UMI-style manipulation data · SenseXperienceUMI 风格操作数据 · SenseXperience
Handheld gripper data for robots面向机器人的手持夹爪数据
SenseXperience UMI Unit records camera, jaw angle, IMU, and audio in sync. EmbodiFlow turns those sessions into training-ready datasets—no robot on site.SenseXperience UMI 单元同步记录相机、开合角度、IMU 与音频。EmbodiFlow 将采集会话转为可训练数据集，无需机器人在场。
- 160 ± 5 gWeight · single hand重量 · 单手
- 60° · 155 mm · 5 kgJaw · opening · force开合 · 开口 · 夹持力
- 2048×1536 · 30 HzRGB · 170° · rolling shutterRGB · 170° · 卷帘快门
- 240 HzEncoder · 1.0 ± 0.2°编码器 · 1.0 ± 0.2°
- 240 HzIMU · ±16 g · ±2000°/sIMU · ±16 g · ±2000°/s
- 2 chOmnidirectional mic全向麦克风
Why UMI-style data为什么是 UMI 风格数据
Manipulation demonstrations without a robot on site无需机器人在场的操作示范
Hardware-agnostic capture such as UMI lets operators demonstrate manipulation with a handheld gripper. For VLA models, Physical AI, and world models, this embodiment-agnostic human data is a critical training foundation.UMI 等与硬件无关的采集方式，使人直接用手持夹爪完成操作示范。对 VLA、物理 AI 与世界模型而言，这类与本体无关的人体数据是关键训练基础。
In-the-wild collection真实场景采集
Compact hardware for scalable collection in homes and workplaces—no robot cell required.结构紧凑，可在家庭与工作场所规模化采集，无需机器人工位。
Gripper-native signals原生夹爪信号
Jaw angle from a magnetic encoder is recorded with the camera and IMU streams.磁编码器测得的夹爪开合角度与相机、IMU 一并记录。
Long-tail coverage长尾覆盖
Human demonstrations cover diverse tasks, complex strategies, and unstructured environments.人类示范覆盖多样任务、复杂策略与非结构化环境。
Recorded signals采集信号
Time-synchronized streams per grasp每次抓取的时间同步数据流
Gripper camera夹爪相机
170° RGB view of the interaction zone, 2048×1536 at 30 Hz.170° RGB 对准交互区域，2048×1536 @ 30 Hz。
Jaw angle夹爪开合
Magnetic encoder at 240 Hz, accuracy 1.0 ± 0.2°.磁编码器 240 Hz，精度 1.0 ± 0.2°。
Motion运动
On-board IMU at 240 Hz: accelerometer ±16 g, gyroscope ±2000°/s.板载 IMU 240 Hz：加速度计 ±16 g，陀螺仪 ±2000°/s。
Audio音频
Two-channel omnidirectional microphone for contact and event cues.双声道全向麦克风，记录接触与事件声音。
System composition系统组成
UMI, wrist, and egocentric modulesUMI、腕部与第一视角模块
SenseXperience baseline combines four modules. Each can deploy independently or as one unified system.SenseXperience 基线由四个模块组成；各模块可独立部署，也可组成统一系统。
UMI Unit
Gripper Data Acquisition Module夹爪数据采集模块
Handheld gripper with camera, encoder, IMU, and microphone.集成相机、编码器、IMU 与麦克风的手持夹爪。
Wrist Unit
Wrist-mounted camera腕部相机
1920×1080 @ 30 Hz, 170° FoV, 62 g.1920×1080 @ 30 Hz，170° 视场角，62 g。
Ego Unit
Egocentric view第一视角
Stereo global-shutter RGB with IMU for task context.双目全局快门 RGB 与 IMU，提供任务语境。
Compute Unit
Main control主控
1 TB SSD, 2.5G Ethernet, 5× USB-C.1 TB SSD，2.5G 网口，5× Type-C。
From capture to training data从采集到训练数据
From gripper sessions to exportable datasets从夹爪采集到可导出数据集
Capture采集
SenseXperience wearable modules record synchronized video, IMU, audio, and interaction signals in real environments.SenseXperience 可穿戴模块在真实环境中同步记录视频、IMU、音频与交互信号。
Process后处理
IO-Agent-Lab algorithms perform timestamp alignment, pose estimation, and gesture recognition.IO-Agent-Lab 算法完成时间对齐、位姿估计与手势识别。
Review & QA审核与质检
EmbodiFlow automated and manual QA filters incomplete or low-quality samples before training.EmbodiFlow 自动与人工质检在训练前过滤不完整或低质量样本。
Export导出
Export LeRobot, HDF5, and MCAP datasets compatible with ROS and common robot-learning workflows.导出 LeRobot、HDF5、MCAP 格式，兼容 ROS 与主流机器人学习工作流。
FAQ
Frequently asked questions常见问题
What is UMI?什么是 UMI？
UMI (Universal Manipulation Interface) collects robot manipulation demonstrations with a handheld gripper and attached camera, so operators can gather data in real environments without a robot present.UMI（Universal Manipulation Interface）用手持夹爪与相机采集机器人操作示范，可在无机器人的真实环境中采集数据。
What does the UMI Unit record?UMI 单元记录哪些数据？
A 170° RGB camera (2048×1536 at 30 Hz), gripper jaw angle from a magnetic encoder at 240 Hz, IMU at 240 Hz, and two-channel audio. It combines with wrist and egocentric units for additional views.170° RGB 相机（2048×1536 @ 30 Hz）、240 Hz 磁编码器夹爪角度、240 Hz IMU 与双声道音频；可与腕部、第一视角单元组合。
Which export formats are supported?支持哪些导出格式？
Through EmbodiFlow: LeRobot, HDF5, and MCAP, compatible with ROS and common robot-learning workflows.经 EmbodiFlow 导出 LeRobot、HDF5、MCAP，兼容 ROS 与主流机器人学习工作流。
Can grippers and sensors be customized?夹爪和传感器可以定制吗？
Yes. SenseXperience is modular: cameras, grippers, sensors, and pipelines can be customized, with options for motion capture and tactile sensing.可以。SenseXperience 模块化设计，相机、夹爪、传感器与流程均可定制，并可加入动捕与触觉。
Configure a UMI capture setup for your tasks.按任务配置 UMI 采集方案。
Describe your tasks, robots, and data requirements—IO-AI will help configure SenseXperience.说明任务、机器人与数据需求，艾欧智能协助配置 SenseXperience。