可穿戴通用操作接口，让机器人学精细抓握更直观高效
RealDexUMI: A Wearable Universal Manipulation Interface for Dexterous Robot Learning
- 用可穿戴手套实时映射人手动作，免重定向控制
- 在8个真实任务中平均成功率88.75%，支持跨形态迁移
- 集成视觉与触觉传感，采集数据与部署零间隙匹配
学习精细操作需要保留微小手物交互的示范数据，且在部署时仍可执行。现有方法要么通过重定向或体感转换损失可执行的灵巧性，要么依赖机器人特定遥操作，成本高且难以直观获取接触感知数据。我们提出 RealDexUMI，一种基于共享灵巧末端模块的可穿戴通用操作接口，集成轻量灵巧手、手内视觉和指尖触觉传感。掌侧同构遥操作手套将人手指输入映射为机器人关节指令，实现无重定向、实时、直观且精确的手部控制。共享的手部与传感模块实现采集与部署间零间隙末端数据匹配，包括手内观测、触觉信号、接触状态和手部动作一致。在8个涵盖精细、接触密集、长时程及双臂操作的真实机器人任务中，基于 RealDexUMI 数据训练的策略平均成功率达 88.75%，可泛化至未见初始姿态，并在三种不同机器人形态间迁移。
原文摘要 · Abstract (English)
Learning dexterous manipulation requires demonstrations that preserve fine hand-object interactions while remaining executable at deployment. Existing pipelines either lose deployable dexterity through retargeting or embodiment conversion, or rely on robot-specific teleoperation that is costly to scale and often lacks intuitive, contact-aware control for dexterous data collection. We present RealDexUMI, a wearable universal manipulation interface built around a shared dexterous end-effector module that integrates a lightweight dexterous hand, in-hand vision, and fingertip tactile sensing. A palm-side isomorphic teleoperation glove maps human finger inputs to robot-hand joint commands, enabling real-time, retargeting-free, intuitive, and precise hand control. The shared hand and sensing modules yield zero-gap end-effector data, with matched in-hand observations, tactile signals, contacts, and hand actions between collection and deployment. Across eight real-robot tasks spanning fine-grained, contact-rich, long-horizon, and bimanual manipulation, policies trained on RealDexUMI data achieve an average success rate of 88.75%, generalize to unseen initial poses, and transfer across three embodiments. Website: https://research.beingbeyond.com/realdexumi
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。