Complex Systems

Research of Complexity in Cellular Automata through Evolutionary Algorithms Download PDF

Emmanuel Sapin
Faculty of Computing, Engineering, and Mathematical Sciences,
University of the West of England,
Bristol, BS16 1QY, UK

Olivier Bailleux

Jacqueline Chabrier
Universite de Bourgogne, 9 avenue A. Savary, B.P. 47870,
Dijon, 21078 Cedex, France

Abstract

This paper presents an evolutionary approach to searching for cellular automata that accept gliders. The proposed technique is based on a specific fitness function taking into account spatial evolution, the number of living cells, and the presence of gliders. The results show that the genetic algorithm is a promising tool for finding cellular automata with specific behaviors, and then could prove to be decisive for identifying new automata that support universal computation.