Papers

无参考数据的人形机器人滑板强化学习方法

Reference-Free Humanoid Skateboarding via Termination-Shaped Reinforcement Learning

Why we picked it

不靠人类示范,纯靠强化学习约束设计就能让 K1 人形机器人踩滑板前进,比之前需要动捕数据或解析模型的方案更省事。做足式机器人全身控制、人机物体交互研究的人可以看看这个 termination shaping 思路,能直接搬到别的任务上。

arXiv 论文《Reference-Free Humanoid Skateboarding via Termination-Shaped Reinforcement Learning》提出一套免示范数据的深度强化学习框架,让 K1 等人形机器人学会滑板。此前的滑板方案要么依赖人类动作示范,要么需要系统辨识的解析转向模型,该方法用 model-free 学习绕开了这两类前提。

Body

技术上,框架将任务奖励与'利用阻断式终止'(exploit-blocking terminations)结合,设置多项约束:禁止踩板脚离板、禁止双脚同时站在板上、限制过度扭转、限制与滑板分离、限制滑板运动不稳。这些约束在训练过程中占比不断变化,形成从站位控制到持续机器人—滑板协同运动的隐式课程。

滑板属于'物体介导运动'(object-mediated locomotion)——机器人需同时协调身体与一个可移动的被动外部物体,比走路、跑步更难。该方法对做多物体交互控制的机器人团队有直接参考价值。

Original · arXiv cs.LG

Reference-Free Humanoid Skateboarding via Termination-Shaped Reinforcement Learning

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.

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include K1, reinforcement learning, humanoid robots, sim-to-real.

Source attribution: curated from arXiv cs.LG (@Ganeshvar Soenessardien, Michael Przystupa, Kevin Sebastian Luck).