Research Of Kalman Filtering Algorithm In Vehicle Navigation System

Along with the development of economy and the progress of technique, electronic technology, communication technology, computer technology and artificial intelligence are playing a more and more important role in solving traffic problems. In order to improve the flexibility and safety of the traffic and maximize the use of existed road system, modern information technology is adopted to control and manage the road network and urban traffic.As a core part of Intelligent Transportation Systems (ITS), Vehicle Navigation System (VNS) is considered to be a potential measure to alleviate traffic problems. However, there are still some problems in VNS, such as low precision and low reliability of vehicle location, slow speed and low veracity of vehicle target filter tracking. Some Kalman filtering algorithms are discussed in this paper. The main contents are as follows:(1) This paper generally introduces the content of Intelligent Transportation Systems, the development and implementation of Vehicle Navigation System firstly. Then, Vehicle Navigation System is analyzed detailedly based on the study of its constitution and principle.(2) On the basis of the presentation of basic Kalman filtering algorithm, Federated Kalman Filtering (FKF) algorithm is introduced in detail, and its optimization analysis is also analyzed. A model of FKF algorithm is proposed and applied to vehicle location and navigation. Then, GPS/DR navigation model is established and simulated by applying the FKF algorithm. Simulation results show that this method can improve the accuracy of vehicle location and navigation effectively.(3) The Kalman filter tracking algorithm in Vehicle Navigation System is studied. Vehicle target model and target tracking filter are constructed based on Kalman filtering algorithm. VD-Kalman and IMM-Kalman filtering algorithm are proposed, then, Matlab simulation is made based on the two filtering algorithms respectively for the Vehicle Navigation System. The simulation results of these two methods referred above can improve filtering accuracy and reliability effectively, fault-tolerance and accuracy are also raised greatly at the same time.