HANDOFF: 通过蒸馏互补教师实现人形机器人任务空间全身控制
HANDOFF: Humanoid Agentic Task-Space Whole-Body Control via Distilled Complementary Teachers
AI总结：
提出HANDOFF框架，通过多教师KL蒸馏和上下文门控机制，将全身运动跟踪、行走和跌倒恢复三个专家策略融合为混合专家学生策略，实现基于紧凑显式接口的全身控制，在Unitree G1上达到先进的速度跟踪性能并扩展了操作工作空间。
摘要：
对于要在现实世界中部署的人形机器人，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。现有的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，该接口直观、通用、模块化且具有足够的表达能力，适用于多种操作技能。为此，我们引入了HANDOFF，这是一个单一的人形全身控制器，遵循该接口，并通过多教师KL蒸馏，在上下文条件门控方案下，从三个互补专家（具有安全过滤数据的全身运动跟踪、行走和跌倒恢复）中蒸馏出混合专家学生。在Unitree G1上，HANDOFF达到了最先进的速度跟踪性能，并提供了最大的鲁棒操作工作空间之一。我们进一步通过多个自然语言驱动的任务执行演示了硬件可行性，这些任务由VLM驱动的智能体规划器提供支持，无需特定任务数据或控制器微调。
英文摘要：
Humanoid loco-manipulation benefits from one controller to coordinate locomotion, arm movements, and fall recovery. These behaviors are difficult to learn together from scratch because they require different skills and objectives. We present HANDOFF, a training architecture that distills motion-tracking, locomotion, and recovery teachers into one 29-DoF policy, using context-conditioned KL losses over separate action slices and a soft mixture of experts. Commanded velocity continuously blends body supervision, while a binary recovery flag assigns the full action to the recovery teacher. The resulting controller takes a compact 10-D task-space command rather than a dense kinematic stream and does not switch policies at runtime. On the Unitree G1, HANDOFF matches state-of-the-art velocity tracking and obtains the largest robust manipulation workspace in a matched adapted-interface benchmark. The same controller executes natural-language-driven, multi-stage loco-manipulation tasks in simulation and hardware experiments without further data collection or fine-tuning.
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。
