Research Of Improved CPF Algorithm For Intergrated Train Positioning

At present,scientists put forward GNSS / INS new positioning technology by combining INS navigation navigation and GNSS satellite navigation.This kind of positioning technology combines the advantages of INS and GNSS two positioning technologies,and at the same time makes up the two positioning techniques Can improve the reliability of the navigation system,but also improve the accuracy of the navigation system.It is a kind of new positioning technology with broad development prospect,which has important application in the development of navigation facilities.In the development of GNSS / INS positioning technology,the emphasis is on the innovation of information fusion technology,the research of new filtering technology for effective location information extraction,while avoiding other factors of interference.Therefore,based on the background of CTCS3 train control system,this paper studies the combination of GNSS / INS train and analyzes the effectiveness of the filtering structure and algorithm used in combinatorial positioning technology,and analyzes the effects of particle filter,Kalman filter and so on.Filtering algorithm and the existing problems of detailed analysis of the existing problems.Aiming at the problem that the filter technology can not meet the complex high-speed train combination location environment,this paper proposes a train fusion information fusion technology based on improved CPF algorithm.The improved CPF algorithm is introduced in the particle filter frame,and the Markov chain Monte Carlo(MCMC)method is introduced into the resampling process through the generation of the density function.Through this improvement,the particle degradation problem is solved and the filtering performance is improved The The improved algorithm is simulated by MATLAB.The results show that the improved CPF has smaller position error and velocity error,which can improve the positioning accuracy of the train during nonlinear motion.