A variable structure multiple model filtering for SINS/DVL integrated solution
Wang Lei, Haitao Gao · 2017
This paper describes the implementation of an intelligent navigation system, based on the integrated use of the Doppler velocity log (DVL) and strapdown inertial navigation system (SINS), for autonomous underwater vehicle (AUV) applications. A variable structure multiple model (VSMM) filtering method is presented to be used to fuse the data from the SINS sensors and to integrated them with the DVL data. The goal is to solve the problem with respect to measurement noise with unknown or randomly varying statistics properties when AUV is in uncertain or tough environment. In the presented VSMM approach, expected-mode augmentation methodology is utilized to get more accurate model set for the estimating process of IMM algorithm. Compared to IMM algorithm, the novel algorithm can capture more subtle changes of the system mode. Simulation results show that the proposed algorithm can improve the precision and stability of the navigation algorithm significantly.