Swarm Rule-base Development using Genetic Programming Techniques
Michael A. Kovacina, Daniel W. Palmer, Ravi Vaidyanathan · 2003
Abstract — The programming of a swarm of autonomous agents to perform a given task is a time-consuming process. This work proposes the development of a genetic programming system to self-evolve rule-bases for swarms of autonomous agents. The genetic programming system utilizes finite state machines to represent candidate algorithms. The currently implemented system has been used to evolve rule-bases for the problem of object collection, a standard problem in the area group coordination. Index Terms—genetic programming, swarm I.