Task Allocation of Multiple AUVs and Simulation in Large Range Underwater Environment

This paper studies the task allocation problem of underwater multi-AUV system. The mathematical model for task allocation is established based on the characteristics of underwater multi-AUV system executing tasks. On this basis, we propose a task allocation method for multi-AUV system based on Self-Organizing Maps (SOM) neural network. The task allocation scheme is obtained by considering the competitive nature of SOM and addressing issues such as multiple task types, underwater currents, and moving task points that exist in the actual working environment of underwater multi-AUV system. The fitness function of the SOM task allocation algorithm is improved and validated through simulation comparisons. And the proposed algorithms in this paper are extended to larger underwater environment. Complex simulations involving multiple task types and large task volumes are conducted in large-scale underwater environment to validate the performance of the improved algorithm.