Enhanced Whole-Arm Robot Teleoperation Using a Semi-Autonomous Control Policy

Abstract

Previous work in robot teleoperation focused on the movement of a robot's end-effector by a human operator. However, a lack of pose control in teleoperation resulted in the robot arm frequently colliding with obstacles. Furthermore, even with pose control, it is still difficult for the robot to quickly and accurately move to the target due to mechanical discrepancies between human and robot. This paper proposes a semi-autonomous method to teleoperate the robot arm by integrating whole-arm teleoperation in joint-space control and autonomous end-effector position control. The proposed method is validated through experimental work on a robot arm with 6 degrees of freedom, with results showing significant improvement in human control for reaching for objects safely, quickly, and accurately.