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
- CoRL paper: flow-matching VLAs can run 1-step Euler inference with no retraining — kaixhin · 2026-10-26
- Robotic dexterity should be judged on speed, force control and tool use, not copying humans — DJiafei · 2026-10-26
- Shenzhen company combines humanoid robots with industrial robot arms — RoboBalaji · 2026-10-26
- Realbotix, a sex-doll maker, unveils Echo humanoid with lifelike face — CyberRobooo · 2026-10-26
- Track bolt robot tightens railway fasteners in 6–7 seconds with sub-mm precision — RoboBalaji · 2026-10-26
- Tesla FSD (Supervised) Approved in Latvia, Its 10th European Country — XFreeze · 2026-10-26