HANDOFF: Humanoid Agentic Task-Space Whole-Body Control via Distilled Complementary Teachers

论文概要

研究领域: ML
作者: Lizhi Yang, Junheng Li, Nehar Poddar
发布时间: 2026-06-04
arXiv: 2606.06493

中文摘要

人形机器人在现实世界部署时，命令空间（即任务规划与全身控制之间的接口）的选择至关重要。现有全身控制器通常需要密集的运动学或空间参考，而规划器难以从任务语义中合成这些参考。我们提出了一种紧凑、显式的接口，直观、通用、模块化且足够表达多样操作技能。为此，我们引入HANDOFF——一种单一的人形全身控制器，通过多教师KL蒸馏和上下文条件门控机制，从三个互补专家（全身运动跟踪+安全过滤数据、运动、跌倒恢复）蒸馏为混合专家学生。在Unitree G1上，HANDOFF达到SOTA速度跟踪性能，并提供最大的鲁棒操作工作空间之一。我们还通过VLM驱动的智能体规划器展示了多个自然语言驱动的任务部署，无需任务特定数据或控制器微调。

原文摘要

For a humanoid robot to be deployed in the real world, the choice of command space (i.e., the interface between task planning and whole-body control) is crucial. Existing whole-body controllers typically demand dense kinematic or spatial references that planners struggle to synthesize from task semantics. We instead propose a compact, explicit interface that is intuitive, general, modular, and expressive enough for diverse manipulation skills. To this end, we introduce HANDOFF, a single humanoid whole-body controller that follows this interface and is distilled via multi-teacher KL distillation under a context-conditioned gating scheme into a mixture-of-experts student from three complementary specialists: whole-body motion tracking with safety-filtered data, locomotion, and fall-recovery....

自动采集于 2026-06-07

#论文 #arXiv #ML #小凯