Higher Degree CKF And Its Application In SINS Large Misalignment Angle Initial Alignment

Initial alignment is one of the key technologies of SINS.The initial alignment accuracy directly affects the system's navigation accuracy.In accordance with the size of the initial attitude misalignment,precise alignment can be divided into a small misalignment and a large misalignment of two cases.The problem of alignment of small misalignment angle has been solved well.However,many problems of initial alignment under large misalignment angle have not been solved well.The most critical problem is the nonlinear filtering algorithm.Commonly used nonlinear filtering is based on the Kalman filter(KF)algorithm developed from the cubature Kalman filter(CKF).The ordinary CKF filtering method can achieve third degree filtering accuracy.In recent years,the high degree CKF algorithm has been proposed.Based on the cubature Kalman filter(CKF),a seventh degree cubature Kalman filter(7th-CKF)algorithm is proposed for the first time in the environment of Gaussian noise and non-Gaussian noise respectively.At the same time,The CKF algorithm is applied to the large initial misalignment of SINS.The performance of each algorithm is verified and compared.The main structure of the thesis is as follows:1?A 7th-order cubature Kalman filter(7th-CKF)algorithm is proposed.According to the integral approximation theory,the filtering order of CKF is expanded,the seventh-order spherical radius theory is deduced,and the sampling criterion of seven-order volume sampling is proposed.Finally,a 7th-CKF algorithm is proposed.2?A composite embedded Kalman filter(CECKF)algorithm is proposed.According to the embedded criterion,through the embedded approximation to the middle-order term,the composite embedded theory is deduced,and the general form of any order composite algorithm is analyzed.Finally,the CECKF algorithm is proposed.3?A seventh degree quadrature cubature Kalman filter(7th-CQKF)algorithm is proposed.According to the Gaussian Laguerre orthogonal criterion,the quadrature point is taken for the radius integral,which eliminates the uncertainty of the higher order lower radius integral approximation and reduces the complexity of higher order expansion.Finally,a 7th-CQKF algorithm is proposed.4 ? A seventh degree simplified quadrature cubature Kalman filter algorithm(7th-SCQKF)is proposed.According to the simplified CKF theory,the seventh-order simplified CKF algorithm is deduced to improve its performance.Then,the seventh-order radial integrals are orthogonally collected,and finally the 7th-SCQKF algorithm is proposed.5?The adaptive Gaussian mixed CQKF algorithm(PGM-ACQKF)is proposed for non-Gaussian noise.The characteristics of the Gaussian mixture noise model are analyzed.It is found that the displacement parameter is the key factor affecting the Gaussian mixture noise.The adaptive displacement parameter method is proposed to track the variation of Gaussian mixture noise.Then the PGM-ACQKF algorithm is proposed.6?The four algorithms of 7th-CKF,7th-CQKF,7th-SCQKF and CECKF proposed in this paper are applied to the initial alignment of SINS large misalignment angle,and the validity of each high-order algorithm in practical application is verified.The filtering accuracy and calculation of the four algorithms are analyzed.