基于遗传算法的除草机器人多机路径规划研究  被引量:1

Research on multi-machine path planning of weeding robot based on genetic algorithm GUO Junhui

在线阅读下载全文

作  者:吴坚 马浩杰 张同锋 WU Jian;MA Haojie;ZHANG Tongfeng(School of Mechanical and Energy Engineering,Zhejiang University of Science and Technology,Hangzhou 310023,Zhejiang,China;Zhejiang Wason Cold Chain Technology Co.,Ltd.,Hangzhou 311100,Zhejiang,China)

机构地区:[1]浙江科技大学机械与能源工程学院,杭州310023 [2]浙江微松冷链科技有限公司,杭州311100

出  处:《浙江科技大学学报》2024年第5期357-368,共12页Journal of Zhejiang University of Science and Technology

基  金:浙江省食品物流装备技术研究重点实验室开放基金项目(KF2022002yb)。

摘  要:【目的】针对单个小型行间除草机器人续航能力差,无法独立完成大面积稻田除草任务的问题,提出了一种基于多染色体优化遗传算法(multi-chromosome optimized genetic algorithm,MGA)的多机协同路径规划方法。【方法】首先,根据稻田秧苗分布情况,将能耗最高的除草机器人的移动距离最小化作为优化目标,建立了多机协同路径规划模型;其次,设计了一种多染色体遗传算法,并引入逐代竞争机制、配对交换机制及自适应变异算子,以提升算法的最优解质量和鲁棒性;最后,在不同田形的模拟地图和稻田栅格地图上进行了多机协同路径规划仿真试验,并与传统遗传算法(genetic algorithm,GA)及自适应遗传算法(adaptive genetic algorithm,AGA)进行对比。【结果】在不同田形地图仿真试验中,多染色体优化遗传算法表现出了较高的最优解质量和鲁棒性,明显优于传统遗传算法和自适应遗传算法。在稻田栅格地图仿真试验中,优化遗传算法生成的能耗最高的除草机器人的移动距离相较于传统遗传算法缩短了9.7%,相较于自适应遗传算法缩短了8.0%;平均路径长度相较于传统遗传算法缩短了6.1%,相较于自适应遗传算法缩短了3.8%。搜索结果的标准差相较于传统遗传算法降低了34%,相较于自适应遗传算法降低了40%。【结论】多染色体优化遗传算法能有效缩短能耗最高的小型行间除草机器人的移动距离,有助于其在续航能力范围内完成作业任务。[Objective]In response to the problem of poor endurance of individual small inter-row weeding robots,which hinders their ability to independently and effectively complete large-area paddy field weeding tasks,a multi-machine cooperative path planning method was proposed on the basis of multi-chromosome optimized genetic algorithm(MGA).[Method]Firstly,based on the distribution of rice seedlings in the paddy field,the optimization objective of minimizing the movement distance of the most energy-consuming weeding robot was established to create a multi-machine cooperative path planning model;secondly,a multi-chromosome genetic algorithm was designed,incorporating mechanisms such as generational competition,paired exchange,and adaptive mutation operators to enhance the algorithm's quality of optimal solutions and robustness;finally,simulation experiments of multi-machine cooperative path planning were conducted on simulated maps of different types of farmland and grid maps of paddy fields,with results compared to traditional genetic algorithms(GA)and adaptive genetic algorithms(AGA).[Result]In the simulation experiment on maps of different types of farmland,the multi-chromosome optimized genetic algorithm demonstrates superior quality of optimal solutions and robustness,outperforming traditional genetic algorithms and adaptive genetic algorithms.In the simulation experiment on grid maps of paddyfields,the optimized genetic algorithm reduces the movement distance of the most energy-consuming weeding robot by 9.7%compared to traditional genetic algorithms and by 8.0%compared to adaptive genetic algorithms;the average path length decreases by 6.1%compared to traditional genetic algorithms and by 3.8%compared to adaptive genetic algorithms.Compared to traditional genetic algorithms,the standard deviation of search results decreases by 34%;compared to adaptive genetic algorithms,it decreases by 40%.[Conclusion]Multi-chro mosome optimized genetic algorithm can effectively reduce the movement distance of small inter-row weed

关 键 词:除草机器人 遗传算法 全覆盖路径规划 多机协同 

分 类 号:S224.15[农业科学—农业机械化工程] TP242[农业科学—农业工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象