Research On Data Simulation Of Train Integrated Positioning System

Train positioning is one of the key technologies of Train Traffic Control System and Train Integrated Positioning System using multi-sensor data fusion theory is an important development direction of train positioning. As the actual conditions, it's difficult to obtain the error and failure data and reappearance of the failure mode, which result in great constraints of development and verification of Train Integrated Positioning System. Therefore, it's of great importance to research and develop the Train Integrated Positioning Simulation System. This thesis mainly focused on the research of data-level simulation based on the Train Integrated Positioning Simulation System.First of all, mathematical models of the sensors which train positioning usually adopted are established according to the basic principles of train positioning, error models and failure characteristics are analyzed as well.Secondly, the structure of the Train Integrated Positioning Simulation System are designed followed the rules of the actual Train Integrated Positioning System. The main characteristics and major contents of the simulation system are described and analyzed from three levels which are data-level, signal-level and device-level and the major clue of the data simulation of the Train Integrated Positioning Simulation System is proposed.Relationship of speed, distance and time are obtained by using Train traction calculation and direction information of train traveled are extracted from the original actual data. With this information, the train operation characteristics and motion model are acquired. Based on the train traction calculation together with the sensor error model and failure characteristics, simulated multi-sensor data are obtained by data simulation.Finally, the multi-sensor data are obtained with the simulation function of SimGEN which is the software of the GNSS (Global Navigation Satellite System) simulator. The motion model SimGEN used is established through the train operation characteristics and the simulation time is synchronized using the time of train traveled as the coordinate. Compared the simulated data with these data, the correctness of the simulated data is validated. Also, the data of accelerator and gyroscope are used to fusion calculation with the algorithm of data fusion. The outcome represented the result of GNSS/IMU based train integrated positioning is compared with the original data and proved the simulated data are usable.