We introduce perioperation, a paradigm for robotic data collection that sensorizes and records human manipulation while maximizing the transferability of the data to real robots. We implement this paradigm in DEXOP, a passive hand exoskeleton designed to maximize human ability to collect rich sensory (vision + tactile) data for diverse dexterous manipulation tasks in natural environments. DEXOP mechanically connects human fingers to robot fingers, providing users with direct contact feedback (via proprioception) and mirrors the human hand pose to the passive robot hand during manipulation. The force feedback and pose mirroring make task performance more natural for humans compared to teleoperation, increasing both speed and accuracy. We evaluate DEXOP across a range of dexterous, contact-rich tasks, demonstrating its ability to collect high-quality demonstration data at scale. Learning experiments demonstrate that data collected with \dexop leads to significantly higher task performance per unit time compared to traditional teleoperation, making DEXOP a powerful tool for advancing robot dexterity.
We show some examples of precise dexterous manipulation that mainly involves fingertip of the hand. All videos are played with 1X speed.
We show some examples of whole-hand dexterous manipulation that involves both the fingers and the palm. All videos are played with 1X speed.