Analysis of Autonomous Underwater Vehicle (AUV) navigational states based on complex networks

Abstract To determine the navigational states of an autonomous underwater vehicle (AUV), a data analysis approach of AUV navigation based on complex networks is proposed in this study. First, the noise in AUV navigation data is eliminated by the projection-density peaks clustering algorithm (Pro-DPCA), and the weighted complex networks of the de-noising data are constructed. The nodes of networks characterize AUV navigation states. Subsequently, we compute the topological statistics of the complex networks to obtain the fluctuation patterns of the AUV navigational data. This is used to analyse AUV navigational states. For verifying the approach, the heading data at different depths is analysed in our experiments. The experimental results indicate that the topological statistics of the complex networks accurately describe the navigational states of AUV at different depths.