浙江农业学报 ›› 2025, Vol. 37 ›› Issue (11): 2364-2375.DOI: 10.3969/j.issn.1004-1524.20240875

• 生物系统工程 • 上一篇    下一篇

基于YOLOv5-7.0草莓采摘分级机器人的设计与试验

邱琰1,2(), 叶自然2, 谭向峰2, 代梦迪2, 葛世豪3, 阮贇杰3, 赵先亮1,*(), 孔德栋2,*()   

  1. 1.浙江科技大学 生物与化学工程学院,浙江 杭州 310023
    2.浙江省农业科学院 数字农业研究所,浙江 杭州 310021
    3.浙江大学 生物系统工程与食品科学学院,浙江 杭州 310058
  • 收稿日期:2024-10-12 出版日期:2025-11-25 发布日期:2025-12-08
  • 作者简介:邱琰(1997—),男,江苏睢宁人,硕士,主要从事农业机械化自动化研究。E-mail:87359948@qq.com
  • 通讯作者: *赵先亮,E-mail:zlzhao@zust.edu.cn;孔德栋,E-mail: kongdd@zaas.ac.cn
  • 基金资助:
    浙江省农业科学院地方合作项目(SY202301);浙江省农业关键核心技术攻关项目

Design and experiment of strawberry picking and grading robot based on YOLOv5-7.0

QIU Yan1,2(), YE Ziran2, TAN Xiangfeng2, DAI Mengdi2, GE Shihao3, RUAN Yunjie3, ZHAO Xianliang1,*(), KONG Dedong2,*()   

  1. 1. School of Biological & Chemistry Engineering, Zhejiang University of Science and Technology, Hangzhou 310023, China
    2. Institute of Digital Agriculture, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China
    3. College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China
  • Received:2024-10-12 Online:2025-11-25 Published:2025-12-08

摘要:

针对当前草莓植物工厂中采摘分级工作量大、人工操作难等问题,设计了一种适用于植物工厂的智能草莓采摘分级机器人。分析了整机结构及工作原理,基于理论分析与选型计算优化设计了阿克曼式底盘移动控制系统、伸缩式电动剪刀及仿人型机械手;本文利用YOLOv5-7.0算法实现了草莓分级判定与果梗生长方向预测,最后开展了整机采摘分级性能测试。试验结果表明,该机器人工作性能稳定,采摘与分级动作连贯,草莓平均分级准确率达到87.44%,实现了精准定位和轻柔抓取目标,显著提高了工作效率。

关键词: 农业机器人, 草莓, 植物工厂, 采摘, 分级

Abstract:

In order to solve the problems of large workload and difficult manual picking in strawberry plant factories, an intelligent strawberry picking and grading robot for plant factories was designed. The structure and working principle of the robot were analyzed, and its key components were optimized based on theoretical analysis and selection calculations, including the Ackermann type chassis movement control system, telescopic electric scissors, and the structure of humanoid underactuated manipulator. The YOLOv5-7.0 algorithm was employed to achieve strawberry grading and predict the direction of peduncle growth. Ultimately, a performance test of the machine’s picking and grading capabilities was conducted. The test results showed that the robot exhibited stable performance and consistent actions in picking and grading, with an average accuracy of 87.44% for strawberry grading. This achievement enables precise targeting and gentle grasping of the fruit, significantly enhancing work efficiency.

Key words: agricultural robot, strawberry, plant factory, picking, grading

中图分类号: