A Dynamic Task Allocation Method for Unmanned Aerial Vehicle Swarm Based on Wolf Pack Labor Division Model

The dynamic task allocation problem for Unmanned Aerial Vehicle (UAV) swarms is characterized by dynamic uncertainty, high nonlinearity, and multimodality, which increasingly becomes a focal point and challenge within the realm of task allocation. This study investigates the mixed interaction modes of “individual-individual,” “individual-environment,” and “swarm-environment,” drawing inspiration from the labor division observed in wolf packs. We design a mechanism for alliance formation that is predicated on both individual democratic choice and centralized swarm decision-making, integrating the concept of alliance formation with a labor division model. Moreover, we propose a democratic-centralization model (DCM) that incorporates a bottom-up labor division approach aligned with the wolf pack alliance framework. This model is adeptly adapted to the dynamic task allocation for UAV swarms, employing a strategy of “individual cyclic competition plus overall coordinated decision-making.” We compare simulation experiments using the DCM with three contemporary, leading methods addressing the dynamic allocation problem for UAV swarms tasked with dynamic operations. The results demonstrate that our model can efficiently orchestrate the dynamic allocation of UAV swarms, exhibiting considerable dynamic adaptability, collaborative efficiency, and robustness.