Sha Yi

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.

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.

Tactile sensing isn't a universal win for robot manipulation; its effectiveness hinges on the sensor type, material, and the task at hand.