Research On BDS/GPS/GLONASS Multi-System Fusion Positioning

BeiDou Navigation Satellite System(BDS)has been offering 24-hour navigation services in the Asia-Pacific region and the construction of the global coverage constellation is already in progress.In the meantime,other Global Navigation Satellite Systems(GNSS),such as Global Positioning System(GPS)and Global Navigation Satellite System(GLONASS),are also accelerating the modernization process and constantly improving their existing systems.China also signed interoperability agreements for satellite navigation systems respectively with Russia and the United States,which marks the cooperation between the world's major satellite navigation system will be deepened.Therefore,how to make better use of more and more visible satellites in the sky will become a new research focus.Traditional method of single-system positioning often encounters the problem of a significant decrease in the positioning accuracy when the number of visible satellites decreases due to occlusion.Moreover,when less than 4 satellites are visible,the 3-D positioning will not be accurate and reliable.In this thesis,the study of multi-system fusion positioning can not only solve this problem effectively,but provide a theoretical basis for the practical application in many fields in the future.The research contents of this thesis mainly include the following five aspects:1.The similarities and differences among time and coordinate systems of BDS,GPS and GLONASS were analyzed in order to unify them for fusion positioning.2.A multi-system positioning model based on pseudorange-Doppler observation was established to reduce the effect of error on positioning accuracy,which includes measure noise and process noise.Meanwhile,the rectification models of ionospheric and tropospheric errors were applied in pseudorange measurement.3.Unscented Kalman Filter(UKF),instead of Iterative Least Squares(ILS)or Extended Kalman Filter(EKF),was applied to positioning estimation algorithm.4.Ant Colony Optimization(ACO)was applied to optimizing covariance matrix as a universal method,which can enhance the stability of the filtering algorithm and further improve the positioning accuracy.5.Experiments indicate that the multi-system fusion positioning algorithm is effective both under good and extreme observation conditions and both in static and dynamic positioning.It can also obtain relatively precise positioning results under the extreme observation conditions where single-system positioning is invalid.