Impedance Control of Exoskeleton Suit Based on Adaptive RBF Neural Network

Abstract

Exoskeleton suit is a typical human-machine system. Control the exoskeleton suit to track the pilot's moving trajectory as well as to minimize the human-machine interaction force. The suit will help decrease the pilot's power consumption and assist the pilot to carry heavy load. Impedance control was introduced to the control of exoskeleton suit. As the control laws that based on the dynamic model without model uncertainty compensation will increase the human-machine force, a RBF neural network with adaptive learning algorithm was used to compensate the model uncertainty. The stability analysis of the control law was given and the simulation results show the feasibility and validity of the proposed control law.