Abstract
This paper examines the Argumentation Negotiation of agent as a mechanism for solving the intersection priority problem. We integrated multi-agent automatic negotiation mechanism and argumentation reasoning to apply in selecting vehicle priority problem. Agent's belief is evolved by argumentation-base negotiation, so that an agent can adopt the environment easily and select even better actions. When agents are initially created, they have little knowledge and experience with relatively low capability. They strive to adapt themselves to the changing environment. It is an advantage for agents have a good way to learn and evolve their knowledge. This paper addresses evolution of intelligent agents in transport information system. Fuzzy theory is proposed as evolution mechanisms, and Fuzzy soft goal is introduced to facilitate the evolution process. We have built an agent system to demonstrate our approach.