Research On The Technologies Of GPS/DR Integrated Positioning For Vehicle

With the development of the Economy and Society, the traffic jams and environmental pollution of the city get more and more serious. At present,the vehicles positioning based on GPS is the most advanced and popular style to satellite positioning system.To obtain accurate positioning information is the most basic requirement of the vehicle navigation system (VNS). But the GPS signal is very hard to receive and easy to lose because of the environmental condition, such as the tall building, the multi-level crossing bridge, the tunnel, the radio jamming and etc. So the GPS cannot make accurate navigation and localization. On the other hand, in open country, the woods, the cave and the heavy weather often make blackout area to the GPS. In view of the GPS insufficiency, it is necessary to introduce DR to assist and to guarantee the continuity and accuracy of localization.From the perspective of theoretical research, this paper discusses a newest branch of ITS, the Location system of Intelligent Vehicle navigation. And it makes a comprehensive and systematic discussion and summary from the aspects of theory,approach and application. This paper also analyzes the composition of GPS, location rationale, error sources, the composition of DR and its own location rationale. Finally, it studies a data fusion algorithm of GPS/DR integrated positioning.While this article mainly elaborates the integrated positioning algorithm, it also analyzes the traditional auto-adapted federated kalman filter algorithm and proposes the improvement of auto-adapted federated filter algorithm. It consists the auto-adapted federated filter algorithm, the SageHusa auto-adapted filter algorithm and so on. The superiority of this algorithm is the theoretical simplicity and high processing speed and positioning precise. The algorithm takes full advantage that the GPS positioning data in traditional algorithm is diverges to obtain restrain, and enhancing the combination positioning precise. Moreover this algorithm has also carried out on the computer simulation experiment through Matlab language. A series of localization data and simulation result confirm the feasibility and validity of this algorithm, and prove that the improved algorithm can enhance the precision and reliability of the vehicles positioning compared to the traditional algorithm.