Publishing Systems, Grid-based Traffic Information

Traffic Information System is one of the most important parts in Intelligent Transportation System. It provides real-time and reliable traffic information for other systems of ITS, which can be used as references for travelers and traffic system management. By taking advantage of the abilities of grid, which can integrate the computing resources, storage resources, information resources and knowledge resources, Traffic Information System is effectively realized in the distributed, isomeric and data-varying transportation system.This paper proposes the architecture of Integrating Traffic Information System based on the platform of grid. By making use of the Globus Toolkit, firstly, the system provides Traffic Information Issuing System grid-based, and realizes release and querying of the traffic information under the gird environment. Secondly, it realizes the hierarchical management of information and the constructs hierarchical structure of Traffic Information System. Finally, by using the safety services provided by grid, the system constructs the safe architecture of Integrating Traffic Information System, which realizes the safety services of Traffic Information System.By using release and querying services of Traffic Information Issuing System, the traffic information system can query the best computational resource and real-time traffic data, which realizes the traffic flow forecasting based on generalized neural networks. The characteristics of traffic forecasting service factory also make different users share the results of traffic forecasting at the same time and on the same road. By making use of the mutually invoking among the independent services such as data depositing, adopting services, best resource services and parameters-studying services, the concurrent execution of the forecasting services at different nodes are realized, which remarkably reduces the average waiting time of users and makes the problems of the classical model that cannot meet the needs for the distributed and Real-time characteristic resolved.