Skip to Main Content
Agent is one of the representative subjects for research in the field of artificial intelligent. And the architecture of agent is a problem under hot discussion. The classes of agent architecture are introduced in detail and the main approaches to improve the performance of the hybrid agent is discussed in this paper, based on the analysis on the architecture of agent. A coordination controller model of agent based on fuzzy Petri nets is proposed, aim at the problem exist in the coordination controller in the hybrid agent model. And some experiments are given out to illustrate the working process of agent coordination controller based on fuzzy Petri nets.