The Research Of Map Matching Algorithm In The Vehicle Positioning And Navigation System

Along with the maturation of Geographic Information System,Satellite Positioning Technology as well as the Communication Technology day by day, the development and application of Vehicles Monitoring System based on certain positional means in Intelligence Traffic System domain is being valued by people day by day and demonstrates the enormous economy and the social efficiency. Taking the Vehicle Positioning and Navigation System as the background, numerous treatments are carried on to deal with the GPS measurement data and adopt the way of GPS/DR integration positioning in the thesis. Finally, the technology of map matching is utilized to reveal the vehicle's position in the digital map, which remedies the deficiency of positioning by the sensor only, raises the positional precision and dependability, and have practical meaning of using. The research's core of the thesis is to design rational and high-efficient map matching algorithm.Because GPS signal receives have error before the map matching, the GPS signal should be dealed with and the point with big error should be deleted and smoothed. Based on in case of stationary time series, we analyze the error data of GPS's positioning and carry on modeling to the positioning data and reveal the origin of error. About the part of GPS signal's processing, the merit and lack of interrelated filter algorithm are discussed and then state equation and observation equation of Federated Kalman Filtering based on Current Statistical Model are builded.The map matching algorithm is discussed and how to apply the algorithm to positioning and navigation system. Synthesizing relevant map matching algorithms and improving them, a kind of algorithm based on Probability Statistics Method is adopted. The match-weight value of each candidate highway section is calculated in the area of error according to vicinity principle,similarity principle and similarity principle of the crossing, and then the maximal match-weight value the best highway section is choosed.The step of algorithm is divided into initialization course, track course and re- initialization course. When the vehicle come to the crossing, corresponding treatment is carrid on to ensure that the vehicle is matched with the road correctly.Finally, MapObjects method is used under the environment of VC6.0 to realize the algorithm, to simulate actual conditions such as GPS observation signal,speed and orientation of vehicle and so on, and then to bring forth the effect picture of map matching. From the effect of matching, it is good. When vehicle comes to the crossing, we execute re- initialization course and make some especial treatment such as displaying one or several more points, which confirms the vehicle is displayed corretly.Through simulation realization of the software, the algorithm of the thesis is simple and applied. It meets the real-time character requirement, deals with complicated road conditions better and meets the requirement of vehicle positioning.