Closed‐Loop Soft Robot Control Frameworks with Coordinated Policies Based on Reinforcement Learning and Proprioceptive Self‐Sensing (Adv. Funct. Mater. 51/2023)

Abstract

Soft Robot Control Frameworks In article number 2304642, Jongho Lee and co-workers report bioinspired design and control frameworks for high degrees of freedom soft robots. They mimic the biological control and hardware system by applying simulation-based reinforcement learning to find control policies for multichannel soft robots with embedded proprioceptive actuators (shape memory alloy wires). The new approach is successfully tested in soft robotic flatworms, demonstrating both forward and backward locomotion, and has shown the potential to be applied to diverse soft robot configurations.