Degree Study On High Accurate Strapdown Algorithm For Moving Vehicles

The strap-down inertial navigation system (SINS) represents the future developing direction of the inertial technology. Strap-down attitude algorithm plays an important role in design of SINS. Because inertial sensors are bounded to the moving vehicles, arbitrary movements of vehicles may cause errors. Therefore the optimization of strap-down algorithm is very important to eliminate the errors due to arbitrary movements. In this thesis, a systematic analysis of the reasons that causes errors existing in SINS has given, and the strategies to reduce the error caused by attitude algorithm has been discussed. The research is important for implementation of SINS.Firstly, the fundamental theory of the SINS has been introduced,including the principle and error models of SINS, the attitude algorithms, and the algorithms for navigation information updating.Secondly, the errors caused by movements of the vehicles, including coning error, sculling error, and scrolling error, have been analyzed in detail. Compensation methods for these errors are discussed. An error criterion is derived to evaluate the coning error. Optimal algorithms based on rotation vectors are presented to minimize coning errors under typical coning motions. A novel algorithm based on Miller'algorithm is presented as well.Finally, the error criterion for sculling compensation algorithms is derived. According to the criterion, the optimal compensation algorithms based on simulation results under typical sculling motion have discussed.In this thesis, analysis and compensation of errors in SINS has been discussed systematically. The simulation results of the error models and algorithms shows that present optimal algorithms have improved the accuracy of the SINS. The results of the thesis are useful for further study or design of SINS.