Science Robotics study: motion tracking model trained on 100M+ frames enables natural humanoid whole-body movement
zhengyiluo · x · 2026-09-09
A new Science Robotics study demonstrates that a super-scaled motion tracking model trained on over 100 million frames of data enables natural, whole-body movements in humanoid robots. By treating motion tracking as a scalable learning problem with massive motion-capture data, the work marks a notable step toward general whole-body control for humanoids.
More from Embodied
- Robot arms opened to the internet: the best and worst of remote teleoperation — Lumpy_Week7304 · 2026-09-09
- Robotics class hands out 12 pre-built SO-101 arms so students jump straight to calibration — YuXiang_IRVL · 2026-09-09
- IFA Berlin Has Record Number of Humanoid Robots as Consumer Tech Giants Embrace Them — VraserX · 2026-09-09
- Synaptrix demos non-invasive brain-computer interface Atlas: full cursor control from EEG, no implants — burny_tech · 2026-09-09
- Video Prompts as Robot Programs: S1 Follows 4 Recipes in One Kitchen — deepakpathak · 2026-09-09
- Musk endorses 350-mile FSD challenge drive to Starbase ahead of Starship Flight 14 — elonmusk · 2026-09-09