基于边缘检测算法的运动建模与目标点定位分析  

Kinematic modeling and target point localization analysis based on edge detection algorithm

作　　者：刘晨明 杨旭东 魏宇豪 李金昌 吴志东[1,2,3] LIU Chen-ming;YANG Xu-dong;WEI Yu-hao;LI Jin-chang;WU Zhi-dong(School of Mechanical and Electronic Engineering,Qiqihar University,Heilongjiang Qiqihar 161006,China;The Engineering Technology Research Center for Precision Manufacturing Equipment and Industrial Perception of Heilongjiang Province,Heilongjiang Qiqihar 161006,China;Heilongjiang Province Collaborative Innovation Center for Intelligent Manufacturing Equipment Industrialization,Heilongjiang Qiqihar 161006,China)

机构地区：[1]齐齐哈尔大学机电工程学院,黑龙江齐齐哈尔161006 [2]黑龙江省精密制造装备及工业感知工程技术研究中心,黑龙江齐齐哈尔161006 [3]黑龙江省智能制造装备产业化协同创新中心,黑龙江齐齐哈尔161006

出　　处：《齐齐哈尔大学学报(自然科学版)》2025年第3期19-25,共7页Journal of Qiqihar University(Natural Science Edition)

基　　金：教育部产学合作协同育人项目(220602817220926);黑龙江省“双一流”学科协同创新成果项目(LJGXCG2024-P25);黑龙江省高等教育教学改革研究一般研究项目(SJGY20220410);齐齐哈尔大学教育科学研究项目(GJQTYB202212)。

摘　　要：针对机械臂自动抓取过程中抓取点不确定问题,提出一种基于图像识别与物块面心坐标计算相结合的抓取方法。利用边缘检测算法确定物块顶面中心所在区域,获取目标物块顶面中心的二维坐标。在Open MV摄像机下方放置红外测距传感器,测量Open MV摄像机与目标物块顶面之间的距离,从而获得物块顶面面心的三维坐标。实验结果表明,基于物块面心坐标准确抓取成功率比随机抓取成功率高20%,将计算得到的待抓取坐标与实际抓取坐标进行对比,x轴方向平均相对误差为2.9%,z轴方向平均相对误差为3.1%。该方法可有效解决机械臂自动抓取过程中抓取点不确定问题,确保抓取的精度和稳定性。Aiming at the problem of uncertain grasping points in the process of robotic arm automatic grasping,this study proposes a grasping method based on the combination of image recognition and object block face-centered coordinate calculation.The edge detection algorithm is used to determine the area where the center of the top surface of the object block is located,and the two-dimensional coordinates of the center of the top surface of the target object block are obtained.An infrared ranging sensor is placed under the Open MV camera to measure the distance between the Open MV camera and the top surface of the target object,so as to obtain the three-dimensional coordinates of the center of the top surface of the object.The experimental results showed that the accurate grasping success rate based on the face-centered coordinates of the object block is 20%higher than the random grasping success rate.The calculated coordinates to be grasped are compared with the actual grasping coordinates.The average relative error in the x-axis direction is 2.9%,and the average relative error in the z-axis direction is 3.1%.This method can effectively solve the problem of uncertain grasping points in the automatic grasping process of the manipulator and ensure the accuracy and stability of the grasping.

关 键 词：图像识别 机械臂抓取 面心坐标 边缘检测算法 逆运动学 

分 类 号：TP241.3[自动化与计算机技术—检测技术与自动化装置]