Heterogeneous Multi Agents Learning Using Improved Genetic Network Programming

抄録

The heterogeneous multi agent system is a system that cooperates intensively in one place by two or more agents and achieves the task. There is Genetic Network Programming as an automatic generation technique of the heterogeneous multi agent system. In this paper, the authors proposes Improved Genetic Network Programming as automatic generation technique of the heterogeneous Multi Agent system. And the authors propose both co-evolution technique and non-co-evolution technique as agent's evolution technique and compare the performance of two techniques.