arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.30876cs.RO
收藏

LocoManiLab:用于人形机器人运动操作中的数据、策略、控制器与部署的统一测试平台

LocoManiLab: A Unified Testbed for Data, Policies, Controllers, and Deployment in Humanoid Loco-Manipulation

Institute of Artificial Intelligence (TeleAI), China Telecom(中国电信人工智能研究院(TeleAI));University of Science and Technology of China(中国科学技术大学);The University of Hong Kong(香港大学);Zhejiang University(浙江大学);Shanghai Jiao Tong University(上海交通大学);ShanghaiTech University(上海科技大学);Gamma Robotics(伽马机器人公司)

Dewei Wang, Jinrui Han, Jiyuan Shi, Chenyun Zhang, Xianbo Lu, Jiabao Wu, Jiqin Feng, Jipeng Kong, Xinzhe Liu, Ping Luo, Yuchen Xiao, Xingxing Wang, Chenjia Bai, Xuelong Li

AI总结:

本研究提出LocoManiLab统一测试平台,覆盖数据、策略等全流程,通过234项家庭场景任务、51.5小时遥操作等实验,验证了模拟评估的可靠性,为人形机器人运动操作研究提供支撑。

摘要:

人形机器人必须在日常环境中协调运动与操作,但不同策略、控制器和任务设置下的进展难以评估。LocoManiLab提供了一个集成测试平台,涵盖数据收集与生成、策略训练、评估及物理部署。其任务套件覆盖5类家庭场景中的234项原子及组合任务,依托51.5小时的模拟遥操作支撑。评估内容包括:12种策略在共享控制器下跨任务、场景划分及扰动水平的表现,5种控制器下的2种策略,以及演示规模与数据源混合的研究。成功率在未见过的任务和场景上有所下降,且随视觉扰动增加普遍降低,不同策略在泛化性和鲁棒性上存在差异。在数据生成方面,对人形运动生成模型进行了运动操作方向的微调,以研究其输出与任务执行要求的对齐情况及在数据生成流水线中的集成。3项模拟运动操作任务与真实世界设置高度匹配,可在两个领域评估3种策略,相似的策略排名验证了基于模拟评估的可靠性。

英文摘要:

Humanoid robots must coordinate locomotion and manipulation in everyday environments, yet progress is difficult to assess across differing policies, controllers, and task settings. LocoManiLab provides an integrated testbed spanning data collection and generation, policy training, evaluation, and physical deployment. Its task suite covers 234 atomic and compositional tasks across five household scene categories, supported by 51.5 hours of simulated teleoperation. The evaluation covers twelve policies with a shared controller across task and scene splits and perturbation levels, two policies across five controllers, and studies of demonstration scale and data-source mixtures. Success rates fall on unseen tasks and scenes and generally decline as visual perturbations increase, with differences in generalization and robustness across policies. For data generation, a humanoid motion generation model is fine-tuned for loco-manipulation to study the alignment of its outputs with task execution requirements and its integration into the data generation pipeline. Three simulated loco-manipulation tasks closely match real-world setups, enabling evaluation of three policies in both domains. Similar policy rankings support the reliability of simulation-based evaluation.

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。

↑