Carmelo Sferrazza

Humanoid locomotion learned from scratch and transferred to hardware in 10x less time, thanks to a novel off-policy flow RL approach.

TactX enables zero-shot transfer of tactile manipulation policies across different sensor types, boosting success rates from 27.5% to 45.9% in complex tasks.

Synthesizing 48,000 interaction trajectories without human input enables a humanoid robot to learn complex loco-manipulation tasks effectively.

Large-scale human motion data can not only train robot controllers but also optimize the physical designs of robot hands, achieving superior performance in real-world applications.

Humanoid robots can now perform vision-based parkour, chaining together dynamic skills like climbing, vaulting, and rolling, adapting to real-time obstacle changes.