SkillBlender：通过技能融合实现多功能人形机器人全身运动与操作

SkillBlender: Towards Versatile Humanoid Whole-Body Loco-Manipulation via Skill Blending

June 11, 2025

作者: Yuxuan Kuang, Haoran Geng, Amine Elhafsi, Tan-Dzung Do, Pieter Abbeel, Jitendra Malik, Marco Pavone, Yue Wang

cs.AI

摘要

人形机器人凭借其灵活性和类人形态，在多样化环境中执行日常任务方面展现出巨大潜力。近期研究通过优化控制或强化学习，在全身控制与移动操作方面取得了显著进展。然而，这些方法需要对每项任务进行繁琐的特定调优才能达到满意效果，限制了其在日常场景中应对多样化任务时的通用性和扩展性。为此，我们提出了SkillBlender，一种新颖的分层强化学习框架，旨在实现多功能的人形移动操作。SkillBlender首先预训练目标导向、任务无关的基础技能，随后动态融合这些技能，以最少的任务特定奖励工程完成复杂的移动操作任务。我们还引入了SkillBench，一个并行、跨实体、多样化的模拟基准，包含三种实体、四项基础技能及八项具有挑战性的移动操作任务，并配备了一套平衡准确性与可行性的科学评估指标。大量模拟实验表明，我们的方法显著优于所有基线，同时自然规范行为以避免奖励欺骗，从而在日常生活场景中为多样化的移动操作任务生成更准确、更可行的动作。我们的代码与基准将开源，以促进未来研究。项目页面：https://usc-gvl.github.io/SkillBlender-web/。

English

Humanoid robots hold significant potential in accomplishing daily tasks across diverse environments thanks to their flexibility and human-like morphology. Recent works have made significant progress in humanoid whole-body control and loco-manipulation leveraging optimal control or reinforcement learning. However, these methods require tedious task-specific tuning for each task to achieve satisfactory behaviors, limiting their versatility and scalability to diverse tasks in daily scenarios. To that end, we introduce SkillBlender, a novel hierarchical reinforcement learning framework for versatile humanoid loco-manipulation. SkillBlender first pretrains goal-conditioned task-agnostic primitive skills, and then dynamically blends these skills to accomplish complex loco-manipulation tasks with minimal task-specific reward engineering. We also introduce SkillBench, a parallel, cross-embodiment, and diverse simulated benchmark containing three embodiments, four primitive skills, and eight challenging loco-manipulation tasks, accompanied by a set of scientific evaluation metrics balancing accuracy and feasibility. Extensive simulated experiments show that our method significantly outperforms all baselines, while naturally regularizing behaviors to avoid reward hacking, resulting in more accurate and feasible movements for diverse loco-manipulation tasks in our daily scenarios. Our code and benchmark will be open-sourced to the community to facilitate future research. Project page: https://usc-gvl.github.io/SkillBlender-web/.