Genetic Network Programming and Its Application to the Multiagent System

Abstract

Recently many studies have been made on automatic design of complex systems by using the optimization technique such as Genetic Algorithms (GA), Evolution Strategy (ES), Evolutionary Programming (EP) and Genetic Programming (GP). In this paper, a new method named Genetic Network Programming (GNP) which describes the behavior sequences of agents as a network is proposed in order to acquire intelligent behavior sequences. GNP includes judgement nodes and processing nodes in the network. The proposed system can evolve itself by an evolutionary method using mutation and crossover.