遗传算法
分类
任务(项目管理)
计算机科学
算法
农业
实时计算
人工智能
排序算法
数学优化
跟踪(教育)
钥匙(锁)
算法设计
理论(学习稳定性)
自动化
数据挖掘
工程类
进化算法
领域(数学)
农业生产力
过程(计算)
作者
Yuan Bo Qin,Cunjun Li,Chaoting Wang,Maodong Fu,Chanling Pan,Ruibo Chen
标识
DOI:10.1016/j.atech.2025.101700
摘要
Agricultural Multi-Robot Systems (MRS) in smart agriculture are in a continuous stage of research and innovation. The scheduling of multiple intelligent and unmanned agricultural machines is an important direction for the future development of smart agriculture. While UAV applications in agriculture have gained attention, systematic solutions for multi-UAV task allocation under complex spatiotemporal and operational constraints remain under-explored, particularly those that are readily deployable in real production environments. In this paper, a Dual-Population Non-dominated Sorting Genetic Algorithm II (NSGA-II) UAV task allocation algorithm is proposed to address the challenge of task allocation for UAV operations in agricultural multi-robot systems. First, the algorithm simulates actual operational processes and designs a Simulated Farm Operation Assignment initial population generation method to improve the validity and objective values of the initial solutions. Then, a frequency mutation scheme is adopted to balance UAV workloads and enhance population diversity. Finally, a dual-population mechanism is introduced, combined with dynamic population proportions, to improve the convergence speed and global search capability of the algorithm at each stage. Experimental results on actual and public agricultural datasets show that the improved algorithm achieves better allocation results, demonstrating its applicability in handling large-scale farmland task allocation problems and its potential for real-world implementation in precision agriculture systems.
科研通智能强力驱动
Strongly Powered by AbleSci AI