SGEO: Equalization Optimizer with Hybrid Learning Strategies for UAV Path Planning in Complex 3D Environments
Meng Zheng, Qing He, Q. H. He · Research Square · 2024
Abstract Unmanned Aerial Vehicle (UAV) route planning is an intricate issue that requires the comprehensive consideration of multiple factors and the combination of suitable strategies to achieve efficient and safe flight paths. Its purpose is to plan a suitable navigational path for the drone to achieve a specific task or avoid obstacles. In practical applications, UAVs are usually required to accomplish tasks in a variety of complex environments, resulting in the feasible flight paths will be reduced and path planning for UAVs becomes difficult. Accordingly, we design a simple yet useful planner, called Self-adaptive Golden Equilibrium Optimization algorithm (SGEO). The proposed algorithm combines the new self-adaptive approach, the acceleration control factor and the Golden-SA strategy in order to balance the exploitation and exploration capabilities. The novel self-adaptive strategy is used to expand the global search range of the algorithm and enhance its ability for global exploration; the acceleration control factor is introduced to control the parameters a1 and a2 and improve the convergence ability of the algorithm; the Golden-SA strategy helps the candidate particles achieve better balance between different dimensions based on the golden ratio and sine function, this enables the particles to explore the search space more comprehensively. The limitations in UAV route planning are translated into the objective function, and four more algorithms are introduced to compare with SGEO in two different circumstances. The simulation findings reveal that SGEO excels at three-dimensional path planning for UAV in complicated environments.