On the evolutionary optimization of chaos control - A brief survey

Abstract

This work represents the brief introduction into the issues of evolutionary optimization of discrete chaotic systems. This work introduces and compares evolutionary approach representing tuning of parameters for an existing control method either with the standard cost function using the numerical desired state as the one of the input or blackbox type cost function, as well as meta-evolutionary approach representing synthesis of a whole control law by means of Analytic Programming (AP). The main part of this work is focused on the proper development of the cost function used in evolutionary process. As an example of discrete chaotic system, one-dimensional Logistic equation was used. For the experiments following soft computing tools were utilized: Symbolic regression tool Analytic Programming and evolutionary algorithms Self-Organizing Migrating Algorithm (SOMA) and Differential Evolution (DE).