Task allocation for Multi-AUV system: A review

Systematization, and clustering, as the next key direction for AUVs to perform their functions, is becoming an area of enthusiasm among related researchers. And this must be inseparable from task allocation. Task allocation, as we all know, is the implementation of an AUV subtask allocation based on the position, attitude, and status of each AUV in the region after the multi-AUV system is assigned the overall task. Due to the different forms of tasks, AUVs with the same function may be required to perform a certain task, and there is also the possibility that a complex task requires heterogeneous AUV clusters to cooperate to accomplish it. To flexibly cope with different situations, corresponding allocation algorithms have been designed by predecessors. The purpose of this paper is to analyze the existing literature on the models and algorithms of task allocation for multi-AUV system and to provide a macroscopic impression for researchers in this field to improve their algorithms more comprehensively. The paper roughly classifies the algorithms into centralized and distributed. Also, it covers the enumeration of key algorithms, the analysis of the principles, the application situations, and the comparison of advantages and disadvantages; finally, we look forward to the future directions worthy of further research in this field and provides references. •When researchers pay more attention to path-planning of AUVs, the review is written in the field of task allocation of multi-AUV systems firstly.•The paper covers the enumeration of key algorithms, the analysis of the principles, the application situations, and the comparison of advantages and disadvantages.•We choose five algorithms to simulate task allocation with different scales.