A modified rapid alignment method of SINS based on measurement augmentation

Abstract

The precision and convergence time of initial alignment are two most important indicators for inertial navigation system, but they are usually contradictory. It is necessary to achieve considerable accuracy in a very short period for the launching of weapons such as artillery, torpedo and so on. However, the conventional methods need several minutes to complete the initial alignment. In this case, a Kalman filter method based on measurement augmentation (MAKF) is proposed to speed up convergence, and theoretical analysis of observability verifies the feasibility of the proposed method. Furthermore, a large misalignment angle model and a dimensionality reduction Cubature Kalman Filter using measurement augmentation (MACKF) are also employed to solve the invalidity of the linear model under the large misalignment angle. Finally, simulations and turntable experiment results show that the proposed method can speed up the convergence substantially. Even with initial misalignment angle of 70°, the azimuth error angle can be reduced within 1° in 5 s and 0.5° in 10 s. It has important engineering application value.