Underwater Adaptive Height-Constraint Algorithm Based on SINS/LBL Tightly Coupled

Abstract

The application of strap-down inertial navigation system/long baseline system (SINS/LBL) has been proven to be an effective method to solve the accumulated position error of autonomous underwater vehicles (AUVs). However, height positioning in the SINS/LBL system is generally unstable. Acoustic differential technology uses single- or double-difference mode to weaken the influence of system errors on signal propagation time, but the vertical positioning result of this method is unstable. Moreover, the accumulative errors of inertial navigation will cause the height positioning errors to diverge over time. In view of the situation that the height positioning results are prone to divergence, the SINS/LBL tightly coupled (SLT) system is studied and an adaptive method based on the improved Sage–Husa adaptive filter is proposed. This method adaptively adjusts the system noise matrix and then utilizes depth information of the pressure sensor (PS) to adaptively estimate the vertical direction errors. To verify the feasibility of the proposed algorithm, the simulation experiments compare the positioning results of three fusion methods that are centralized Kalman filter (CKF), federated Kalman filter (FKF), and adaptive height-constraint Kalman filter (AHKF). Simulation results show that the AHKF algorithm improves the precision and stability of underwater SLT navigation system significantly. To some extent, the AHKF algorithm addresses the problem of accumulative errors in the vertical direction in the SLT system. Also, the improvement of height positioning accuracy is hugely helpful to AUV complete mission successfully.