Comparative Analysis of Trajectory Optimization Methods for Unmanned Aerial Vehicles in Dynamic Environments

The optimization of the trajectory of unmanned aerial vehicles (UAVs) is crucial to maximize operational efficiency and resource conservation. five trajectory optimization approaches for UAVs: the Dinkelbach Algorithm, Mixed-Integer Non-Convex Optimization, Newton’s Method, Genetic Algorithm, and Conjugate Gradient Method. Non-convex methods offer algorithmic flexibility, while the Dinkelbach Algorithm is renowned for its robustness and ability to solve mixed-integer optimization problems. Using detailed simulations, we evaluate the performance of these five approaches in terms of computation time, precision of the optimal trajectory, and resilience to environmental disturbances. Our results reveal distinct advantages for each method, thus providing valuable insights for the development of efficient trajectory planning systems tailored to the specific needs of UAV missions in diverse and dynamic environments.