Multi-stage task allocation strategy for UAV clusterin multi-object tracking

Multi-unarmed aerial vehicle (UAV) collaborative mission planning is a key technology for the intelligent development of UAV clusters at this stage, where mission assignment under multiple constraints is a core part of UAV mission planning technology. The poor planning ability of an UAV swarm often leads to resource waste and revenue reduction during mission execution, while the adoption of task allocation in the multi-target tracking of UAV clusters guarantees the highest profit. Algorithms of traditional pigeon-inspired optimization(PIO) and contract network protocol reflect defects of low timeliness and high time costs, respectively. Compared with these methods, the multilevel PIO algorithm maximizes the efficiency and profitability of the overall tracking task. A staged UAV cluster task allocation architecture was constructed to complete the global optimal initialization before tracking, and then a parallel auction contract network was employed to further perform the local optimal redistribution of each UAV. The simulation results suggest that this scheme not only guides cluster allocation in stages but also decreases the negotiatory time of a UAV cluster to in-crease the integral profitability under multiple constraints, such as prohibited areas.