Co-Evolving Communicating Classifier Systems for Tracking
In this paper we suggest a general approach to using the genetic algorithm (GA) to evolve complex control systems. It has been shown  that although the GA may be used to evolve simple controllers, it is not able to cope with the evolution of controllers for more complex problems. We present an architecture of co-evolving communicating classifier systems  as a general solution to this, where the only restriction is that each classifier system is responsible for one simple behaviour. Thus the ecology of subproblems evolves its own organisational structure at the same time its constituents evolve their solutions. Whether this structure ends up as a democratic soup, a hierarchy, or something in between, is determined by co-evolution rather than prescribed a priori by a creator. We use the trail following “tracker task” to compare the performance of a single classifier, responsible for the control of the whole system, evolved for this task with the performance of a co-evolved controller using our approach. The resulting interactions of the classifier systems are also examined.
KeywordsGenetic Algorithm Classifier System Fitness Landscape Tracker Task Artificial Life
Unable to display preview. Download preview PDF.
- Fogarty T C, ‘Evolving Controllers’, IEE Digest, 106, 1992.Google Scholar
- Holland J H, ‘Escaping brittleness: the possibilities of a general purpose machine learning algorithm applied to parallel rulebased systems’, Machine Learning II, Los Altos, 1986.Google Scholar
- Jefferson D, Collins R, Cooper C, Dyer M, Flowers M, Korf R, Taylor C and Wang A, ‘Evolution as Theme in Artificial Life: The Genesys/Tracker System’, Artificial Life II, Addison-Wesley, 1989.Google Scholar
- Hillis W D, ‘The Connection Machine’, MIT Press, 1985.Google Scholar
- Bull L, ‘Classifiers and the Tracker Task’, Internal Report, Bristol Polytechnic, 1992.Google Scholar
- Smith D C, ‘The Biology of Symbiosis’, Arnold Edward, 1987.Google Scholar
- Kauffman S A and Johnsen S, ‘Co-Evolution to the Edge of Chaos: Coupled Fitness Landscapes, Poised States, and Co-Evolutionary Avalanches’, Artificial Life II, Addison-Wesley, 1989.Google Scholar
- Dorigo M and Schnepf U, ‘Geneticsbased Machine Learning and Behaviour Based Robotics: A New Synthesis’, IEEE Trans. on Sys., 22, 6, 1992.Google Scholar
- Tinbergen N, ‘The Study of Instincts’, Oxford University Press, 1951.Google Scholar
- Booker L, ‘Classifier Systems that Learn Internal World Models’, Machine Learning, 3, 3, 1988.Google Scholar