The GPS/INS integrated navigation method based on adaptive SSR-SCKF cubature Kalman filter

There are many methods aiming at the nonlinear problem of GPS/INS integrated navigation, such as EKF and UKF, however, these methods have low positioning accuracy and instability. On the study of SCKF and nonlinear model of GPS/INS integrated navigation, aiming at the issues that the state equation of GPS/INS is nonlinear while the measured equation is linear, and the measured noise changes owing to the changing number of visible satellites or multipath. Therefore, this paper promotes the integrated method based on adaptive SSR-SCKF, which uses the spherical simple-radial cubature rule (SSRCR) to set the cubature sampling points. We also provide a linear measured update process on the basis of singular value decomposition (SVD), and it avoids choosing the cubature sampling points. Combining the moving window method, it can adjust the covariance matrix of measurement noise in real-time. The experiment results show that the proposed method has lower computational complexity, while higher estimated accuracy, numerical stability and better adaptive ability to the changing noise than SCKF in the same conditions.