Ameliorated Moth-Flame Optimization Algorithm for UAV Path Planning Problem
Xiaodong Zhao, Boye Li, Xianliang Zhang · 2025
Aiming at the problem of unmanned aerial vehicles (UAV) path planning in complex environments, this paper proposes an ameliorated Moth-Flame Optimization algorithm (AMFO). The innovations of AMFO lie in: firstly, a control parameter is designed between flames and moths to balance the algorithm's exploitation and exploration capabilities. Secondly, updating the moth positions by combining golden sine search and spiral search strategies, thereby improving the algorithm's search accuracy. Finally, employing a Gaussian mutation mechanism to enhance the algorithm's ability to escape local optima. Through experimental comparisons on 23 standard test functions, AMFO demonstrates superior convergence speed and accuracy in most optimization problems. Additionally, in benchmark scenarios based on real digital model maps, AMFO successfully solves the path optimization problem for UAVs under feasibility and safety constraints. The experimental results indicate that AMFO exhibits significant effectiveness and practicality in UAV path planning in complex environments.