首页|异型圆柱贯穿腔体多轴线定位方法研究

异型圆柱贯穿腔体多轴线定位方法研究

扫码查看
异型圆柱贯穿腔体承载件的多轴线定位是其自动化缺陷检测的前提,本文提出一种基于机器视觉的异型圆柱贯穿腔体承载件的多轴线定位方法,从根源上解决了人工定位测量速度慢、精度低、自动化定位不能批量处理多个特征等问题.首先标定相机内参和机械臂手眼关系,接着采集异型承载件图像并检测端面椭圆,然后识别端面ArUco靶标并估计相机位姿,最后将承载件端面圆心坐标和法向量转换到机器人基座坐标系下,求解轴线数学表征.实验结果表明,该方法可以快速定位承载件轴线特征,异型承载件端面圆心定位精度最高可达1.63mm,端面ArUco靶标姿态估计精度最高可达0.35°,效率远超于人工测量,并可以处理具有多个圆柱端面的异型工件,具有一定的工业应用价值.
Research on Multi Axis Positioning Method of Special Shaped Cylinder Through Cavity
The multi-axis positioning of the special-shaped cylindrical penetrating cavity bearing parts is the prerequisite for its automatic defect detection.It proposes a multi-axis positioning method based on machine vision for a special-shaped cylindrical penetrating cavity bearing parts,which solves the problems of slow manual positioning and measurement,low accuracy,and in-ability to batch process multiple features in automated positioning.Firstly,calibrate the camera internal parameters and the hand-eye relationship of the manipulator,then collect the image of the special-shaped carrier and detect the end ellipse,then identify the ArUco target and estimate the camera pose.Finally,the center coordinates and normal vectors of the end face of the carrier are converted to the robot base coordinate system,and the mathematical representation of the axis is solved.The experimen-tal results show that this method can quickly locate the axis characteristics of the bearing part.The positioning accuracy of the end face center of the special-shaped bearing part can reach up to 1.63mm,and the end face ArUco target attitude estimation accura-cy can reach up to 0.35°,the efficiency is much higher than manual measurement,the method can handle special-shaped work-pieces with multiple cylindrical end faces,which has a certain industrial application value.

Monocular VisionIndustrial RobotAruco TargetTarget DetectionFeature Location

兰孝文、赵波、王少锋、黄顺舟

展开 >

内蒙古科技大学信息工程学院,内蒙古 包头 014010

内蒙古科技大学机械工程学院,内蒙古 包头 014010

上海航天设备制造总厂有限公司,上海 200000

单目视觉 工业机器人 ArUco靶标 目标检测 特征定位

内蒙古自治区科技计划项目

2020GG0160

2024

机械设计与制造
辽宁省机械研究院

机械设计与制造

CSTPCD北大核心
影响因子:0.511
ISSN:1001-3997
年,卷(期):2024.403(9)
  • 7