Research On Higher-order Algorithm Of Strap-down Inertial Navigation System

Strap-down inertial navigation system(SINS), which is widely used in many areas, is a very advanced inertial navigation technology, for its feature of small volume, simple structure,low cost, and good maintainability. To design a high performance algorithm for error compensation to improve the navigation precision of the SINS is very important. This paper makes study on building system error model and filter algorithm of SINS to improve the navigation precision of the SINS.First, this paper makes research on several kinds of attitude model establishment methods, like Euler Angle, quaternion, direction cosine matrix, and Rodrigues parameters method, and analyses the relationship between these methods. This paper proposes dual quaternion to describe attitude for SINS, and established the additive dual quaternion transfer alignment error model on the basis of the quaternion alignment error model. Contrast from the viewpoint of engineering application, this paper makes simulation research on these two models, simulation results verify the additive dual quaternion error model more superiority, provides the reference for new attitude description method and transfer alignment technology of SINS.Then, through detailed analysis for the Bayesian optimal filtering theory, this paper derives the UKF filter algorithm based on unscented transformation rule, Stirling polynomial interpolation filtering algorithm and CDKF algorithm, and discusses the estimation precision problem of these algorithms. Through analysing CKF filtering algorithm based on the spherical-radial cubature rule, this paper proposes a high-order CKF algorithm, and analyses its feasibility and computational advantage in theory, which lays a theoretical foundation for the engineering application of the higher-order CKF algorithm. Then this paper sets up the SINS/GPS integrated navigation positioning system error model for high resolution airborne SAR motion compensation problems, uses higher-order CKF algorithm to calculate and analyze, the simulation results show high order CKF algorithm with high calculation accuracy and fast convergence speed.Finally, this paper makes some optimal design for high-order CKF algorithm. Traditional SINS quaternion norm-preserving calculation needs normalized process, which leads to huge calculation and complicated control program, this paper proposes padé approximation to construct a quaternion norm-preserving recursive formula, to avoid the normalized process. Through simulation test, the simulation results verify the padé approximation algorithm with higher computing precision and faster convergence speed, reduces the amount of calculation and complexity of control program, and has a good application prospect.