基于多点结构光扫描的工件定位方法  

Workpiece Positioning Method Based on Multi-Point Structured Light Scanning

作　　者：程丽[1] 徐超海 教经纶 朱思俊[2,3] CHENG Li;XU Chaohai;JIAO Jinglun;ZHU Sijun(School of Mechanical Engineering,Shenyang University,Shenyang 110044,China;Shenyang Institute of Automation,Chinese Academy of Sciences,Shenyang 110016,China)

机构地区：[1]沈阳大学机械工程学院,沈阳110044 [2]中国科学院沈阳自动化研究所,沈阳110016 [3]中国科学院机器人与智能制造创新研究院,沈阳110169

出　　处：《组合机床与自动化加工技术》2024年第9期178-182,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：工信部渠道项目(MJZ4-4N22)。

摘　　要：在自动化生产线工业机器人抓取工件作业中,工件定位方法直接影响作业质量、效率及系统成本。传统基于线结构光的工件定位方式通常存在专用性强、成本高等特点,难以在劳动密集型产业中推广应用。针对这一问题,利用KD-tree ICP算法,提出一种基于多点结构光扫描的工业机器人抓取工件定位方法。首先,利用模型点云和多点结构光原始点云进行粗配准;其次,提取待定位工件点云特征点,并拟合出工件边界;最后,利用模型边界点云和拟合出工件边界点云进行精配准。试验结果表明,该方法具有较高的定位精度及识别效率,满足工件抓取要求。In the process of grasping workpieces by industrial robots in automatic production lines,the workpiece positioning method directly affects the operation quality,efficiency and system cost.The traditional workpiece positioning method based on line structured light usually has the characteristics of strong specificity and high cost,which is difficult to be popularized and applied in labor-intensive industries.For solving this problem,an industrial robot grasping workpiece positioning method based on multi-point structured light scanning is proposed by using KD-tree ICP algorithm.Firstly,the model point cloud and the multi-point structured light original point cloud are used for rough registration.Secondly,the point cloud feature points of the workpiece to be located are extracted,and the workpiece boundary is fitted.Finally,the model boundary point cloud and the fitted workpiece boundary point cloud are used for fine registration.The experimental results show that the method has high positioning accuracy and recognition efficiency,and meets the requirements of workpiece grasping.

关 键 词：多点结构光 视觉定位 点云配准 工业机器人 

分 类 号：TH164[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]