Role Adaptation and Force, Impedance Learning For Physical Human-Robot Interaction

Abstract

In this paper, the upper limb exoskeleton robot and human collectively accomplish coordination manipulation in unknown environment. A robot's role adaptation law is proposed based on game theory. According to the human's intention which is characterized by the interaction force, the role adaptation law can adjust robot's role to lead or follow the task in the real time. On the other hand, when the robot interact with unstable environment to accomplish contact task, a new controller which concurrently adjust the force, impedance is proposed to compensate environmental disturbance and guarantee the interaction stability. Experiments have been proved the effectiveness and performance of the proposed coordination controller.