Abstract: In view of the fact that many current UAV track planning based on two‑dimensional maps cannot meet the plant protection needs of hilly and mountainous areas, in this article, the classical ant colony algorithm was improved and smoothed. An innovative three‑dimensional trajectory planning method was proposed. The initial pheromone was optimized to speed up the initial convergence of the ant colony; the heuristic function was improved to ensure the rationality of the trajectory, the elite ant colony update strategy was introduced to enhance the exploration ability of subsequent ant colonies, the pheromone update mechanism was dynamically adjusted, ensuring strong explorability in the early stage of the algorithm, and the algorithm's convergence speed could be accelerated. Simulation experiments showed that the improved ant colony algorithm had the advantages of shorter paths, lower energy consumption, and smoother trajectories. It also improved that ant colony algorithm demonstrated good environmental adaptability. Compared with the classical ant colony algorithm and A* algorithm, the improved ant colony algorithm shortened the track length by 40.86% and 2.83% respectively, the total energy consumption of the UAV was reduced by 45.04% and 12.13% respectively, and the algorithm running time was saved by 31.32% and 8.86%, respectively. In a three‑dimensional complex environment with obstacles, the obstacle avoidance planning ability of the improved ant colony algorithm had been verified.

Key words: UAVs for plant protection, improved ant colony algorithm, hilly and mountainous areas, three?dimensional trajectory planning, obstacle avoidance

摘要： 针对当前许多基于二维地图的无人机航迹规划难以满足丘陵山地的植保需求，对经典蚁群算法进行改进与平滑处理，提出一种创新的三维航迹规划方法。优化初始信息素，加快蚁群初期收敛速度；改进启发式函数，确保航迹的合理性；引入精英蚁群更新策略，增强后续蚁群的探索能力；动态调整信息素更新机制，保证算法前期的强探索性并加快算法收敛速度。仿真试验表明，改进蚁群算法具有路径更短、能耗更低、航迹更平滑等优势，同时环境适应性良好。改进蚁群算法相比于经典蚁群算法和A*算法，航迹长度分别缩短40.86%和2.83%，无人机总能耗分别减少45.04%和12.13%，算法运行时间分别节约31.32%和8.86%；在有障碍的三维复杂环境中，验证改进蚁群算法的避障规划能力。

关键词: 植保无人机, 改进蚁群算法, 丘陵山地, 三维航迹规划, 避障