Research On Precise Positioning Methods Of High Speed Train Based On Multi Source Information Fusion

With the promulgation of the Outline of national Transportation Power Construction,high-speed train has gradually become the national infrastructure and popular means of transportation,which has played a huge role in promoting the development of national economy.In order to ensure the safe and efficient operation of high-speed trains,the train control system is very strict to the position information of high-speed trains.At present,the combination positioning technology and high-speed train in high-speed train positioning,has been the focus of relevant research fields at home and abroad,among which the innovation of data fusion filtering technology has important research significance and great challenge.Therefore,this paper uses EKF,UKF and other multi-source information fusion algorithms to study the precise positioning of high-speed trains under combined positioning fusion estimation,and carries out simulation and experimental research.The main work of this paper is as follows:1.Modeling and simulation of tightly coupled GNSS/INS combined positioning for high-speed trains.The divergence of positioning error is caused by satellite signal occlusion in integrated positioning system.In the background of CCTS-3 train control system,a tightly coupled GNSS/INS combined positioning model for high-speed trains was established by the state variables of GNSS and INS positioning system as well as the difference of their output pseudo-range and pseudo-range rate error,which reflected the mapping error feedback of INS positioning system,ensuring the calculation and estimation of the correction of positioning parameters.On this basis,the simulation conditions are set,and the trajectory and speed of high-speed train are simulated in the tightly coupled GNSS/INS combined positioning model.The simulation results show that the tightly coupled GNSS/INS model is suitable for high-speed train positioning.2.Research on high-speed train precise positioning method based on multisource information fusion algorithm.Aiming at the problem that global measurement information can not be obtained by using different sensors to measure data.In this paper,the multi-source information fusion algorithm can not only process multiple same or different types of data,but also calculate the optimal value of tightly coupled combination positioning.In the process of studying the multi-source information fusion algorithm,the EKF and UKF algorithms are applied to the high-speed train tightly coupled GNSS/INS combined positioning model,and the simulation analysis is carried out by MATLAB 2017 B.Aiming at the problem that the observed values in the model contain gross error and the EKF and UKF algorithms are not stable enough,the UKF algorithm is improved and simulated by using the equivalent weight principle in the anti-error estimation theory.The simulation results show that EKF,UKF and the improved UKF algorithm all play an important role in improving the positioning accuracy,among which the improved UKF algorithm can not only reduce the impact of the observation gross error,but also improve the positioning accuracy and stability most obviously.3.In order to increase the authenticity of the simulation results,the highspeed train precise positioning experimental system is built and the experimental results are analyzed.Firstly,the high-speed train positioning simulation system is designed by C# on Visual Studio 2019 platform.Through Ethernet communication and serial communication,the communication with the interface information server of ccts-3 train control simulation system is realized.Then,the train control simulation system is combined to start the safety data network,so that the interface information server can read and locate the database.Finally,the location track of EKF,UKF and improved UKF algorithm is obtained by testing the location of high-speed train through kazi special line.The experimental verification results show that the registration points of EKF,UKF and the improved UKF algorithm are stable and fit the orbit line,and the improved UKF algorithm is more accurate and closer to the orbit line.