基于遗传和蚁群交互算法的穴盘苗稀植移栽路径优化
CSTR:
作者:
作者单位:

石河子大学机械电气工程学院,石河子 832000

作者简介:

蔡继萌,E-mail:cjmimmortal@foxmail.com

通讯作者:

王卫兵,E-mail:wwbshz@163.com

中图分类号:

TP18

基金项目:

国家自然科学基金项目(61763042)


Optimizing path of sparse planting and transplanting plug seedlings based on genetic-ant colony interactive algorithms
Author:
Affiliation:

College of Mechanical and Electrical Engineering, Shihezi University, Shihezi 832000, China

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    针对穴盘苗移栽到低密度穴盘路径规划效率低下问题,基于遗传算法和蚁群算法提出蚁群-遗传(ant colony-genetic optimization algorithm,ACGO)和遗传-蚁群(genetic-ant colony optimization algorithm,GACO)交互算法进行稀植移栽路径优化。通过仿真试验,使用固定顺序法和其他5种算法计算从72-32、72-50、128-50、128-32孔穴盘的移栽路径长度,对比分析不同算法在优化路径长度和计算时间上的差异,并通过相对标准差评估算法的稳定性。结果显示,在72孔到32孔穴盘移栽中,对比固定顺序法,GACO算法的平均路径长度缩短59.3%,平均计算时间为5.15 s,相对标准差约为1.5%;ACGO算法的平均路径长度缩短19.2%,平均计算时间为13.50 s,相对标准差约为1%。进一步研究显示,ACGO算法在200孔移栽至72孔和105孔场景的优化效果弱于贪婪算法,而GACO算法在不同孔数组合和缺苗数下展现出更高的普适性和稳定性。研究表明,ACGO和GACO 2种交互算法均可提升原算法的性能,但GACO算法在处理复杂稀植移栽路径规划问题时表现更为优越。

    Abstract:

    The ant colony-genetic optimization algorithm(ACGO) and genetic-ant colony optimization algorithm(GACO) interactive algorithms based on genetic algorithm and ant colony algorithm was proposed to optimize the path of sparse planting transplanting to solve the problem of low efficiency in planning the path of transplanting plug seedlings to low-density plug trays. The fixed sequence method and five other algorithms were used to calculate the length of transplanting path for plug trays with holes from 72-32,72-50,128-50, and 128-32 holes through simulation experiments. Comparative analyses were conducted on the performance of algorithms in optimizing the length of transplanting path and the time of calculation. The stability of the algorithm was evaluated by the relative standard deviation. The results showed that the GACO algorithm reduced the average length of path in transplanting plug tray with 72 to 32 holes by 59.3% compared with the fixed sequence method, with an average time of calculation of 5.15 seconds and a relative standard deviation of approximately 1.5%. The ACGO algorithm reduced the average length of path by 19.2%, with an average time of calculation of 13.50 seconds and a relative standard deviation of approximately 1%. The results of further studies showed that the optimization effect of ACGO algorithm in the scenarios of transplanting from 200 holes to 72 holes and 105 holes was weaker than that of greedy algorithm, while GACO algorithm had higher universality and stability under different combinations of hole numbers and numbers of missed seedling. It is indicated that both interactive algorithms have improved the performance of the original algorithm, but the GACO algorithm performs better in dealing with problems in planning path of complex sparse planting and transplanting plug seedlings. It will provide a strong reference basis for optimizing the path of sparse planting and transplanting plug seedlings.

    参考文献
    相似文献
    引证文献
引用本文

蔡继萌,王卫兵,曲家灏,郭小龙,李国栋,吴潇雨.基于遗传和蚁群交互算法的穴盘苗稀植移栽路径优化[J].华中农业大学学报,2025,44(4):248-258

复制
分享
相关视频

文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2024-06-30
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2025-07-24
  • 出版日期:
文章二维码