The UKF Train Combination Positioning Information Fusion Algorithm Based on Variational Bayesian Estimation

In order to solve the problem of inaccurate positioning caused by satellite signal interference during train operation, a combination positioning model of BDS/INS was established, and the system state equation and observation equation were established respectively. This paper presents a method of untracked Kalman filter (UKF) information fusion train location based on variable decibels bayesian estimation. Firstly, the nonlinear variable bayesian estimation with inverse Gamma distribution is extended to UKF algorithm framework to form a variable Bayesian UKF. Then it is applied to the BDS/INS train positioning system. The simulation results show that compared with the traditional positioning methods, the proposed algorithm can significantly reduce the positioning errors in complex environments, and fully meet the basic requirements of modern high-speed railway for train positioning accuracy and reliability. Therefore, the algorithm not only provides a new solution for train positioning technology, but also lays a solid foundation for ensuring the safety of train operation and improving the transportation efficiency, and has broad application prospects and important practical significance.