Multi-UAV 3D Trajectory Planning Model Considering Task Priority

Based on the origin-destination points and planned takeoff times of multiple unmanned aerial vehicles (UAVs), and comprehensively considering multiple constraints such as spatiotemporal conflicts, the spatial distribution of restricted zones, and UAV flight performance, this paper establishes a multi-UAV 3D track trajectory collaborative planning model that takes task priority into account and aims to minimize both trajectory mileage cost and delay time cost. To address the computational challenges of the model, a Multi-Strategy Hiking Optimization Algorithm (MSHOA) was designed in this paper. The algorithm first employed a Logistic-Cubic hybrid mapping for population initialization to improve initial population quality; in the global exploration phase, the golden sine strategy and development position update strategy were integrated to perturb individuals so as to avoid local convergence of the algorithm, and the hiker interaction shortcut-searching strategy was applied to accelerate the algorithm’s convergence capability. To verify the feasibility and superiority of MSHOA, 12 benchmark test functions were selected to compare the solution performance of MSHOA with five classical intelligent optimization algorithms, and ablation experiments were implemented to verify the effectiveness of each improved strategy. Meanwhile, a practical UAV track planning case was used to comparatively analyze the influences of departure delay, task priority and spatiotemporal conflict constraints on multi-UAV scheduling results. The research results show that, compared with the traditional model without task priority settings, MSHOA increases total flight time, total delay time, and trajectory mileage by 0.89%, 12.60%, and 1.79%, respectively. Such slight performance loss is caused by the adaptation to multiple practical flight constraints, which can effectively avoid flight conflicts and better fit the engineering scenarios of real UAV collaborative scheduling, thus demonstrating stronger practicality. Compared with HOA, DBO, GWO, PSO and SSA, MSHOA reduces the iteration cost by 56.39%, 82.40%, 46.55%, 91.99% and 81.39%, respectively, and cuts down convergence iterations by 38.46%, 86.09%, 83.33%, 48.39% and 69.23% correspondingly, which further verifies the advantages of the improved algorithm in optimization accuracy and convergence speed.