A Novel SINS/DVL Integrated Navigation Method Based on Different Track Models for Complex Environment

In the underwater navigation system, the Strap-down Inertial navigation (SINS)/Doppler Velocity Log (DVL) system has the advantage of autonomy. However, affected by the measurement range of DVL, there are two operating modes: bottom-track and water-track. To solve the navigation continuity problem of SINS/DVL under different modes, an SINS/DVL navigation method aided ocean current is proposed in this paper. On the one hand, according to different work modes of DVL, state space models based on bottom-track and water-track are established respectively. In order to improve the model match, DVL scale factor error and ocean current information are considered separately in this paper. On the other hand, in order to reduce the influence of complex noises, an improved Variational Bayes Adaptive Kalman Filter (VBAKF) algorithm is proposed. Based on VBAKF, a noise estimator is introduced, and an adaptive adjustment factor is designed. The improved VBAKF can estimate measurement noise and system noise simultaneously. The results of simulation, vehicle and river tests show that the proposed SINS/DVL method is effective and feasible. At the same time, the improved VBAKF algorithm can reduce the external noise interference and improve the navigation accuracy.