Research On Traffic Information Processing Technology Based On Fuzzy Theory

Intelligent Integrated Traffic Information System (IITIS) is an important and necessary subsystem of Intelligent Transportation Systems (ITS), which can promote traffic management level, improve urban traffic status and service level as well as realize the transportation sustainable development. Data processing is the core of IITIS, which is an essential part to get decision support knowledge for travelers and management department form basic traffic origin data.The aim of traffic information processing is to analyze historical, real-time, short-term and long time traffic data such as traffic volume, occupancy, speed, queue, delay and travel time about intersections, links and networks, and obtain useful information for traffic signal timing optimizing, traffic guidance information broadcasting, traffic network organizing and so on. If there is no traffic information processing, traffic management and control will work blindly and can't achieve the intelligent goal.The importance of traffic information processing in IITIS is reviewed firstly, fuzzy theory application in traffic information processing is analyzed, traffic volume prediction algorithm based on fuzzy linear regression model is presented, more then, fuzzy comprehensive judgment is used to analyze traffic status operation, fuzzy control is initially put forward in traffic flow equilibrium assignment, thus providing decision support for microscopic information process and macroscopic management. Finally, a fuzzy expert system for traffic management decision support is described in detail.The above four levels are to describe the traffic information processing in microscopic and macroscopic aspect. According to the point, line and area aspect, they can be divided into:The first level: traffic volume prediction and traffic status judgement from point and line layer.The second level: fuzzy equilibrium control and fuzzy expert system for traffic management decision support from area layer. In terms of local, zone tactics and area strategy, they can be divided into:The local level: to solve the problem of the local traffic volume prediction.They have something to do with short-time traffic and solve the practical task, such as predicting the intersection information (volume).The zone tactics level: how to execute the traffic management and control strategy such as broadcasting the guidance information and managing the emergency incident, this work need to judge the traffic operation status and control the traffic flow equilibrium assignment.The area strategy level: fuzzy expert system can analyse the existing network in order to get manipulative knowledge by using abundant traffic data and traffic expert knowledge.This research consists of seven chapters. The main works are described as follows:1 Traffic information processing and fuzzy theory foundation. The necessity and important meaning of traffic management digital information is discussed, traffic information concept, characteristic and information process goal is put forward, study meaning is further pointed out. Then, general traffic information preprocessing, data fusion and mining technologies and fuzzy theory knowledge are recommended. Thus providing theory and technology foundation for the later traffic information processing technology based on the fuzzy theory.2 Traffic volume prediction algorithm based on fuzzy linear regression. Accurate knowledge of future short time traffic volume on the travel network is critical for traffic information system. It is important and necessary to estimate traffic volume precisely in real–time with historic collected data for Urban Traffic Control System(UTCS) and Urban Traffic Flow Guidance System(UTFGS). The important meaning of traffic volume prediction is discussed. In this research, based on literature review of several methods to estimate future short interval traffic volume, a fuzzy linear regression algorithm is presented to predict urban traffic flow dynamically, the model is put forward and the detail principle is studied. In contrast to other technologies, the fuzzy algorithm is able to integrate statistical volume data and get the fuzzy correlation among them. The algorithm is tested with real traffic data and produces average estimates error of traffic volume with only 6.00%. Applied to Shanghai city, the algorithm exhibits high performance in traffic volume estimation and play a big role in advanced traffic information system. 3 Traffic status mining based on fuzzy comprehensive judgment. Traffic status analysis in the urban traffic network is the foundation of traffic flow guidance and traffic control and the important content of IITIS too. Fuzzy comprehensive judgment method is adopted to analyze and mining traffic operation status, 3 traffic parameters are used such as link average travel speed, average occupancy and average delay of intersection to judge out four kinds of traffic statuses, which can provide decision support for traffic department. The algorithm is tested with simulation traffic data and produces judgment correct rate of traffic status with 80% or so. The simulation result proves that the model in this paper is practical and the algorithm is feasible, which can be used to judge traffic statuses and detect traffic incident.4 Dynamic traffic equilibrium assignment based on fuzzy control. To avoid the complexity and difficulties in solving the optimal traffic assignment model, a fuzzy control algorithm is put forward, the input collected traffic data is described and how to convert them to fuzzy data is also interpreted in detail. Then the fuzzy control rules are listed in table to get the optimal link volumes. In the last, a simulation road network is set up, the quasi algorithm for the equilibrium assignment model is tested and the results are got satisfactorily, in which total travel is increased about 6.77%, total travel time is decreased about 23.06% and average travel speed is improved 38.81% after equilibrium assignment.5 Research of fuzzy expert system for traffic management decision support. Taking into the advantages of fuzzy expert system such as the uncertain, not complete or fuzzy knowledge and the intelligence of expert possessing, in this research, Fuzzy Expert System for Traffic management Decision Support (FES-TDS) is put forward, the function of FES-TDS is analyzed, then, GIS-T database based on integrated information system is designed, importantly, the knowledege base of FES-TDS on the basis of practical experience is put forward, and the mechanism of deducing of FES-TDS is described in detail , including deducing the method, deducing the control strategy and deducing the procedure.6 Concluding remarks of the research. The main research achievements and further study problems are pointed out.