Variable-Length Iterative Learning Control for Live Working Manipulators

Abstract

The rapid development of computer technology and robot technology in recent years makes it possible for robotic arm to replace manual work to carry out dangerous live work, which will greatly improve the efficiency of maintaining electric power facilities, shorten maintenance time and protect workers’ safety. Aiming at the control of manipulators, an iterative learning control algorithm based on variable length and time-varying system is developed in this paper. This method can well solve the problem of different operating time of live-line operation of manipulator, and improve the limitations of original iterative control laws that each iteration time is required to keep same to adapt to the actual working conditions. Specifically, this paper is divided into two parts: Firstly, machine vision is used to locate the spatial relationships among the manipulator base, the manipulator actuator, and the target to be worked; Then, a novel iterative learning control schemes are constructed for the tracking control purpose.