Research on UAV Trajectory Planning Based on Improved Whale Optimization Algorithm

This research endeavors to enhance the whale optimization algorithm for addressing the intricate flight path planning challenges of unmanned aerial vehicles (UAVs) in both 2D and 3D spaces. Addressing the shortcomings of the initial algorithm, including limited initial population diversity and an uneven balance between global and local searches, we introduce a set of novel approaches. For 2D track planning, we incorporate a reverse learning mechanism and a nonlinear convergence factor to bolster the algorithm's initial population diversity and search equilibrium. Experimental data demonstrates that the refined algorithm outperforms in both benchmark tests and 2D raster map simulations. Considering the heightened freedom and complexity of 3D track planning, we advance a multi-strategy fused whale optimization algorithm, amalgamating various strategies to substantially elevate the algorithm's initial solution quality and searching capability, while effectively steering clear of local optima. Additionally, B-spline curves are employed to smooth and optimize the flight path. Three-dimensional simulation outcomes reveal that the proposed algorithm surpasses the improved version across multiple metrics. Future endeavors could delve deeper into the algorithm's applications in dynamic environments and multiobjective optimization.