Research on multi-source information fusion based on sliding window-factor graph for autonomous underwater vehicle

Abstract

Aiming at the high-precision real-time navigation requirements of autonomous underwater vehicle (AUV)-integrated navigation system, an AUV multi-source information fusion algorithm based on sliding window-factor graph (SW-FG) was proposed in this paper. First, the SW-FG model for the AUV-integrated navigation system is constructed using the factor graph (FG) theory. Second, historical factor nodes and estimated variable nodes are selected by a fixed-width window, and the information factors are updated by window sliding and updating. Finally, the forward and backward message passing of the FG are performed within the sliding window to achieve the smoothed optimal estimation of the current navigation state through the weighted combination of the two navigation solutions. Simulation results show that the proposed SW-FG algorithm can significantly enhance the real-time navigation solution accuracy of the AUV-integrated navigation system. Compared with the regular FG filtering algorithm, the horizontal positioning accuracy of the proposed SW-FG algorithm is improved by 17.17%. Semi-physical experimental results verify the reliability and effectiveness of the proposed scheme, and the horizontal positioning accuracy is improved by 7.28%.