Abstract: Due to the nonlinear characteristics of satellite navigation positioning equations, the traditional positioning methods need to linearize the equations before solving, which would reduce the accuracy of positioning. Cubature Kalman filter (CKF) algorithm based on Bayesian estimation and cubature transformation could effectively improve the accuracy of nonlinear filtering estimation. In this paper, cubature Kalman filter algorithm is applied to satellite positioning solution. The transfer and measurement models of the positioning solution system are established. The three-dimensional position coordinates of the receiver are recursively solved by time update and measurement update. The positioning solution algorithm based on cubature Kalman filter is verified by IGS observation data tests, and compared with the least square method and extended Kalman filter algorithm. The results show that the application of CKF algorithm in satellite positioning is of higher precision and faster convergence speed.

Key words: positioning solution, Bayesian estimation, cubature rule, CKF, LSM, EKF