Liu Yangyang, Zhang Pengyang, Ru Yu, Wu Delin, Wang Shunli, Yin Niuniu, Meng Fansheng, Liu Zhongcheng
School of Engineering, Anhui Agricultural University, Hefei, China.
School of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing, China.
Front Plant Sci. 2022 Sep 16;13:998962. doi: 10.3389/fpls.2022.998962. eCollection 2022.
The complex environments and weak infrastructure constructions of hilly mountainous areas complicate the effective path planning for plant protection operations. Therefore, with the aim of improving the current status of complicated tea plant protections in hills and slopes, an unmanned aerial vehicle (UAV) multi-tea field plant protection route planning algorithm is developed in this paper and integrated with a full-coverage spraying route method for a single region. By optimizing the crossover and mutation operators of the genetic algorithm (GA), the crossover and mutation probabilities are automatically adjusted with the individual fitness and a dynamic genetic algorithm (DGA) is proposed. The iteration period and reinforcement concepts are then introduced in the pheromone update rule of the ant colony optimization (ACO) to improve the convergence accuracy and global optimization capability, and an ant colony binary iteration optimization (ACBIO) is proposed. Serial fusion is subsequently employed on the two algorithms to optimize the route planning for multi-regional operations. Simulation tests reveal that the dynamic genetic algorithm with ant colony binary iterative optimization (DGA-ACBIO) proposed in this study shortens the optimal flight range by 715.8 m, 428.3 m, 589 m, and 287.6 m compared to the dynamic genetic algorithm, ant colony binary iterative algorithm, artificial fish swarm algorithm (AFSA) and particle swarm optimization (PSO), respectively, for multiple tea field scheduling route planning. Moreover, the search time is reduced by more than half compared to other bionic algorithms. The proposed algorithm maintains advantages in performance and stability when solving standard traveling salesman problems with more complex objectives, as well as the planning accuracy and search speed. In this paper, the research on the planning algorithm of plant protection route for multi-tea field scheduling helps to shorten the inter-regional scheduling range and thus reduces the cost of plant protection.
山区复杂的环境和薄弱的基础设施建设使植保作业的有效路径规划变得复杂。因此，为改善丘陵和山坡地区复杂茶树植保的现状，本文开发了一种无人机多茶园植保路线规划算法，并与单区域全覆盖喷雾路线方法相结合。通过优化遗传算法（GA）的交叉和变异算子，根据个体适应度自动调整交叉和变异概率，提出了动态遗传算法（DGA）。然后在蚁群优化（ACO）的信息素更新规则中引入迭代周期和强化概念，以提高收敛精度和全局优化能力，提出了蚁群二进制迭代优化（ACBIO）。随后对这两种算法进行串行融合，以优化多区域作业的路线规划。仿真测试表明，本研究提出的蚁群二进制迭代优化动态遗传算法（DGA-ACBIO）在多茶园调度路线规划中，与动态遗传算法、蚁群二进制迭代算法、人工鱼群算法（AFSA）和粒子群优化（PSO）相比，最优飞行距离分别缩短了715.8米、428.3米、589米和287.6米。此外，搜索时间比其他仿生算法减少了一半以上。该算法在解决目标更复杂的标准旅行商问题时，在性能和稳定性方面保持优势，同时具有规划精度和搜索速度。本文对多茶园调度植保路线规划算法的研究有助于缩短区域间调度范围，从而降低植保成本。
