Artificial Intelligence Optimisation of UAV Flight Paths for Enhanced Fog Dispersal Efficiency

This paper introduces a comprehensive study of artificial intelligence (AI) optimisation techniques for unmanned aerial vehicle (UAV) flight path planning for fog dispersal operation. The research relates to the critical challenge of fog-related disruptions in the aviation industry, which result in major economic losses in excess of $100 million annually at major airports around the world. By combining powerful artificial intelligence algorithms, such as reinforcement learning, genetic algorithm and neural network, this study proposes an intelligent UAV based fog dispersal system capable of autonomous path optimisation and real-time adaption. The system uses MATLAB / simulink for the simulation of UAV dynamics and ANSYS Fluent / OpenFOAM for the fog behaviour modelling, which is integrated to machine learning algorithms, to be used for the dynamic navigation and decision-making. Simulation results show that AI-optimised UAV flight paths have 35% better coverage efficiency and 40% decrease in consumption of seeding agents when compared to traditional fixed pattern approaches. The proposed system has significant advantages in terms of visibility, operational efficiency and environmental sustainability, which could revolutionise fog management strategies at airports across the world.