Machine-Learning Assisted Electronic Skins Capable of Proprioception and Exteroception in Soft Robotics.

Abstract

Inspired by natural biological systems, soft robots have recently been developed, showing tremendous potential in real-world applications because of their intrinsic softness and deformability. The confluence of electronic skins and machine learning is extensively studied to create effective biomimetic robotic systems. Based on a differential piezoelectric matrix, this study presents a shape-sensing electronic skin (SSES) that can recognize surface conformations with minimal interference from pressing, stretching, or other surrounding stimuli. It is then integrated with soft robots to reconstruct their shape during movement, serving as a proprioceptive sense. Additionally, the robot can utilize machine learning to identify various terrains, demonstrating exteroception and pointing toward more advanced autonomous robots capable of performing real-world tasks in challenging environments.