Adaptive Kalman Filter with Restriction for High Precise Vehicle-Borne Navigation

In recent years, the industry of vehicle-borne navigation develops rapidly. Navigation products appear constantly. The performance of products has significantly increased, such as accuracy, route guidance, sound and display. But they can only achieve path recognition. They are impossible to identify the driveway. Identifying the driveway will be a challenge for high precision navigation equipment. For vehicle-borne GPS data processing, the thesis designs the adaptive Kalman filter based on the current statistical model of vehicle. The average acceleration and square error can be adaptively updated. Then, based on the characteristic of vehicle driving, the adaptive Kalman filters restricted by road information as brought forward. By many simulations and real data, the Kalman filter is tested. The results prove that the algorithm is useful and fit for the system, and the filter track is smoother than the original one. The positioning precision and reliability are improved effectively.