arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.26169cs.RO
收藏

通过同构遥操作与独立技能路由实现的人形机器人组合操作

Compositional Humanoid Manipulation through Isomorphic Teleoperation and Independent Skill Routing

Shanghai Jiao Tong University(上海交通大学);The University of Hong Kong(香港大学);Xense Robotics(森斯机器人公司);Fudan University(复旦大学)

Yiming Wang, Peixian Yang, Lifeng Zhuo, Lai Wei, Xinling Du, Kun Song, Weiming Wang, Daolin Ma

AI总结:

该研究针对人形机器人操作需复用手臂技能但演示所有组合不现实的问题,提出HS-MoDP策略,结合G1Exo外骨骼实现技能组合,提升了橱柜操作成功率,相关硬件与代码将开源。

摘要:

人形机器人操作需要可复用的手臂技能,但演示所有可能的技能组合并不现实。我们提出人形机器人语义混合专家扩散策略(Humanoid Semantic Mixture-of-Experts Diffusion Policy,HS-MoDP),该策略通过在共享专家池上进行独立语义路由来组合分别演示的手臂技能。此设计在保留手臂特定技能控制的同时,实现了手臂间动作生成参数的共享。语义调度器通过交替操作和重新定位阶段,将学习到的策略与现有的运动控制器集成,扩展了有效工作空间并实现了位置变化时的技能复用。为获取全身接触操作的演示,我们开发了G1Exo,这是一种同构外骨骼,可提供手臂构型的直接关节级控制。在Unitree G1上的实验表明,HS-MoDP的技能组合效果优于基线方法;G1Exo相比末端执行器VR-IK能实现更可靠的接触遥操作,其演示可提升四种策略架构的接触技能学习效果。对于HS-MoDP,将演示来源从VR-IK切换为G1Exo后,完整橱柜操作流程的成功率从35%提升至65%。所有项目硬件设计和源代码将在评审流程结束后以开源形式发布。

英文摘要:

Humanoid manipulation requires reusable arm skills, yet demonstrating every possible skill combination is impractical. We present Humanoid Semantic Mixture-of-Experts Diffusion Policy (HS-MoDP), which composes separately demonstrated arm skills through independent semantic routing over a shared expert pool. This design preserves arm-specific skill control while sharing action-generation parameters across arms. A semantic scheduler integrates the learned policy with an existing locomotion controller through alternating manipulation and repositioning stages, extending the effective workspace and enabling skill reuse across location changes. To acquire demonstrations of whole-body contact manipulation, we develop G1Exo, an isomorphic exoskeleton that provides direct joint-level control of arm configurations. Experiments on a Unitree G1 show stronger skill composition than the baselines. G1Exo enables more reliable contact teleoperation than end-effector VR-IK, and its demonstrations improve contact-skill learning across four policy architectures. For HS-MoDP, switching from VR-IK to G1Exo demonstrations raises full cabinet workflow success from 35% to 65%. All project hardware designs and source code will be released as open source after the review process.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑