| Modern ships are usually equipped with several sets of navigation system, every navigation system has its own characteristics, but there is no any one navigation system which can independently meet the high-tech warfare and ocean sailing. So the new navigation information management requirement is put forward. How to make multi-sensor integrated navigation system have high accuracy and performance and high fault tolerance and real-time performance which is one of the key of navigation technologies.Marine integrated navigation information management system is the core of the navigation information, centralized display centralized control and integrated processing of navigation information and to maximize the effect of each navigation information. in real time to the user departments to provide comprehensive and best of the navigation information. In order to improve the accuracy of ship navigation, real-time and reliability. The multiple sets of navigation devices used in combination with an organic unity, the advantages of using different navigation equipment to make up their own shortcomings, we need more navigation sensor system choose a fault-tolerant, adaptive and parallel processing capabilities of information fusion algorithm; need to design a real-time, space and ease of operation of small embedded database that can store large amounts of redundant data, improve system reliability; with real-time friendly man-machine interface that allows real-time display of the navigation sensor information, information fusion and navigation state.In view of the above problems, the integrated navigation system based on SINS, GPS and Loran C is researched. The adaptive federated filter structure and algorithm based on real-time information fusion principle is proposed and applicated in the integrated navigation system. Aiming at post information fusion processing, the optimal interval smoothing filter algorithm is presented based on Kalman filter, and the corresponding navigation sensor performance assessment is given. The simulation results show that, compared with the traditional Centralized Kalman filter, the adaptive the filter has better fault tolerance and higher accuracy when the local sensors are in trouble. Through the fixed interval smoothing, the whole curve stability is greately improved and the comprehensive filter precision is better than the traditional Kalman filter. So the propsed post information fusion algorithm is effective and feasible, and the simulation experiments verify the evaluation algorithm reasonable and effectiveness and embody a high engineering practical value.This paper adopts the S3C2440 of SAMSUNG company as a main control unit of the system. The software and hardware of ship integrated navigation information management are achieved based on the arm-linux operating system platforms. According to the theory of integrated navigation system and actual situation, the real-time database of Vxworks is established to store the original data, process data and the result. The application of Vxworks database to integrated navigation system is realized. Final system test basically satisfies the requirements of users'system and improves the performance of integrated navigation system. |
