Application of the Naozhou Population of Larimichthys Crocea School Algorithm (DNPFS-OA) in Path Planning for AUV Clusters

Abstract

Aiming at the cooperative path planning problem of AUV cluster in underwater complex environment, this paper proposes a deep distributed optimization algorithm (DNPFS-OA) based on the intelligent behavior of Tanzhou Pseudosciaena crocea. This algorithm is different from the traditional leader-follower architecture. By establishing the deep coupling between the bio-pressure sensing model and the hydrodynamic equation, and combining with the communication-control collaborative optimization framework, the autonomous collaboration of AUV clusters in the three-dimensional strong disturbance environment is realized. The sea test shows that the path length of DNPFS-OA is reduced by 28.3%, the energy consumption is reduced by 35.7%, and the mapping efficiency is improved by 42.6% when the communication packet loss rate is 30%.