HuBE: Cross-Embodiment Human-like Behavior Execution for Humanoid Robots

TL;DR

HuBE is a novel bi-level framework that enables humanoid robots to generate human-like, contextually appropriate behaviors with cross-embodiment adaptability, improving motion similarity and efficiency.

Contribution

The paper introduces HuBE, a bi-level closed-loop framework with a new dataset and bone scaling augmentation for cross-embodiment human-like behavior execution.

Findings

HuBE significantly outperforms baselines in motion similarity.

HuBE enhances behavioral appropriateness across platforms.

The framework improves computational efficiency.

Abstract

Achieving both behavioral similarity and appropriateness in human-like motion generation for humanoid robot remains an open challenge, further compounded by the lack of cross-embodiment adaptability. To address this problem, we propose HuBE, a bi-level closed-loop framework that integrates robot state, goal poses, and contextual situations to generate human-like behaviors, ensuring both behavioral similarity and appropriateness, and eliminating structural mismatches between motion generation and execution. To support this framework, we construct HPose, a context-enriched dataset featuring fine-grained situational annotations. Furthermore, we introduce a bone scaling-based data augmentation strategy that ensures millimeter-level compatibility across heterogeneous humanoid robots. Comprehensive evaluations on multiple commercial platforms demonstrate that HuBE significantly improves motion similarity, behavioral appropriateness, and computational efficiency over state-of-the-art baselines, establishing a solid foundation for transferable and human-like behavior execution across diverse humanoid robots.