Anticipative Iterative Learning Control of Robot Manipulators

Authors:	Polushin Ilia, Carleton University, Canada Tayebi Abdelhamid, Lakehead University, Canada
Topic:	4.3 Robotics
Session:	Robot Manipulators Control
Keywords:	Robot control, Iterative improvement, Learning algorithms

Abstract

Abstract: An iterative learning scheme for the tracking control of robot manipulators without velocity measurement is presented. The proposed learning algorithmis anticipative (noncausal) in the sense that it utilizes “future” values of thetracking error obtained during the previous iteration. The algorithm ensures convergence of the tracking error to a prescribed small domainin finite number of iterations, uniformly in time. Experimental results on a six-degrees-of-freedom (6-DOF) robot manipulator are presented to show theeffectiveness of the algorithm proposed.