arXiv 2504.065852026-03-25cs.RO

通过联合扭矩空间扰动注入实现人形机器人类 locomotion 策略的 sim-to-real

Sim-to-Real of Humanoid Locomotion Policies via Joint Torque Space Perturbation Injection

X Technologies（1X Technologies）；Advanced Institute of Convergence Technology (AICT), Suwon, Republic of Korea（融合技术研究院（AICT），Suwon，韩国）

Junhyeok Rui Cha, Woohyun Cha, Jaeyong Shin, Donghyeon Kim, Jaeheung Park

备注：: This work has been submitted to the IEEE for possible publication

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出了一种新的 sim-to-real 方法，通过在训练阶段向用于正向模拟的输入关节扭矩添加状态依赖扰动，以模拟更广泛的真实差距。实验表明，该方法能提升人形机器人在复杂真实差距下的鲁棒性。

摘要：

本文提出了一种新的 sim-to-real 方法，用于训练具有模拟经验的控制策略。现有 legged robots 的 sim-to-real 方法大多依赖于领域随机化方法，即在训练期间随机化固定的有限模拟参数集。相反，我们的方法在训练阶段向用于正向模拟的输入关节扭矩添加状态依赖扰动。这些状态依赖扰动旨在模拟比随机化固定参数集所捕获的更广泛的真实差距。实验结果表明，我们的方法使人形机器人 locomotion 策略在面对训练域中未见过的复杂真实差距时表现出更高的鲁棒性。

英文摘要：

This paper proposes a novel alternative to existing sim-to-real methods for training control policies with simulated experiences. Prior sim-to-real methods for legged robots mostly rely on the domain randomization approach, where a fixed finite set of simulation parameters is randomized during training. Instead, our method adds state-dependent perturbations to the input joint torque used for forward simulation during the training phase. These state-dependent perturbations are designed to simulate a broader range of reality gaps than those captured by randomizing a fixed set of simulation parameters. Experimental results show that our method enables humanoid locomotion policies that achieve greater robustness against complex reality gaps unseen in the training domain.

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。