Research Of The Train Positioning Technologies Based On GNSS Technology

For modernizing the management and control of China railway, and for increasing the efficiency and safety of transportation, it urgently needs to equip the train control systems which are high performance, more safe and reliable. The orientation of train control system is CBTC (Communication Based Train Control) system. CBTC can reduce the length of blocks section and increase the density of railways. Meanwhile, CBTC needs reliable and low-cost train positioning system urgently.GNSS (Global Navigation Satellite System) technologies can improve the performance of the train positioning system. For instance, GNSS can be used to correct the odometer's errors which are brought by skid and slip. Virtual balise and virtual signal can be achieved. It can reduce the cost of the train positioning system greatly by reducing trackside equipments and workload of maintenance. GNSS (Global Navigation Satellite System) develops quickly. The resources of satellites are more and more abundant. And the performance is also getting better. However, if it is used for train positioning, it still needs some assistive technologies to improve the integrity and accuracy of GNSS. These technologies include the digital track maps, integrity monitoring algorithms, the automatic recognizing of track occupancy, the integration with multi-sensors and so on.Based on the subject supported by the national nature science fund - intelligent transportation information platform for high speed railway and subject supported by the European framework 6- Early Galileo Service In China, the thesis focuses on probing deeply into Train Positioning System based on GNSS technology.The thesis describes the train positioning platform which is established in the research. Based on the field test and analysis of the requirement of train positioning, GNSS augmentation technologies were discussed and a recommendation solution for China railway was presented. A method for creating digital track map which can be used for the train control system was presented and the content of the digital track map were also defined. The thesis also presents an algorithm named multi-trail data refinement using Hough transform, which can reduce the costs of digital map production and speed up track survey. Using the historical positioning information and track information of trains, the thesis presented an algorithm of automatic track occupancy recognition which is based on HMM(Hidden Markov Model), The number of states in CHMM and the sampling rate of positioning information were analyzed, the mixed-density CHMM was established ,and the current track occupancy of the locomotives was and recognized successfully. The initialization problem of the train positioning system was resolved.The thesis presents a passive integrity monitoring algorithm of train positioning system by fault analysis, risk analysis and integrity model analysis. The algorithm adjust safety buffer according to the errors s of train positioning system, and improve the security and integrity. The train inertial feature is analyzed. With low cost inertial sensors, the GPS/INS integrated train positioning was implemented.The contents of the thesis have been tested and verified on the Qinghai-Tibet Railway and Beijing Sanjiadian marshalling yard, the rationality and feasibility have been proved.