Evolution of a multi-agent system with distributed multi-level memory

Conceptually, one of the important applications of evolutionary algorithms is in the field of multiagent systems. But the agents can not always act as the members of an evolutionary algorithm generation and elaborated solutions for mapping these different concepts are needed. In this paper a memory-like structure is proposed for the agents to increase the efficiency of an evolving multiagent system. Then the idea is applied to a group of mobile robots for object displacement where simulation results show the efficiency of the proposed idea.