PSAO: An enhanced Aquila Optimizer with particle swarm mechanism for engineering design and UAV path planning problems

Abstract

Metaheuristic algorithms have become increasingly significant in solving complex optimization problems. To address the limitations of the original Aquila Optimizer (AO), such as insufficient local exploitation ability, low optimization precision, and slow convergence rate, an enhanced Aquila Optimizer (PSAO) for global optimization has been proposed. PSAO uses a better ergodic good point set to initialize the Aquila population and modifies the search method by employing the golden sine operator and the mechanism of self-learning and social learning based on particle swarm, followed by designing a nonlinear balance factor γ as the switching condition of the algorithm. The simulation experiments on benchmark functions and CEC2017 functions have verified that the PSAO has better global optimization ability and stronger robustness compared with other intelligent algorithms. Meanwhile, the contribution of each component that belongs to PSAO has been validated by ablation experiments for the CEC2022 test functions. To further illustrate the practical application potential of PSAO, PSAO is successfully applied to four typical engineering design problems, three of which reach the best fitness values. In addition, utilizing PSAO to solve the UAV trajectory planning problem by considering the objectives of trajectory length, altitude and corner of the flight process, the total flight cost is reduced by 60.22 %, 27.94 %, and 22.41 % compared with AO, PSO and Gold-SA, respectively, which proves its applicability and superiority in solving the real optimization problems.