摘要
为了提高五轴机床转轴位姿标定的效率,提出了一种基于机器视觉求解双转台五轴机床转轴位姿(转轴向量和轴上一点)的方法。首先通过工业CCD相机拍摄不同转轴下的标定板照片,经过图像处理技术得到相机内外参;然后建立机床几何误差模型得到坐标系间的空间转换关系,从而计算得到转轴位姿在内的机床参数的初始解;最后通过levenberg-marquardt(L-M)算法得出最优的机床参数解值,随后,通过加工点跟随实验验证转轴位姿的精度。实验结果表明,所提出的求解方法满足实际加工精度要求,并且测量方法简单、快速。通过采集不同的标定板采图方式和L-M迭代初始值的仿真实验,验证了此方案具有良好的鲁棒性。
In order to improve the efficiency of axis pose calibration of five axis machine tool,a method based on machine vision to solve the axis pose(axis vector and a point on the axis)of five axis machine tool with double turntable is proposed.Firstly,the calibration plate photos under different rotating axes are taken by industrial CCD camera,and the internal and external parameters of the camera are obtained by image processing technology;Then the geometric error model of the machine tool is established to obtain the spatial transformation relationship between coordinate systems,so as to calculate the initial solution of the machine tool parameters including the position and attitude of the rotating shaft;Finally,the optimal solution of machine tool parameters is obtained by levenberg marquardt(L-M)algorithm.Then,the accuracy of the rotation axis pose is verified by the machining point following experiment.The experimental results show that the proposed solution method meets the requirements of the actual machining accuracy,and the measurement method is simple and fast.Through the simulation experiments of collecting different calibration plate mining methods and the initial value of L-M iteration,it is verified that this scheme has good robustness.
作者
王瑞
瞿佳伟
张春雷
WANG Rui;QU Jia-wei;ZHANG Chun-lei(School of Mechanical Engineering,Sichuan University,Chengdu 610065,China;Chengdu Leetro Automation Co.,Ltd.,Chengdu 510107,China)
出处
《组合机床与自动化加工技术》
北大核心
2022年第5期36-40,共5页
Modular Machine Tool & Automatic Manufacturing Technique