R&D Of Dynamic Traffic Information Acquisition And Processing Technology

With the improvement of people's lives and the urbanization process, vehicle population is rising rapidly. Traffic congestion continues to worsen, which gives rise to series social problems such as traffic accidents, environmental pollution and energy consumption.Intelligent Transportation System (ITS) ease the modern transportation contradiction between supply and demand as an important way, Dynamic Traffic Information Platform (DTIP) is the core content in ITS research. For the traffic information from various subsystems, dynamic traffic information platform adopts advanced technologies of data storage, data fusion and comprehensive management, which will realize data utilizing, information sharing and providing for traffic participants, and auxiliary decision-make support for traffic managers.On the basis of the study of the key technologies of ITS, the thesis designs the DTIP around the framework of information platform, the organization and management of road traffic information and the realization of information release platform. Dynamic traffic information acquisition and processing technologies are deeply explored. DTIP is developed to realize traffic monitoring, flux control and high-quality information service. The main contents of the thesis are taken as follows:Firstly, design the platform's architecture. Describing the fuctions of each subsystem, the thesis establishes the logical framework of DTIP composed by information acquisition and processing, information release and database. Then, the thesis analyzes the customer-focused, the information needs of various users and the primary function of DTIP.Secondly, in the implementation of traffic dynamic data acquisition, the thesis introduces the system's classification, methods, advantages and disadvantages, and achieves a heterogeneous acquisition of features.Collecting real-time data of some various types of testing equipments and floating cars, processing data and uniforming data format, and then saving into SQL Server 2000 database,the system meets the city's actual needs of ITS.Thirdly, propose traffic information fusion algorithm of radial basis function (RBF) network based on an artificial immune system and fuzzy clustering. Today multi-sensor detection becomes the main directions of traffic information acquisition and real-time acquisition increases, the thesis designs the traffic information fusion algorithm mixed with artificial immune system, fuzzy clustering and artificial neural network. With the clonal selection theory in the artificial immune systems, a new algorithm with FCM clustering algorithm, RBF structural parameters are optimized. This algorithm can overcome the shortcomings of getting into local optimization from traditional algorithms, increase the convergent rate, and obtain better parallel processing ability.Fourthly, designing the internal information publishing platform, the thesis chooses MapX and Visual C++ to develop the programme. By integrating mass of static and dynamic data, the internal information publishing platform is able to realize information sharing and comprehensive data utilizing, offer kinds of information services to traffic participants and strengthen the management of traffic control.Finally, we give the results of coding, make a conclusion and propose the future research in this direction.