A Hybrid Algorithm for Optimising AUV Path Planning and Precise Node Localization in UWSNs Using Cluster-Based Geometric Search Method

Underwater Wireless Sensor Networks (UWSNs) play a vital role in aquatic monitoring and marine exploration, yet they face persistent challenges due to limited communication bandwidth, energy constraints, and the complex mobility of Autonomous Underwater Vehicles (AUVs). This paper presents a novel hybrid optimization framework that combines an Energy-Aware K-Means clustering algorithm with a terrain-sensitive A* path planning method. The clustering mechanism groups sensor nodes based on both spatial coordinates and remaining energy levels, ensuring balanced data aggregation. Simultaneously, the enhanced A* algorithm navigates the AUV through energy-efficient paths, accounting for underwater terrain variations and current-induced drift. This integrated strategy enhances localization accuracy, reduces unnecessary AUV movement, and significantly extends network lifetime. Simulation results confirm that the proposed approach achieves substantial gains in energy efficiency, reducing energy consumption by over 80%, shortening traversal distance by 87.5%, and improving overall network sustainability by 91.4%. These outcomes demonstrate the effectiveness of the hybrid model in optimizing data collection and node coordination in dynamic, large-scale UWSNs.