摘要
提出了一种基于投影矩阵M的机器人手眼标定方法,即将AX=XB转换为MY=M′YB求解形式,避免了矩阵分解.同时,采用矩阵直积运算和矩阵特征向量运算相结合的方法,将传统的非线性问题转换成线性问题,构成了可利用最小二乘法求解运算的解析形式,提高了机器人手眼标定的精度和稳定性.通过实验验证了该方法的可行性.
This paper presents a new approach for robot hand-eye calibration based on the projection matrix M that yields a homogeneous transform equation MY=MYB. Meanwhile, the matrix direct product and matrix characteristic vector methods are utilized in order that the traditional non-linear calculation can be transformed into linear problems, which generate an analytical form that can be solved by least square methods. Some experiments are given to show the feasibility of the method.
出处
《机器人》
EI
CSCD
北大核心
2005年第4期301-305,共5页
Robot
关键词
机器人手眼标定
投影矩阵M
矩阵直积运算
robot hand-eye calibration
projection matrix M
matrix direct-product operation
