Learning of robot tasks on the basis of passivity and impedance concepts

Abstract

This paper aims at explaining why a simple iterative learning scheme for complicated robot dynamics with strong nonlinearities works well in acquiring any prescribed desired motion over a finite time interval or any desired periodic motion. It is firstly shown that passivity or dissipativity as an input–output property of a given system plays a key role in the capability of learning iteratively a desired motion through repeated practices. It is then shown in the simplest case when the tool endpoint is free to move that a simple iterative scheme of learning enables robots to make a progressive advance in a sense of zero-impedance matching at every trial of operation. In case of impedance control when a soft and deformable finger-tip presses a rigid object or environment, it is shown that, for a given desired periodic force of physical interaction between the soft finger tip and the rigid object, the robot learns steadily the desired periodic tasks.