农机化研究

针对植保无人机在作业过程中如何快速找到一条更优的全局路径，提出了一种改进人工鱼群算法(T-AFSA)。利用Tent混沌映射所生成的混沌序列对种群初始化，丰富了种群的多样性，提高了人工鱼群初始解的质量；引入黄金正弦算法对适应度高的人工鱼个体进行优化，让它们更好地领导种群的觅食和追尾行为；采用自适应策略对人工鱼个体的视野和步距进行改进，平衡了算法的全局搜索能力和局部搜索能力；删除路径中的冗余节点，去除不必要的转折点，找到全局中的更优路径；将所得的路径利用B样条曲线进行平滑处理，有利于植保无人机进行路径跟踪。仿真实验表明：改进算法能够解决传统人工鱼群算法计算精度低和后期收敛速度变慢的问题，可以为植保无人机快速规划出一条从起点到终点与障碍物无碰撞、平滑且距离较短的路径，方案具有可行性和有效性。

Aiming at how to quickly find a better global path for plant protection drones during operation, a modified Artificial Fish Swarm Algorithm(T-AFSA) was proposed. Using the chaotic sequence generated by the Tent chaotic map to initialize the population enriches the diversity of the population and improved the quality of the initial solution of the artificial fish swarm.The Golden Sine Algorithm was introduced to optimize the individual artificial fish with high fitness, so that they can better lead the population's foraging and tail-tracking behavior. The self-adaptive strategy was used to improve the field of view and step distance of the individual artificial fish, which balanced the global search ability and local search ability of the algorithm.Deleted redundant nodes in the path, removed unnecessary turning points, and found a better path in the whole world.The B-spline curve was used to smooth the resulting path, wich was beneficial to the path tracking of the plant protection UAV.The simulation experiments showed that the algorithm in this paper could improve the low calculation accuracy of the traditional artificial fish swarm algorithm and the slow convergence speed in the later stage, which could quickly plan a smooth and short path from the starting point to the end point for the plant protection UAV without collision with obstacles, which proved the feasibility and effectiveness of this improvement scheme.