A Method of Robot Behavior Evolution Based on Intelligent Composite Motion Control

Abstract

Intelligent Composite Motion Control (ICMC) is a methodology for building up robot systems in which robots realize complex, and dexterous behavior autonomously and adaptively by parameter optimization and use of empirical knowledge only if the motion control for basic element motions is given. In this article, ICMC is first reviewed, mainly for the Method of Knowledge Array, which provides a tool for realizing suboptimal motions for new situations by use of empirical knowledge. Behavior evolution based upon ICMC is proposed, i.e., it is shown how robot motions are coordinated from the most basic motions such as joint rotation, and how they evolve into complex behavior such as dexterous ball throwing.