基于共面靶标的线结构光机器人视觉标定方法  

Visual Calibration Method of Linear Structured Light Robot Based on Coplanar Target

作　　者：乔赫廷[1] 张然 王黠 QIAO Heting;ZHANG Ran;WANG Xia(School of Mechanical Engineering,Shenyang University of Technology,Shenyang 110870,China)

机构地区：[1]沈阳工业大学机械工程学院,沈阳110870

出　　处：《机械工程师》2024年第3期21-25,共5页Mechanical Engineer

摘　　要：建立了线结构光视觉系统的数学模型,提出一种基于共面靶标的线结构光视觉系统标定方法,包括相机的内、外参数的标定,线结构光平面在相机坐标系下的标定方法,提出一种利用微分的思想映射到世界坐标系下逼近标定特征点纵坐标的实际值,使用相机标定时的外参将多个特征点转换至相机坐标系下,组成光平面超定方程组,使用最小二乘法估计光平面方程参数。为验证方法的准确性和实用性,实现以CMOS工业相机、线激光发射器组成的视觉系统,通过对棋盘格标定板测量,进行实验。实验结果表明,该标定方法测量得出的相对误差为0.14%,能够满足焊接机器人的精度要求。This paper establishes a mathematical model of the linear structured light vision system,and proposes a calibration method for the linear structured light vision system based on coplanar targets,including the calibration of the internal and external parameters of the camera,and the calibration method of the linear structured light plane in the camera coordinate system.A method using differential idea to approximate the actual value of the ordinate of the calibrated feature points in the world coordinate system is proposed.Using the external reference during camera calibration,multiple feature points are transformed into the camera coordinate system to form the optical plane overdetermined equations,and the parameters of the optical plane equations are estimated by the least square method.In order to verify the accuracy and practicability of the method,the vision system composed of CMOS industrial camera and line laser transmitter is realized.The experiment is carried out by measuring the chessboard calibration board.The experimental results show that the relative error measured by this calibration method is 0.14%,which can meet the accuracy requirements of welding robot.

关 键 词：共面靶标 焊接机器人 视觉标定 线结构光标定 

分 类 号：TP23[自动化与计算机技术—检测技术与自动化装置]