Statistical results of applying a nature-inspired hybrid optimization algorithm to the path planning problem

UAV path planning is a complex optimization problem involving multiple objectives, such as avoiding obstacles while traveling from a starting point to a destination, minimizing energy-time-dependent costs, and ensuring safe flight in a dynamic environment. This problem generally requires multidimensional approaches involving distance, time, energy, and safety. It can be modeled in stationary or moving environments and in 2D/3D spaces. The mathematical model is usually dependent on several constraints, including nonlinear factors such as altitude, angle, velocity, and collision risk. Although these types of problems are attempted to be solved with classical deterministic search methods such as A^* and Dijkstra, these methods often get stuck at sub-optimum solutions due to the dynamic nature of the environment and the need for optimality. Therefore, meta-heuristic optimization algorithms are derivative-free, nature-inspired, population-based, or single-solution-based heuristics. They scan the search space with a general exploration + local exploitation strategy. They are suitable for approaching the global optimum and avoiding getting stuck at local solutions. Because of these characteristics, meta-heuristic algorithms are particularly preferred in high-dimensional, multi-criteria, and dynamic problems such as UAV path planning. This study examines the most studied and recently defined metaheuristic optimization algorithms in the literature and applies them to the route planning problem. The ZHO, ZOA, WHOA, and COA algorithms, previously unused in UAV route planning, were adapted to solve this problem, and their results were statistically analyzed. These results are visualized in tables and graphs to clearly demonstrate their advantages.