UAV Path Optimization Using High-Altitude Data and Mutated Fruit-Fly Strategy

In uncrewed aerial vehicle (UAV) navigation, one of the foremost challenges is devising efficient flight paths that navigate complex environments while conserving energy and avoiding obstacles. The paper proposes a High-Altitude Platform aided Fruit Fly Optimization Algorithm (HAPFOA) as a novel approach to address sub-optimal path planning in dynamic complex environments by leveraging fruit fly's innate foraging strategies and adapting them to the three-dimensional challenges of UAV navigation at high altitudes. This approach applies the Fruit Fly Optimization Algorithm (FOA), where each UAV is analogous to a ‘fly' exploring potential pathways. The paths chosen by these UAVs are evaluated using a comprehensive ’smell' metric that integrates crucial navigational factors such as closeness to the target, obstacle avoidance, energy utilization, and maneuverability. This process is iterative, with each UAV continuously adjusting its flight path based on the ‘smell' metric feedback, aiming to find the route with the best possible ’smell' score. The algorithm helps to avoid obstacles during the vision phase. Cubic spline interpolation is used to refine and smoothen the UAV trajectories to ensure the practical application of these flight paths, making them more applicable and realistic for risky operations. Based on obstacle warnings from HAPUAV, the simulation results show an efficient route formulation by the proposed model with no idle time and a faster convergence rate. Another advantage of HAPFOA is its ability to preserve more energy than other models.