Multi-AUV Data Retrieval Method Based on Multiple Signal Classification Algorithm

This paper investigates an ASV-based multi-AUV data retrieval algorithm for subsea data collection application scenarios, in which AUVs float to the surface after completing their data collection tasks on the seafloor and wait for retrieval by ASV. The main optimization objective of the algorithm is the waiting time of the AUVs to guarantee the freshness of the data collected. First, a multiple signal classification algorithm is used to predict the emergence points of AUVs; subsequently, an improved particle swarm optimization (PSO) algorithm is designed to plan the paths for the ASV to visit several AUVs emergence points and guide the ASV to retrieve the AUVs and their collected data. The algorithm proposed in this paper shows excellent performance in terms of energy consumption and data collection in the simulation scenario.