The Improvement And Application Research Of Particle Filtering Algorithm In GPS/DR Integrated Vehicle Navigation System

In order to improve transport efficiency and enhance security, all countries have carried out on the research of vehicle navigation system. The key to vehicle navigation system is to choose what kind of positioning method. GPS and DR is the most common two positioning methods, They each have advantages and disadvantages, effective combination of the two kinds of positioning methods can get better locate results than the only one system. The core issue to achieve GPS/DR combination is the design of data fusion; therefore, a good use of data fusion methods to improve the accuracy of data integration became the focus of research. The main works are as follows:First of all, the vehicle navigation system and the development of filtering theory are summarized. Meanwhile, the GPS/DR integrated vehicle navigation principle is introduced.Secondly, generate filtering estimation for GPS/DR integrated vehicle navigation system based on EKF, but the application of EKF, the nonlinear system must be linear-based, leading to large errors for filtering results.Thirdly, based on the EKF, research the application of PF in GPS/DR integrated vehicle navigation system. A common problem with PF is the degeneracy phenomenon, in order to solve this problem, we rely on two methods:Good choice of importance density and Use of resampling. In this thesis, based on these two aspects, generate the importance density using EKF with full use of the observation information, so that it is closer to the true state. Also, in the systematic resampling method, it is easy to introduce errors, thereby; limit the variable error within a certain range based on the concept of norm, and set a threshold to measure the degeneracy of the algorithm, so that it is more reasonable for the choice of particle with power value. Finally, based on these two aspects, propose an improved PF. Then, the matlab tool is used and the estimation value is compared with true value from the eight filter results from eastern position and velocity based on the GPS/DR integrated vehicle navigation system. Analyze the advantage and disadvantage of EKF algorithm, PF algorithm and improved PF algorithm.At last, the thesis is summarized and the open problems for further research are also discussed.