AUV Multi-source Polar Integrated Navigation Based on Factor Graph Iterative Smoothing

A highly accurate navigation system is a prerequisite for the safe navigation of the AUV in the polar regions. During the mission, the AUV is subject to the complex ocean environment in the polar regions and the availability of its auxiliary navigation system changes with the surrounding ocean environment. With the increase in navigation sensors and the complex environment of the polar regions, conventional combined navigation algorithms will not be capable of meeting the accuracy requirements of navigation. In this paper, An AUV multi-source integrated navigation algorithm based on iterative smoothing of factor graph is designed to address the problem of increasing navigation accuracy errors due to the time-varying characteristics of observation noise of sub-navigation sensors and the fixed weight of measurement information assignment and the non-real-time nature of global optimization in standard factor graph. The algorithm first adds a belief function to the standard factor graph algorithm to suppress the time-varying measurement noise of sub-navigation sensors. Then, a sliding window is added to the improved factor graph algorithm to ensure real-time performance. Finally, based on the high-precision characteristics of inertial navigation in a short period of time, inertial navigation iteration is performed on the nodes that are about to be marginalized within the sliding window to improve navigation accuracy. Simulation experiments demonstrate the effectiveness of the algorithm.