An Ecological Analytic Approach to Evolutionary Multi-Agent Systems

抄録

An evolutionary Multi-Agent System (MAS) is discussed as an autonomous decentralized system in which each agent is capable of adapting to its embodied environment by means of artificial evolution. This type of system is recognized as a promising approach to designing an adaptive MAS, although it is difficult to analyze the strategies that each agent achieved through the process of artificial evolution. Moreover, it is also difficult to trace how the necessitated functions are emerged through the interactions with each other. In this paper, inspired by biology, the ecological method is experimentally applied to a MAS in order to grasp how they evolved and what they achieved. A food collecting problem is examined to illustrate how to use the ecological method to grasp characteristics of agents in a MAS.