Multi-UAV Cooperative Path Planning Using a Behavior-Adaptive Aquila Optimizer Under Multiple Constraints

Addressing the challenges of high dimensionality, strong nonlinearity, and multiple constraints in multi-UAV cooperative path planning, this paper proposes a Behavior-Adaptive Aquila Optimizer (EAO) achieved by enhancing Aquila Optimizer (AO). EAO constructs a multi-strategy cooperative framework that integrates a periodic diversity maintenance mechanism, a diversity-based dynamic neighborhood guidance mechanism, a narrowed exploitation behavior based on neighborhood differential evolution, and a search-state-aware adaptive behavior selection mechanism. Through dynamic behavior adjustment during the search process, the proposed algorithm improves search performance and stability. To validate its effectiveness, EAO was systematically evaluated on the CEC2017 and CEC2020 benchmark suites and compared with the original AO and 13 representative high-performance optimization algorithms. Parameter sensitivity analysis, an ablation study, and an exploration-exploitation experiment were also conducted. The results show that EAO achieves the best overall performance ranking. Furthermore, EAO was applied to multi-UAV cooperative path-planning simulations in complex environments that considered UAV dynamic constraints. Comparative experiments with five competitive algorithms demonstrate that EAO achieves superior performance in terms of path-planning fitness, number of effective trajectories, and runtime. Compared with AO, EAO improves the average fitness by 80.42%, 81.25%, 81.34%, and 84.84% across different map environments, confirming its feasibility and effectiveness for multi-UAV cooperative path planning.