一种高精度的机器人手眼标定算法  

An Algorithm for Robot Hand Eye Calibration with High Accuracy

作　　者：胡为[1] 刘冲 傅莉[1] 陈新禹 HU Wei;LIU Chong;FU Li;CHEN Xin-yu(School of Aerospace Engineering,Shenyang Aerospace University,Shenyang 110136,China;School of Automation,Shenyang Aerospace University,Shenyang 110136,China)

机构地区：[1]沈阳航空航天大学航空航天工程学部,沈阳110136 [2]沈阳航空航天大学自动化学院,沈阳110136

出　　处：《火力与指挥控制》2018年第9期19-24,共6页Fire Control & Command Control

基　　金：国家自然科学基金(61074090);中航创新基金(cxy2013SH16);辽宁省自然科学基金(2015020061);辽宁省自然科学基金(联合基金)资助项目(2015020069)

摘　　要：提出了一种基于矩阵直积的机器人手眼标定改进算法。应用矩阵直积、Moore-penrose逆以及最小二乘法对机器人手眼关系方程CX=XD进行线性化处理,并运用F范数分析了标定结果的测量精度。该方法与传统两步法和一般性矩阵直积的手眼标定算法相比,能够克服由于误差传递引起的精度下降问题。结果表明:基于矩阵直积改进算法的标定精度高于另外两种现有算法,并且具有强鲁棒性的特点。An improved algorithm for robot hand-eye calibration based on kronecker product is proposed.The linearity of the hand-eye relation equation CX=XD is solved by using kronecker product,Moore-penrose inverse and least squares method,and the measurement accuracy of the calibration result is analyzed by F norm.This method can overcome the problem of precision degradation due to error propagation compared with traditional two-step method and general kronecker product hand-eye calibration algorithm.The results show that the calibration accuracy of the improved algorithm based on kronecker product is higher than that of the other two algorithms,and it has strong robustness.

关 键 词：手眼标定 矩阵直积 传统两步法 最小二乘法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] TJ57[自动化与计算机技术—计算机科学与技术]