Study And Simulation Of Strapdown Inertial Navigation System

Strapdown inertial navigation system (SINS) is a type of inertial navigation systems. Compared with platform inertial navigation system, strapdown inertial navigation system has the advantages of low volume, low weight and low cost. Strapdown inertial navigation system leads the trend of inertial navigation systems.In this paper, the principles of SINS navigation are introduced, and the method of initial alignment of strapdown inertial navigation system, the method of calculation of navigation, method of generating flight orbit data are discussed in detail. Then the paper develops the whole system with MATLAB/Simulink. Because the system error of SINS accumulates with time, pure SINS can hardly meet the requirements of longtime, distant navigation. To solve this problem, GPS/SINS integrated navigation system is commonly used to improve the accuracy and reliability. Simulation of GPS/SINS integrated navigation system based on Kalman filter algorithm is studied in this paper. The result shows that the simulating system is steady and reliable and the simulating result is satisfying.The initial alignment is one of the most important key techniques of the SINS, which impacts directly on the precision of the SINS. Because of the low precision of gyros in low-accuracy SINS, the accuracy of estimate of azimuth angle misalignment is not very high; and the Kalman filter is commonly used in initial alignment, which shows low robustness in practical application, thus an approach which can improve the accuracy and robustness for initial alignment of low-accuracy SINS is urgently needed. Through the introduction of magnetic heading, a new method of initial alignment has been developed for low-accuracy SINS by using H_∞filter, which can solve the problem. An error model of SINS with magnetic heading is proposed. The observability and controllability of the model is analyzed, and concluded that the system is observable and controllable. The H_∞filter is designed, and the simulation results illustrate that this method can restrain colored noises and modeling error effectively. This approach can be used in initial alignment of low-accuracy SINS.