Scalable and General Whole-Body Control for Cross-Humanoid Locomotion

arXiv cs.RO / 4/15/2026

💬 OpinionSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper addresses cross-embodiment whole-body control for humanoid robots, aiming to remove the need for robot-specific training while maintaining robustness across different designs.
It proposes XHugWBC, a training framework that uses physics-consistent morphological randomization, semantically aligned observation/action spaces, and policy architectures that model robots’ morphological and dynamical properties.
The approach trains a single generalist policy over a broad distribution of randomized humanoid embodiments, yielding motion priors that enable zero-shot transfer to unseen robot designs.
Experiments reported include twelve simulated humanoids and seven real-world robots, showing strong generalization and robustness of the resulting universal controller.

Abstract

Learning-based whole-body controllers have become a key driver for humanoid robots, yet most existing approaches require robot-specific training. In this paper, we study the problem of cross-embodiment humanoid control and show that a single policy can robustly generalize across a wide range of humanoid robot designs with one-time training. We introduce XHugWBC, a novel cross-embodiment training framework that enables generalist humanoid control through: (1) physics-consistent morphological randomization, (2) semantically aligned observation and action spaces across diverse humanoid robots, and (3) effective policy architectures modeling morphological and dynamical properties. XHugWBC is not tied to any specific robot. Instead, it internalizes a broad distribution of morphological and dynamical characteristics during training. By learning motion priors from diverse randomized embodiments, the policy acquires a strong structural bias that supports zero-shot transfer to previously unseen robots. Experiments on twelve simulated humanoids and seven real-world robots demonstrate the strong generalization and robustness of the resulting universal controller.