摘要
为解决机器人因运动学参数误差而导致的绝对定位精度损失,提出一种基于双结构光视觉的关节机器人运动学标定方法,具体研究了双结构光光源配合双目视觉在求取光源发射装置空间位置中的应用、推导了出光源装置位置误差与连杆参数误差的误差方程。通过采集机器人若干末端的机器人坐标系位姿与世界坐标系位姿数据,再基于最小二乘理论求取机器人坐标系与世界坐标系的单应性矩阵,将之后采集的点通过单应性矩阵转换到机器人坐标系下的位姿。再通过误差方程来获取连杆参数误差,并对其真实参数进行修正,从而提高机器人定位精度。最后讨论了该方法在SK-10R关节机器人上进行标定的实验结果,结果表明该方法能实现机器人空间位姿的标定,在低成本的条件下大幅提高机器人的绝对定位精度。
In order to solve the absolute positioning accuracy loss caused by the kinematic parameters error of the robot,calibration method for robot kinematics parameters based on structured light and vision has been proposed in this paper,which mainly did research on the application of using double-structured light source and double vision camera to solve the position of light source in space,and we have also inferred the error equation of light source position’s error and manipulator’s parameters error.Firstly,sample several the data of position and orientation from robotics coordinate system and world coordinate system in a articulated robot,and then solve the Homography matrix between robotics coordinate system and world coordinate system via the Least Squares Theory,and finally we revised the actual parameters through Homography,to improve robotics positioning precision.The experimental results are discussed using this method in the SK-10R robot,which shows it can achieve the calibration of robot’s position and orientation in space.This method can improve the absolute precision for a robot in a low cost.
作者
吴孟桦
胡晓兵
赵清祥
向海
WU Meng-hua;HU Xiao-bing;ZHAO Qing-xiang;XIANG Hai(School of mechanical Engineering,Sichuan University,Chengdu 610065,China)
出处
《组合机床与自动化加工技术》
北大核心
2020年第6期37-40,46,共5页
Modular Machine Tool & Automatic Manufacturing Technique
基金
2017年智能制造综合标准化与新模式应用(2017ZZ018)
四川省支撑计划(2017GZ0146、2018GZ0125)。
关键词
机器人
标定
结构光
视觉
robot
calibration
structured light
vision
