FADA adapts humanoid robots to new conditions in ~2 minutes, no rewards needed
GuanyaShi · x · 2026-10-24
Researcher Angchen Xie introduces FADA: humanoid controllers often fail when conditions change—a heavy backpack, a soft floor, a steep slope. FADA adapts a humanoid to new conditions using only the robot's own experience, keeping the task goal while changing only how it executes.
Key claims: about two minutes to adapt, with no rewards, no demonstrations, and no simulator retuning. Scaling researcher rosarioscaling boosted the paper, calling the direction very promising and closely aligned with their own philosophy, needing just a few key improvements.
More from Embodied
- AGCO to pilot Agility Robotics' humanoid Digit at Kansas farm machinery plant in 2027 — CyberRobooo · 2026-10-24
- High Refractive Index Glass Hits 2.1 With 97% Transmittance, Raising Eyebrows for AR Optics — jwt0625 · 2026-10-24
- Archer to launch air taxis connecting Austin airports with F1 circuit — whurley · 2026-10-24
- microagi promises a billion robots by 2029; skeptics say scene understanding won't be solved by 2027 — AlexTensor · 2026-10-24
- Tesla registers 69 more Cybercabs, bringing its Robotaxi fleet to 420 vehicles — mitchdeg · 2026-10-24
- CMU RI PhD open-sources full robotics faculty job market application materials — sh_reya · 2026-10-24