Complete Coverage Search for Multiple Autonomous Underwater Vehicles Based on Neuronal Activity Reassignment

Multiple autonomous underwater vehicles (MAUVs) cooperative systems are extensively extensively utilized for conducting search missions of wrecked aircraft or submarines across vast oceanic areas. Ensuring safe and efficient complete coverage path planning is a critical challenge for MAUVs in performing these search tasks. In this paper, we present an efficient MAUVs cooperative adaptive coverage path planning algorithm. Specifically, we introduce the neuronal activity reassigned (NAR) algorithm as the initial step to devise optimal navigation paths for a single AUV. Through the utilization of both the AUV steering number optimization strategy and the deadlock point escape algorithm based on the glasius bio-inspired neural network (GBNN), the path planning process is jointly optimized. Then, the adaptive balance evolution (ABE) algorithm is designed with the AUV energy constraints as a guide to match each AUV to a subtask sea area with task volume differentiation by dividing the task sea area in multiple iterations based on the residual capacity difference of individual AUVs. Finally, the simulation and experiment results demonstrate that the MAUVs cooperative adaptive coverage path planning algorithm proposed in this paper offers several advantages, including optimal task balance, extensive coverage, minimal path repetition rate, reduced occurrence of deadlock points, and decreased steering times. These benefits are instrumental in facilitating the MAUVs system’s efficient execution of refined search tasks.