WholeBodyWAM: Generalizing Pre-trained World-Action Priors to Humanoid Loco-Manipulation via WBC-Grounded Coordination

2026-09-15

AI Summary

本文通过引入WholeBodyWAM解决人形机器人全身操作问题，该方法基于预训练的世界-动作先验，并结合全身控制器语义进行协调，以实现更泛化的操作。

Abstract

World Action Models (WAMs) offer a promising approach to general-purpose robot manipulation by jointly modeling visual dynamics and actions. However, most WAM studies focus on tabletop or arm-centric manipulation, while humanoid loco-manipulation remains less explored. To address this gap, we introduce WholeBodyWAM, which jointly predicts future visual dynamics, manipulation actions, and whole-body control intents for generalizable humanoid loco-manipulation. It preserves pre-trained world-action priors while grounding heterogeneous whole-body controller (WBC) semantics and coordinating whole-body behavior. Extensive experiments show that WholeBodyWAM achieves an overall simulation task success rate of 91.9%, with a 0.23 improvement in real-world out-of-distribution task progress and a 70% reduction in success-rate variance across WBCs relative to the respective baselines. These results suggest a path toward scalable humanoid whole-body intelligence by extending pre-trained world-action priors through structured WBC grounding and coordination, rather than relearning whole-body behavior from scratch. Project page: https://wholebodywam.github.io/.

Problem

• WholeBodyWAM
• humanoid loco-manipulation
• world-action priors
• whole-body control

Innovation

• WholeBodyWAM
• World-Action Priors
• Whole-Body Control
• Humanoid Loco-Manipulation
• WBC-Grounded Coordination