Genetic network programming with automatically defined groups for assigning proper roles to multiple agents

In this paper, we apply a Genetic Network Programming (GNP) Architecture using Automatically Defined Groups (ADG) to a multi-agent problem where cooperation of agents are required. GNP is a kind of evolutionary methods inspired from Genetic Programming (GP). While GP has a tree architecture, GNP has a network architecture with which an agent works in the virtual world. In GNP with ADG, each agent is assigned to a group according to its role to complete some task of a cooperative problem. We consider two types of problems in this paper: one problem is to assign an appropriate role to each agent according to its ability, and the other is to assign a proper role to each agent with the same ability. While the first problem has the specific conditions as for the ability of an agent, the latter is a general problem. We show the effectiveness of GNP with ADG through computer simulations on the two types of load transportation problems.