A Vehicle Information Management System Based On Multi-Agent

With the constant improvement of people's life standard and the rapid development of automobile industry, vehicles enter into the human life deeply. Also, the people's requirements of motor vehicles increase highly. The vehicle's safety, comfort, simplicity of operation, and its functional completeness and the quality of after-sale service should all be taken into consideration. On the other hand, with the rapidly increasing of the automotive electronics industry, related products such as car navigation systems, multimedia systems also influence the market. However, current research on the automobile after-sales services fail to address the issues such as systematic after-sale service management system, automated diagnostic systems, maintenance systems, accident alarm systems and etc. In this paper, we present a vehicle information management system based on multi-agent, which is mainly designed for the automotive after-sales service management. Constructed on the GPRS, the system not only has its basic information management functions, but also has its automatic diagnostics, auto maintenance reminding, accident alarming and other functions.From the perspective of software engineering, the framework of the vehicle information system can be is divided into three parts:hardware, communications network and software. In the first part, we introduce the system architecture and its workflow. Then, we also describe the mechanism of communication between the system agents and the design of the communication process, which include the functional design of modules and transport protocol. The functional requirements of software components and modules are also discussed. Finally, we specify in details how to use multi-agent technology to manage vehicle information systems. From the perspective of multi-agent system model, the system is divided into car-side agent, server-side agent and client-side agent. Each agent contains a number of sub-agent, and they have different functions but are closely related to each other. Through the communication and interaction between agents, system goals will be achieved in a more efficient, automated, and systematic manner.The system has been successfully developed in an automobile electronic company, and unanimously affirmed by the users. I believe that after further improvement it will be used widely.