Skip to content

论文arXiv 2611.260542 天前

K1人形机器人无参考滑板控制:终止塑形强化学习框架 ​

本文由 AI 根据公开内容整理,可能有误,请以原文为准:Reference-Free Humanoid Skateboarding via Termination-Shaped Reinforcement Learning(arXiv 2611.26054)

核心内容 ​

针对人形机器人在物体介导运动(Object-mediated Locomotion)中的挑战,研究团队提出一种无参考数据的深度强化学习框架,成功使 K1 人形机器人学会滑板。该方法摒弃了传统依赖人类演示或解析转向模型的系统辨识流程,采用 Model-free 学习范式,通过将任务奖励与 Exploit-blocking Terminations 结合,解决了被动可移动平台与地面交互的复杂动力学控制问题。

技术核心在于设计了一系列约束性终止条件,包括支撑脚脱离板面、双脚站立、过度扭转、人机分离及板体不稳定等。这些约束在训练过程中动态变化,迫使策略网络规避作弊行为,从而形成从基础站姿稳定到持续滑行的隐式课程学习路径。该架构验证了在缺乏先验数据条件下,仅凭环境交互反馈即可实现人形机器人与外部被动工具的复杂协同运动。

关键看点 ​

  • 无参考学习:采用 Model-free RL 框架,无需人类演示数据或解析转向模型即可训练 K1 机器人滑板技能
  • 终止塑形机制:引入 Exploit-blocking Terminations,将支撑脚脱落、过度扭转等违规行为设为训练终止条件
  • 隐式课程演进:利用约束条件的动态分布变化,自动引导策略从静态姿态控制向持续板机协同运动过渡
  • 物体介导运动:突破传统步行/跑步局限,解决人形本体与被动可移动平台同时交互的动力学耦合难题

工程与产业观察 ​

该研究为具身智能工具使用提供了纯RL范式参考。Exploit-blocking Terminations 机制有效解决了Sim-to-Real中常见的奖励欺骗问题,证明无需昂贵动捕演示数据,仅靠精心设计的终止约束即可习得复杂物体交互技能,降低了人形机器人非标准运动能力的开发门槛。

英文摘要 (Abstract) ​

Humanoid locomotion research has advanced rapidly on walking and running, but \emph {object-mediated} locomotion, in which a robot must coordinate its body with a passive external object, remains difficult. One challenging task is skateboarding, which requires simultaneous interaction with the ground and a passive but movable platform to achieve forward motion. Current methods for tackling skateboarding with humanoid robots either require human demonstrations or utilise analytic steering models that require system identification. In this paper, we introduce a deep reinforcement learning framework that does not rely on prior data such as demonstrations and enables robots such as the K1 humanoid to learn to skateboard using model-free learning. The method combines task rewards with exploit-blocking terminations. We introduce a number of constraints restricting behaviours such as losing the planted board foot, standing with both feet on the board, excessive twisting, separation from the board, and unstable board motion. Their changing prevalence during training produces an implicit progression from stance control towards sustained robot--board locomotion.

作者:Ganeshvar Soenessardien, Michael Przystupa, Kevin Sebastian Luck

基于本页技术方向关联的整机参数、核心零部件、实测报告与同赛道选型横评。

外部来源

本页事实对应的公开页面,以原文为准。(点击将离开本站)

  1. arXiv 2611.26054:Reference-Free Humanoid Skateboarding via Termination-Shaped Reinforcement Learning ↗访问 2026-11-20