野外徒步:一种可扩展的感知跳跃框架用于人形机器人
Hiking in the Wild: A Scalable Perceptive Parkour Framework for Humanoids
AI总结:
本文提出了一种可扩展的感知跳跃框架,通过结合地形边缘检测和脚部体积点,实现人形机器人在复杂地形中的稳健徒步。
摘要:
在复杂、非结构化环境中实现稳健的人形徒步需要从反应式本体感觉转向主动式感知。然而,整合外源感觉仍是一个重大挑战:基于地图的方法因状态估计漂移而受挫;例如,基于激光雷达的方法在处理躯干抖动方面表现不佳。现有的端到端方法往往在可扩展性和训练复杂性方面存在问题;具体而言,一些先前使用虚拟障碍物的工作是按情况单独实现的。在本工作中,我们提出了\textit{Hiking in the Wild},一种可扩展、端到端的跳跃感知框架,专为稳健的人形徒步设计。为确保安全和训练稳定性,我们引入了两种关键机制:一种结合可扩展的\textit{地形边缘检测}与\textit{脚部体积点}的脚部安全机制,以防止在边缘上发生灾难性滑动;以及一种\textit{平面补丁采样}策略,通过生成可行的导航目标来缓解奖励黑客问题。我们的方法利用单阶段强化学习方案,将原始深度输入和本体感觉直接映射到关节动作,不依赖外部状态估计。在全尺寸人形机器人上的广泛实地实验表明,我们的策略能够在速度高达2.5米/秒的情况下实现复杂地形的稳健穿越。训练和部署代码已开源,以促进可重复研究和在真实机器人上最小化硬件修改的部署。
英文摘要:
Achieving robust humanoid locomotion in complex, unstructured environments requires transitioning from reactive proprioception to proactive perception. However, integrating exteroception remains a significant challenge: existing approaches often fail to achieve high-speed, stable locomotion across diverse, unstructured terrains. In this work, we present Hiking in the Wild, an extensible, end-to-end perceptive legged parkour framework designed for robust humanoid hiking. Our approach maps depth inputs and proprioception directly to joint actions, without relying on external localization and mapping. To ensure safety and training stability, we introduce two key mechanisms: a foothold safety mechanism combining extensible Terrain Edge Detection with Foot Volume Points to reduce catastrophic slippage on edges, and a Flat Patch Sampling strategy that mitigates reward hacking by generating feasible navigation targets. Extensive experiments on the Unitree G1 humanoid robot demonstrate that our policy enables robust traversal of complex terrains at speeds up to 2.5 m/s.
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。