UMI: Handheld Grippers Turn Human Demos into Hardware-Agnostic Robot Policies
gabriberton · x · 2026-11-03
Researchers from Stanford, Columbia and Toyota Research Institute released Universal Manipulation Interface (UMI), an RSS 2024 Best Systems Paper finalist.
- Handheld grippers enable portable, low-cost, information-rich collection of in-the-wild human demonstrations, including bimanual and dynamic manipulation.
- A carefully designed policy interface with inference-time latency matching and relative-trajectory action representation lets policies transfer directly from human demos to robots.
- Learned policies are hardware-agnostic and deployable across multiple robot platforms.
- By only changing training data, UMI unlocks zero-shot generalizable dynamic, bimanual, precise, and long-horizon behaviors, validated in comprehensive real-world experiments.
- Paper, code, hardware guide, data collection instructions and 3D-printing/assembly tutorials are all open-sourced.
Related event: UMI: The Data-Collection Foundation Behind Many Robot Startups(6 posts)→
More from Embodied
- Apps still not optimized for iPhone Duo's dual screen weeks after launch — evielync · 2026-11-03
- UMI: the data collection paper many robotics startups are built on — gabriberton · 2026-11-03
- New startup Shiqu Tech unveils android robot head "Unit 01" — TinfoilTricorn · 2026-11-03
- Demo app for 'vibe designing' robot clothing simulates cloth physics, thermodynamics and gaits — kscottz · 2026-11-03
- Yacine MTB says baby nuked his time but a Sunday off yielded 'ridiculous progress' on his drone hobby project — yacineMTB · 2026-11-03
- Would you fight a robot? Turing Award winner David Patterson joins the joke — davidpattersonx · 2026-11-03