Optimal target impedance selection of the robot interacting with human

Abstract

Human–robot interaction is an important issue in robotic researches which is the key in many rehabilitation and robot-assisted therapy applications. Impedance control can properly handle soft interaction of robots with the environment. Optimal target impedance selection can increase the performance of the overall system and guarantee the stability. The target impedance cannot be selected without proper knowledge about the stiffness and inertia parameters of the human. In this paper, a systematic analysis is done to introduce a method to estimate the human stiffness and consequently adjust the robot target stiffness. Then, particle swarm optimization is used to find the damping and inertia parameters of the robot to minimize the peak of the interaction force. Also, no assumption is made for the passivity of the human dynamic. The passivity analysis of the human–robot system is investigated. The novelty of this paper is in introducing a practical approach to select the robot target impedance. Finally, experimental results on a lower limb exoskeleton are provided to validate the proposed approach.