Practical limitations of wave variable controllers in teleoperation

Abstract

Wave variable controllers can make a force reflecting teleoperator insensitive to communication time delays by encoding information before transmission. They are based on passivity and assume no information about the delays or the environment. However, like most robotic control systems, they model the master and slave devices as passive inertias and ignore amplifier and sensor dynamics. We review these modeling assumptions and show that non-idealities of the mechanisms, actuators, and sensors can violate passivity and impose practical limitations on the wave based controllers. We propose a method for quantifying this activeness and dissipating the corresponding energy to restore passivity.