Development Of Vehicle Integrated Navigation System Based On Information Fusion

The land vehicle navigation equipment is required that it should offer not only real-time, high-precision, high-reliability navigation information, such as position quickly, but also should have the functions of self-determining, adaptability to the change of conditions, anti-jamming, miniaturizing and good property to price ratio under the complex conditions.This paper analyses and designs the GPS/GLONASS/INS integrated navigation system used for land vehicle, which has special working condition and applied requirement. The software of prototype and correlative experiments had been accomplished. This project has important theoretical and practical significance for promoting the development of vehicle navigation equipment.This paper establishes the mathematical model in integrated navigation system firstly. Then an Adaptive Kalman filter algorithm based on ARMA model is used in the system. This method estimates filter gain matrix K directly, avoids calculating system noise covariance matrix Q and measure noise covariance matrix R , it improves stability of the system. When integrated navigation system works in real-time, there maybe arise fault information during the data acquisition due to some uncertain factors, these fault datum will.degrade the navigation accuracy of whole system. To solve this problem, A method to judge data validity based on the fuzzy logic is presented . This method gives an integrated evaluation by means of possibility theory and variable weight analysis method, then several different processing modes were given according to different judgement result. This method takes full advantages of redundancy and complementarity of different observations and makes the result more accurate and reliable. At last,on the basis of the research mentioned above, the master plan and software development of integrated navigation system has been achieved. Relevant experiments have been done to the system, including static experiments and dynamic experiments in mobile vehicle. The experiment results show that the scheme and algorithms of integrated navigation system are usable and effective.