Research On Some Key Issues Of Intelligent Transportation System Based On LBS

In recent years, with the wide popularity of positioning equipments such as GPSCand the rapid development of the mobile applications, various applications based on LBS is rising rapidly, influencing and changing people s way of life. As one of most closely tied applications to LBS, application of intelligent transportation has been providing rich data to LBS as well as presenting great challenges on how to manage and utilize these data to better sever intelligent transportation system. This dissertation focused on the main three factors of transportation system which are human, vehicles, and roads Ctook ridesharing as the main line, researched the key issues of the intelligent transportation system based on LBS, proposed the specific solutions and application systems. The major work and contributions of this dissertation are as following:(1) We proposed an algorithm for mining regular routes from historical trajectory data, and recognized the travel modes based on the characteristic of the mined routes, thus help to understand and predict users travel behavior. In the dissertation, we proposed a series of trajectory characteristics and mining methods including the support route, support directed edges and the regular stay points, and verified the effectiveness of the algorithm with the historical user data and practical user experience.(2) We designed a recommendation system for ridesharing commuting with private vehicle, and proposed two key algorithms. In this system, we focused on the characteristics of commuters, proposed a complete solution contains data storage, user retrieval and path planning; proposed an incremental riders match algorithm, which guarantees the ridesharing quality as well as improves the vehicle occupancy with a method of preference equalization testing; proposed a path planning algorithms with the better selected pickup/drop-off spots between the passenger and the driver, to improve the efficiency and successful rate of the carpool with the improved vehicle detour and passenger walking routes. We verified the effectiveness of the system based on the real user data.(3) We proposed a traffic signal priority control strategy based on the weights of different vehicles. This strategy has comprehensively considered the travel demands of common social vehicles, and provides approximate priority for ridesharing vehicles. Aiming at the situation that traffic experiment is difficult to carry out in reality, we designed and realized a scaled-down traffic experimental platform. This platform based on the scaled-down 3D road scenarios, using scaled-down autonomous vehicles as the main body of traffic, to realize the demonstration and verification of the traffic activity. Based on the scaled-down traffic experimental platform, the experiment of ridesharing signal priority is carried out. The experiment results show that, with an appropriate weight, we can further improve the travel efficiency in intersections besides the effective incentive of ridesharing to reduce the traffic flow.