Overview
- Field: Machine Learning / Robotics
- Authors: Lizhi Yang, Junheng Li, Nehar Poddar
- Published: 2026-06-04
- arXiv: 2606.06493
- Whole-body motion tracking with safety-filtered data
- Locomotion
- Fall recovery
Summary
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. The authors instead propose a compact, explicit interface that is intuitive, general, modular, and expressive enough for diverse manipulation skills.
To this end, they introduce HANDOFF, a single humanoid whole-body controller that follows this interface. HANDOFF is distilled via multi-teacher KL distillation under a context-conditioned gating scheme into a mixture-of-experts student from three complementary specialists:
Results
On the Unitree G1 humanoid, HANDOFF achieves state-of-the-art velocity tracking performance and provides one of the largest robust manipulation workspaces. The authors also demonstrate multiple natural-language-driven task deployments using a VLM-driven agentic planner, without task-specific data or controller fine-tuning.
--- *Auto-collected on 2026-06-07.*