Abstract:

This paper proposes a method of initial alignment of shipborne Strapdown Inertial Navigation System (SINS) based Adaptive Network based Fuzzy Inference Systems (ANFIS). It provides position, velocity and heading angle corrections for Inertial Navigation System (INS) and improves the accuracy and robustness of the whole system. To achieve this objective, first, wavelet is used to denoise the output of the sensors; then the position, velocity and heading angle of INS are taken as the input of ANFIS; after training the corresponding corrections are output. The expected outputs of trained samples are the results of wave multiresolution analysis. Test results show that using ANFIS the error of horizontal alignment is within 1′ and that of azimuth alignment is within 2′, which are superior to the results of Kalman and RBF under the same conditions. The proposed method can provide reliable and effective alignment for shipborne SINS.

Key words: computer application, stapdown inertial navigation system(SINS), initial alignment, wavelet multiresolution analysis(WMRA)