-Chia Chang, Yan Gu arxiv

Motion mimicking, i.e., encouraging the control policy to mimic human motion, facilitates the learning of complex tasks via reinforcement learning (RL) for humanoid robots. Although standard RL frameworks demonstrate imp…

Reinforcement Learning