A Cooperative Area Allocation and Adaptive Coverage Planning Method for Multi‐AUV Underwater Search and Rescue

This paper proposes an integration method for area allocation and coverage detection in multi‐AUV collaborative underwater search and rescue missions, which addresses challenges such as complex region allocation and insufficient search efficiency. Specifically, Improved Sparrow Search Algorithm based on Deep Q ‐Network (ISSA‐DQN) is designed to address region connectivity and platform heterogeneity issues during multi‐AUV mission allocation, ensuring optimal assignment of mission regions to individual AUVs. These optimizations enhance overall collaborative efficiency in the AUV swarm system. Furthermore, Multi‐AUV Boundary Adaptive Neural Network Coverage Algorithm (MBANCP) is proposed to further enhance global coverage search performance. This algorithm generates efficient coverage paths for complex, irregular areas, enabling complete detection of the target area and localization of search and rescue targets. Through comparative validation with existing area allocation and coverage detection methods, experiments demonstrate that these proposed methods significantly improve execution efficiency and search performance for multi‐AUV collaborative Search and Rescue (SAR) missions, providing an efficient and reliable solution for area‐based SAR operations in complex underwater environments.