Vision-force fusion robot dressing handles moving arms: 85% arm coverage across 264 real trials
stepjamUK · x · 2026-09-10
A paper under review tackles a gap in robot-assisted dressing research: most methods assume a static arm, which real users can't maintain. The system fuses visual and force feedback into a single policy, trained in simulation and fine-tuned with real-world data, adapting to natural, uncooperative arm movement and handling long-sleeved garments with added friction and occlusion.
Tested across 12 participants and 264 real dressing trials, it successfully dressed an average of 85% of each person's arm length despite the arm moving throughout.
More from Embodied
- Spiced self-play accepted at CoRL: just 30 minutes of human data biases agents to right conventions — EugeneVinitsky · 2026-09-10
- Tesla FSD Tested in Heavy Rain in Slovenia, Impresses Local Owners — yunta_tsai · 2026-09-10
- Cyborg Rescue Roaches Could Deliver Lifesaving Injections in Disasters — ChuckDBrooks · 2026-09-10
- Maven Robotics Exits Stealth With $100M Series A to Win Deployment Deals — TechCrunch AI · 2026-09-10
- MicroSLAM tops LaMaria benchmark, turning cheap egocentric video into robot training environments — lukas_m_ziegler · 2026-09-10
- Musk endorses owner who 'retired from driving' after buying a Tesla FSD — elonmusk · 2026-09-10