A Multi-Mode Switching Variational Bayesian Adaptive Kalman Filter Algorithm for the SINS/PNS/GMNS Navigation System of Pelagic Ships

Abstract

The ocean-going environment is complex and changeable with great uncertainty, which poses a huge challenge to the navigation ability of ships working in the pelagic ocean. In this paper, in an attempt to deal with the complex uncertain interference that the environment may bring to the strap-down inertial navigation system/polarization navigation system/geomagnetic navigation system (SINS/PNS/GMNS) integrated navigation system, the multi-mode switching variational Bayesian adaptive Kalman filter (MMS-VBAKF) algorithm is proposed. To be more specific, to identify the degrees of measurement interference more effectively, we design an interference evaluation and multi-mode switching mechanism using the original polarization information and geomagnetic information. Through this mechanism, the interference to the SINS/PNS/GMNS navigation system is divided into three cases. In case of slight interference (case SI), the variational Bayesian method is adopted directly to solve the problem that the statistical characteristics of measurement noise are unknown. By the fixed-point iteration mechanism, the statistical properties of the measurement noise and the system states can be estimated adaptively in real time. In case of interference-tolerance (case TI), the estimation of the statistical characteristics of measurement noise need to weigh the measurement information at the moment and a priori value information comprehensively. In case of excessive interference (case EI), the SINS/PNS/GMNS integrated navigation system will perform mode switching and filtering system reconstruction in advance. Then, the information fusion and navigation states estimation can be completed. Consequently, the reliability, robustness, and accuracy of the SINS/PNS/GMNS integrated navigation system can be guaranteed. Finally, the effectiveness of the algorithm is illustrated by the simulation experiments.