arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2603.06775cs.RO
收藏

HybridMimic: 人形机器人运动模仿的混合强化学习-质心控制

HybridMimic: Hybrid RL-Centroidal Control for Humanoid Motion Mimicking

Department of Computer Science, Purdue University(普渡大学计算机科学系);School of Mechanical Engineering, Purdue University(普渡大学机械工程学院)

Ludwig Chee-Ying Tay, I-Chia Chang, Yan Gu

AI总结:

HybridMimic通过结合强化学习与质心模型控制,有效提升人形机器人在分布外环境下的运动模仿性能。

摘要:

运动模仿,即鼓励控制器模仿人类运动,通过强化学习(RL)帮助人形机器人学习复杂任务。尽管标准RL框架在运动能力方面表现出色,但它们在部署过程中往往绕过了对机器人动力学的显式推理,这种设计选择在机器人遇到分布外环境时可能导致物理上不可行的指令。通过整合模型驱动的原则,混合方法可以提高性能;然而,现有方法通常依赖于预定义的接触时间,限制了其通用性。本文介绍了HybridMimic,这是一种框架,其中学习的策略通过预测连续接触状态和期望质心速度动态调节基于质心模型的控制器。该架构利用质心动力学的物理基础,生成在域转移下仍可行的前馈扭矩。通过物理指导的奖励,策略被训练以高效利用质心控制器的优化,通过输出精确的控制目标和参考扭矩。通过在Booster T1人形机器人上的硬件实验,HybridMimic将平均基底位置跟踪误差比最先进的RL基线减少了13%,证明了动态感知部署的鲁棒性。

英文摘要:

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 impressive locomotion agility, they often bypass explicit reasoning about robot dynamics during deployment, which is a design choice that can lead to physically infeasible commands when the robot encounters out-of-distribution environments. By integrating model-based principles, hybrid approaches can improve performance; however, existing methods typically rely on predefined contact timing, limiting their versatility. This paper introduces HybridMimic, a framework in which a learned policy dynamically modulates a centroidal-model-based controller by predicting continuous contact states and desired centroidal velocities. This architecture exploits the physical grounding of centroidal dynamics to generate feedforward torques that remain feasible even under domain shift. Using physics-informed rewards, the policy is trained to efficiently utilize the centroidal controller's optimization by outputting precise control targets and reference torques. Through hardware experiments on the Booster T1 humanoid, HybridMimic reduces the average base position tracking error by 13\% compared to a state-of-the-art RL baseline, demonstrating the robustness of dynamics-aware deployment.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑