Learning control of elastic joint robot and its application to the industrial robot manipulator

Abstract

Learning control, a methodology for obtaining a reference input pattern that enables an actual output of a robot manipulator to coincide with a desired one, is modified and applied to a manipulator with an elastic-joint robot. This type of learning control is suitable for actual industrial robot control systems that have independent feedback controllers with each joint. Learning control has been implemented for a single-link elastic-joint robot. Improved high-speed motion has been achieved. Experiments on an industrial manipulator show that the problem of vibration encountered in high-speed motion is mitigated.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>