Research on Task Allocation of Heterogeneous UAVs Based on DLR-CBBA

This paper considers the problem of heterogeneous UAVs task allocation in a distributed multi-UAV system under strict time constraints, where other constraints need also be considered. Due to the limited capability of each UAV, multiple UAVs are required to perform a task simultaneously. The presented algorithm termed double layer cycle consensus based bundle algorithm (DLC-CBBA) is an extension to consensus-based bundle algorithm(CBBA). This paper improves the baseline CBBA from the aspects of capability matrix, objective function, task bundle construction, consistency processing, and so on. Extensive simulation results show that conflict-free task assignments are gained based on DLC-CBBA, and the convergence time is much less than the discrete particle swarm optimization (DPSO) when performing multiple-UAV tasks.