Portal de Periódicos da CAPES

Evolving the Ability of Limited Growth and Self-Repair for Artificial Embryos

2003; Springer Science+Business Media; Linguagem: Inglês

10.1007/978-3-540-39432-7_31

1611-3349

Felix Streichert, Christian Spieth, Holger Ulmer, Andreas Zell,

Gene Regulatory Network Analysis

In this paper we address the problem of limited growth and the difficulty of self-repairing in the field of Artificial Embryology. For this purpose, we developed a topological simulation environment for multiple cells which is continuous and structure-oriented, with a dynamically connected network of growing cells and endogenous communication between these cells. The cell behavior is simulated based on models of gene regulatory networks like Random Boolean Networks and S-Systems. Evolutionary Algorithms are used to evolve and optimize the parameters of the models of the gene networks. We compare the performance of Random Boolean Networks and S-Systems when optimized by Evolutionary Algorithms on the problem of limited growth and two types of cell death with and without signaling of cell death on the problem of self-repair.