A data fusion algorithm for DR/GPS/Map-Match integrated navigation system

In this paper, a data fusion algorithm for DR/GPS/map-match (MM) is proposed and implemented. DR is used as the primary navigation system while GPS and MM are used as aiding systems. The algorithm consists of 3 sub-algorithms, which are DR-GPS data fusion algorithm, map-match algorithm, and DR parameter correction algorithm. The data fusion algorithm first fuses the DR output and GPS which yields the input of the MM system. The MM algorithm chooses the proper road and project the vehicle position onto the road, which is based on a fuzzy road choosing theory that takes the distance and azimuth difference between the vehicle and alternative road into consideration. A similarity flag which represents the reliability of the MM data is proposed to enhance the robustness of the data fusion between DR and MM. The MM data is used not only to update positioning information of the DR but also to correct the parameters such as gyroscope scale factor and odometer scale factor, which greatly helps prevent the DR error from going wild.