Control system optimization using genetic algorithms within the SoftLab toolkit
Lisa M. Desjarlais, Mohammad-R. Akbarzadeh-T, Craig W. Wright · 1999
This paper presents techniques for using genetic algorithms within the SoftLab software environment to automate the optimization of control systems. SoftLab is a framework for the design and simulation of control systems, using soft computing components. It allows users to construct systems by connecting computational modules into dataflow networks. We discuss the set of modules provided for implementing genetic algorithms and demonstrate how even the problem-specific aspects of the genetic algorithm can be easily specified through interpreted scripts, avoiding the need to compile or link in order to set up a new test problem.