Humanoid Hanoi: Investigating Shared Whole-Body Control for Skill-Based Box Rearrangement
🤖 AI Summary
This work addresses the degradation of control robustness in humanoid robots during long-horizon tasks involving skill reuse. To this end, the authors propose a framework based on reusable skill sequences that enables skill composition through a shared, task-agnostic whole-body controller (WBC). By integrating skill-level task orchestration with closed-loop execution data aggregation under domain randomization, the approach effectively mitigates distributional shifts induced by skill composition. Experiments on the Humanoid Hanoi simulation environment and fully autonomous box rearrangement tasks on the Digit V3 humanoid robot demonstrate that the proposed shared WBC architecture significantly outperforms non-shared baselines, enhancing consistency, stability, and overall robustness in long-horizon manipulation.
Application Category
📝 Abstract
We investigate a skill-based framework for humanoid box rearrangement that enables long-horizon execution by sequencing reusable skills at the task level. In our architecture, all skills execute through a shared, task-agnostic whole-body controller (WBC), providing a consistent closed-loop interface for skill composition, in contrast to non-shared designs that use separate low-level controllers per skill. We find that naively reusing the same pretrained WBC can reduce robustness over long horizons, as new skills and their compositions induce shifted state and command distributions. We address this with a simple data aggregation procedure that augments shared-WBC training with rollouts from closed-loop skill execution under domain randomization. To evaluate the approach, we introduce \emph{Humanoid Hanoi}, a long-horizon Tower-of-Hanoi box rearrangement benchmark, and report results in simulation and on the Digit V3 humanoid robot, demonstrating fully autonomous rearrangement over extended horizons and quantifying the benefits of the shared-WBC approach over non-shared baselines.
Problem
Research questions and friction points this paper is trying to address.
humanoid manipulation
whole-body control
long-horizon tasks
skill composition
robustness
Innovation
Methods, ideas, or system contributions that make the work stand out.
shared whole-body control
skill-based framework
long-horizon manipulation
data aggregation
humanoid robot
🔎 Similar Papers
No similar papers found.
