Abstract: Aiming at the problem of 3D drone path planning for transportation materials in the affected mountainous areas, an elite ant colony algorithm is proposed. Firstly, the algorithm limits the range of pheromone concentration through an extreme value limiting strategy to prevent the algorithm from falling into a local optimum in the early stage. Secondly, the elite strategy is used to improve the pheromone concentration update formula and strengthen the influence of high-quality individuals on the population. Pheromone diffusion strategies are introduced to strengthen communication and collaboration between individuals closer to each other, in order to prevent the stagnation of the algorithm caused by insufficiently close ant individuals. Finally, the elite diffusion ant colony algorithm, traditional ant colony algorithm, genetic algorithm and firefly algorithm are applied to four examples of disaster-resistant drone transportation in mountainous areas, proving the superiority and effectiveness of the improved algorithm. The algorithm has a good adaptability to the problem of 3D path planning for UAV.

Key words: elite, diffusion, ant colony algorithm, unmanned aerial vehicle, three-dimensional path optimization