DBO-AWOA: An Adaptive Whale Optimization Algorithm for Global Optimization and UAV 3D Path Planning

Abstract

The rapid expansion of unmanned aerial vehicle (UAV) applications in complex environments presents significant challenges in 3D path planning, particularly in overcoming the limitations of traditional methods for dynamic obstacle avoidance and computational efficiency. To address these challenges, this study introduces an adaptive whale optimization algorithm (DBO-AWOA), which incorporates chaotic mapping, nonlinear convergence factors, adaptive inertia mechanisms, and dung beetle optimizer-inspired reproductive behaviors. Specifically, the algorithm utilizes ICMIC chaotic mapping to enhance initial population diversity, a cosine-based nonlinear convergence factor to balance exploration and exploitation, and a hybrid strategy inspired by the dung beetle optimizer to mitigate stagnation in local optima. When evaluated on the CEC2017 benchmark suite, DBO-AWOA demonstrates superior convergence precision and robustness, achieving the lowest minimum and average values across 72% of test functions. In 3D path-planning simulations within mountainous environments, DBO-AWOA generates smoother, shorter, and safer trajectories compared to existing variants, with fitness values reduced by 5-25%. Although the algorithm demonstrates slight instability in highly dynamic hybrid functions, its overall performance marks an improvement in global optimization and UAV path planning.