Guidance Priority Adaptation in Human-Robot Shared Control

Abstract

Demanded in highly unstructured environment like surgery, rehabilitation and teleoperation, shared control architecture allows operators to retain abilities and priority to control a human-robot interaction system (HRI), in which the robot provides stiffness and precision, and the human operator decides tactical maneuvering like obstacle avoidance and emergency takeover. By estimating haptic intention of human and adaptively restraining the operation force in unexpected direction, the adaptive shared control law is desired to continuously switch the lead priority to either the human or the robot. An adaptive admittance control strategy is thus developed based on the guidance virtual fixture (VF) /active constraint to promote the operation performance in collaborative tasks. By synthesizing adaptive admittance control and virtual fixture guidance, this comes true that the human operator’s intention and predefined guidance trajectory are combined in shared control. Both theoretical analysis and simulation results validate that the proposed control strategy has potential to provide stability and flexibility in the specific tasks for the HRI system.