A proposal of a new local improvement mechanism for genetic algorithm and a creative thinking support
Y. Miyata, Takeshi Furuhashi, Y. Uchikawa · 2002
The genetic algorithm (GA) is one of the basic models of evolution and is one of the effective tools for constructing evolvable/adaptive complex systems. This paper presents a new local improvement mechanism for the GA which is very efficient in local improvement. This paper describes a creative thinking support using the proposed method. The evolution of the arm morphology and the throwing motion for throwing a ball under limited link torques is simulated. 1. Introduction The genetic algorithm (GA)[1] is one of the basic models of evolution and is one of the effective tools for constructing evolvable/adaptive complex systems. The authors have proposed a new approach to GA which is efficient in local improvement of chromosomes [2][3]. This new algorithm divides each chromosome into some portions, then applies the genetic operation to each portion such as reproduction, mutation and selection. When the chromosome consists of many loci, this method is very efficient in searching the solu...