A Solution of Synthesized Optimal Control Problem for Interaction of Robots by Evolutionary Computations

Abstract

A paper considers a synthesized optimal control problem. To solve this problem, it is necessary to solve the problem of control system synthesis for stabilization of control object relatively a point in the state space, and then to solve the second problem of optimal control to find some stabilization points for movement from initial position to terminal one with the best value of the quality criterion. For the first problem a network operator method was used. It finds a mathematical expression of a feedback control system in the form of oriented calculating graph by evolutionary algorithm. For solution of the second problem a particle swarm optimization was applied. In experiment the control object is a pair of robots that move a cart joined by rigid hinged hitch from initial point to terminal one trying to avoid obstacles.