Abstract
[Objective] In the handling process of the automatic guided vehicle (AGV), accurate visual positioning of tray holes, particularly those that are small, distorted, and of low contrast, poses a significant challenge. Therefore, a visual positioning method for tray holes in AGVs based on improved YOLOv 5s is proposed. [Methods] Combining the channel shuffling operation improvement of ShuffleNetV 2 and the CBAM attention mechanism improvement, the basic YOLOv 5s framework is improved to focus on the sub-pixel level boundary blurred hole position areas in the deformation key region. Based on the SIoU loss function, attention is paid to the micro hole positions, and the spatial three-dimensional coordinates of the tray hole positions in the camera coordinate system are calculated to obtain the transformation relationship from the camera coordinate system to the hole position area coordinate system. The improved YOLOv 5s framework is utilized to output the three-dimensional coordinates of the tray hole positions in the AGV robotic arm coordinate system. [Results] The experimental results show that the experimental method can effectively capture sub-pixel level positioning accuracy boundaries, with an absolute error of less than 0.03 cm and a relative error of less than 0.83%. The F1 score and mAP index are 95.2% and 94.8%, respectively. The number of floating-point operations, parameter count, and model volume are 4.8 G, 2.6 M and 4.28 MB, respectively. [Conclusion] The experimental method effectively solves the visual positioning problem of small, deformed, and low contrast holes in the tray holes that need to be positioned, and improves the efficiency of AGV tray handling.
Publication Date
1-23-2026
First Page
79
Last Page
85
DOI
10.13652/j.spjx.1003.5788.2025.60141
Recommended Citation
Sheng, CUI; Fangli, TANG; Liangyu, ZHENG; Weili, ZENG; and Weiwei, QU
(2026)
"Visual positioning method for AGV tray holes based on improved YOLOv 5s,"
Food and Machinery: Vol. 42:
Iss.
1, Article 11.
DOI: 10.13652/j.spjx.1003.5788.2025.60141
Available at:
https://www.ifoodmm.cn/journal/vol42/iss1/11
References
[1] 闫冠宇,蔡泽平,梁家海,等.一种基于全局视觉的 AGV 目标检测与定位算法 [J].北部湾大学学报,2024,39(4):68-75,86.YAN G Y,CAI Z P,LIANG J H,et al.An AGV object detection and localization algorithm based on global vision [J].Journal of Beibu Gulf University,2024,39(4):68-75,86.
[2] 王 辉,袁 斌,王 伟 博,等.基 于 视 觉 和 RFID 复 合 导 航 的 AGV设计与研究 [J].机床与液压,2022,50(7):87-91.WANG H,YUAN B,WANG W B,et al.Design and research of AGV based on vision and RFID integrated navigation [J].Machine Tool & Hydraulics,2022,50(7):87-91.
[3] 吕开旺,王明睿,刘振国,等.视觉引导机器人协作 AGV 上下料系统研究 [J].制造业自动化,2022,44(1):123-126.LYU K W,WANG M R,LIU Z G,et al.Research on vision guided robot cooperating with AGV loading and unloading system [J].Manufacturing Automation,2022,44(1):123-126.
[4] 雷杏子,王树才,龚东军,等.基于 YOLOv 5s的筐装禽蛋上料机器人视觉定位方法 [J].华中农业大学学报,2024,43(3):302-310.LEI X Z,WANG S C,GONG D J,et al.A method for visually positioning loading robot of basket-packed poultry eggs based on YOLOv 5s[J].Journal of Huazhong Agricultural University,2024,43(3):302-310.
[5] 余杰先,张中华,陈凌玲,等.移动机器人在自动化生产线上下料中的定位方法 [J].机电工程技术,2023,52(10):196-200.YU J X,ZHANG Z H,CHEN L L,et al.Positioning method of mobile robots in the process of automatic production line loading and unloading [J].Mechanical & Electrical Engineering Technology,2023,52(10):196-200.
[6] 刘泽平,刘明兴,麻方达,等.基于移动目标检测和目标追踪的 全 局 视 觉 AGV 的 定 位 算 法 [J].青 岛 大 学 学 报 (自 然 科 学版),2022,35(1):85-92.LIU Z P,LIU M X,MA F D,et al.A positioning algorithm for global vision guided AGV based on moving target and target tracking [J].Journal of Qingdao University (Natural Science Edition ),2022,35(1):85-92.
[7] 刘蕾.基于 2D视觉的复合 AGV 机器人上下料的定位方法 [J].信息技术与信息化,2022 (7):58-61.LIU L.A positioning method for loading and unloading of composite AGV robots based on 2D vision [J].Information Technology and Informatization,2022 (7):58-61.
[8] 邹斌,张聪.基于 Faster R-CNN 的密集人群检测算法 [J].计算机应用,2023,43(1):61-66.ZOU B,ZHANG C.Dense crowd detection algorithm based on Faster R-CNN [J].Journal of Computer Applications,2023,43(1):61-66.
[9] 李哲,于海生,吴贺荣,等.基于新型 RRT 算法与视觉定位的机 械 臂 工 件 抓 取 [J].组 合 机 床 与 自 动 化 加 工 技 术,2022 (12):148-153.LI Z,YU H S,WU H R,et al.Workpiece grasping of manipulator based on improved RRT algorithm and visual positioning [J].Modular Machine Tool & Automatic Manufacturing Technique,2022 (12):148-153.
[10] 伍子嘉,陈航,彭勇,等.动态环境下融合轻量级 YOLOv 5s的视觉 SLAM [J].计算机工程,2022,48(8):187-195,205.WU Z J,CHEN H,PENG Y,et al.Visual SLAM with lightweight YOLOv 5s in dynamic environment [J].Computer Engineering,2022,48(8):187-195,205.
[11] 王泷,纪元霞,吴红刚,等.基于改进 YOLOv 5s的白酒摘酒酒度检测方法 [J].食品与机械,2025,41(7):65-71.WANG L,JI Y X,WU H G,et al.Detection of alcohol content in liquor gathering based on improved YOLOv 5s[J].Food & Machinery,2025,41(7):65-71.
[12] 张海峰,芦新春,冯博,等.基于改进 YOLOv 5s的海水鱼种类识别 [J].食品与机械,2024,40(8):84-92.ZHANG H F,LU X C,FENG Y R.Marine fish species recognition based on improved YOLOv 5s[J].Food & Machinery,2024,40(8):84-92.
[13] 魏 晶 鑫,陈 中 举,许 浩 然.融 合 Fasternet 与YOLOv 5模 型 的鸡蛋外观检测 [J].食品与机械,2024,40(8):105-112,165.WEI J X,CHEN Z J,XU H R.Detection of egg appreance based on Fasternet and YOLOv 5 model [J].Food & Machinery,2024,40(8):105-112,165.
[14] 黄友锐,王照锋,韩涛,等.结合轻量化 YOLOv 5s的动态视觉SLAM 算法 [J].电子测量技术,2024,47(11):59-68.HUANG Y R,WANG Z FE,HAN T,et al.Dynamic visual SLAM algorithm combined with lightweight YOLOv 5s[J].Electronic Measurement Technology,2024,47(11):59-68.
[15] LIU Q,TANG X Y,HUO J.Attitude measurement of ultraclose-range spacecraft based on improved YOLOv 5s and adaptive Hough circle extraction [J].Applied Optics,2024,63(5):1 364-1 376.
[16] 李爽,曹阳,沈琴琴,等.改进 YOLOv 5s的小目标车辆检测算法[J].计算机仿真,2025,42(2):155-161.LI S,CAO Y,SHEN Q Q,et al.Small target vehicle detection algorithm based on improved YOLOv 5s[J].Computer Simulation,2025,42(2):155-161.
[17] 刘锡琳,潘文松,张爱军.基于改进 YOLOv 5s的轮毂气门孔检测算法 [J].电子设计工程,2024,32(19):140-144,149.LIU X L,PAN W S,ZHANG A J.Hub valve hole detection algorithm based on improved YOLOv 5s[J].Electronic Design Engineering,2024,32(19):140-144,149.
[18] 蒋畅江,刘朋,舒鹏.基于改进 YOLOv 5s的动态视觉 SLAM算法 [J].北京航空航天大学学报,2025,51(3):763-771.JIANG C J,LIU P,SHU P.Dynamic visual SLAM algorithm based on improved YOLOv 5s[J].Journal of Beijing University of Aeronautics and Astronautics,2025,51(3):763-771.
