2D-HMM/KF based alignment method for SINS under mooring condition

In order to restrain high-frequency stochastic perturbation of initial alignment under marine mooring condition, a novel 2D-HMM/KF (two dimensional hidden Markov model/Kalman filter) based initial alignment algorithm in inertial frame was proposed. High-frequency stochastic noises were pre-filtered through 2D-HMM/KF, and approximate periodic medium-frequency perturbation could be eliminated by integral computation, hence, the low-frequency gravity vector information was drew from raw data in body inertial frame and used to obtain the initial attitude matrix. Time-delay problem did not exist in the proposed filter compared with conventional digital low-pass filter. Mooring alignment experiment results show that the proposed initial alignment algorithm under mooring condition can improve azimuth accuracy by one time with better robustness.