3D Path Planning for UAVs Based on Improved Red Kite Optimization Algorithm in Complex Threat Environments

Abstract

Aiming at the problems of slow convergence speed, low optimization accuracy, susceptibility to local optima, and insufficient stability of the traditional Red Kite Optimization Algorithm (ROA) for unmanned aerial vehicle (UAV) path planning in complex three-dimensional environments, this paper proposes an Improved Red Kite Optimization Algorithm (IROA). Firstly, the Personal History Memory Mechanism (PHMM) is introduced, which replaces random perturbation with weighted historical experience to enhance the directional search capability of the algorithm. Secondly, a cosine-based nonlinear adaptive weight strategy is adopted to dynamically balance global exploration and local exploitation. Finally, a greedy selection strategy is embedded to ensure the monotonically non-deteriorating quality of population solutions. In the experimental section, the performance of the improved algorithm is first verified using benchmark test functions, and then four types of UAV flight path-planning scenarios are constructed in the simulation environment for comparison. The results demonstrate that IROA outperforms ROA, WOA, HHO, DBO, WHO, and other algorithms in terms of optimal value, mean value, and stability indicators. Its convergence speed and optimization accuracy are significantly improved, and high-quality paths can be generated even in obstacle scenarios of varying complexity, which verifies the feasibility of the proposed algorithm for three-dimensional UAV path planning.