The Obtain Of OD Information In City Based On Sensing Data

Currently, thorough traffic planning and traffic control are important parts for a city's normal operation. For urban transportation, OD information are regarded as crucial data. At present, method to obtain OD information is still conducting sample survey based mainly on questionnaire, which not only wastes time and energy, but also faces huge problems in organization. As the urbanization of our country accelerates, urban traffic system becomes increasingly complex, a more economic and available method for OD information is needed to be found. At the moment, method based on sensing data, including mobile phone signaling data link, GPS data of floating car, is an effective trial. On the base of researches to floating car, mobile location technology that used in obtaining traffic information, this paper do following works:1. Design Section OD matrix Acquisition algorithm and acquisition algorithm of path between OD sections according to GPS data of floating car. OD matrix algorithm, which based on data structure table after data handling and map matching, picks up OD information record of floating car in time in t according to variation of car state, and then represents these information. Compared with method that estimates OD matrix through traffic flow detection, this algorithm is more effective and more advanced in real-time property; acquisition algorithm of path between OD sections is designed to discuss expected traffic routes of residents, which tracks the paths of floating car, calculates these paths and thus plays an auxiliary part in traffic planning.2. Put up with algorithm that determines the coverage of mobile phone base station. The algorithm regards base station in planar map as n discrete points, and then uses Voronoi polygons and Delaunay triangulation principle to determine specific coverage of each point. Since the mobile phone data positioning produces relatively large errors, this paper design OD matrix obtaining method between sections, which describes quantitative relationship of mutual traffic between sections.This paper conducts experiments to demonstrate the effectiveness of above methods and then finds the following laws of residents travel: 1) Travel time of residents changes between weekdays and weekends. 2) Travel laws of residents in commercial districts make differences during weekdays, weekends and festivals. 3) Passenger flow volume of railway station increases sharply during festivals.These foundings demonstrate the effectiveness of algorithms designed in this paper and then make a solid foundation for future traffic planning.