Multi-Agent Task Allocation with Multiple Depots Using Graph Attention Pointer Network

The study of the multi-agent task allocation problem with multiple depots is crucial for investigating collaboration. Although many traditional heuristic algorithms can be adopted to handle concerned problem, they are not able efficiently obtain optimal or suboptimal solutions. To this end, a graph attention pointer network built in paper deal problem. Specifically, multi-head mechanism employed feature extraction nodes, and parallel two-way selection output introduced further improve performance cooperation efficiency allocation. Experimental results provided show that presented outperforms algorithms.