On finding the optimal genetic algorithms for robot control problems

Abstract

Describes a C++ package used to analyze a class of genetic algorithms. The parameters of the best genetic algorithms have been searched by a genetic algorithm. The ultimate goal of the work is to find out if it would be possible to utilize genetic algorithm techniques in certain difficult and complex robot control problems, such as task planning, adaptation, error detection, and recovery to create a flexible robot control system. A traveling salesman type problem is used as a test and an example of applications of genetic algorithms. The sequence coding problem is solved by using link sets, which allows a pure genetic algorithm approach. This preserves a clear separation between the problem solved and the genetic algorithm itself.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>