摘要
构建了由线结构光测头和机器人组成的视觉测量系统.利用齐次坐标变换的方法建立了测量系统的数学模型.针对数学模型中的手眼关系,使用半径已知的球体,采用定点变位姿的方法,进行粗略的估计.考虑到机器人的绝对精度较低,将定点信息和尺度信息作为优化条件,对计算机器人的运动学参数和手眼关系做进一步寻优.实验表明文中的方法能够有效提高测量机器人的重复精度.
A platform composed of a line-structured-light vision scanner and an industrial robot is established for 3D surface digitalization.A mathematical model for the measuring system is constructed by using the homogenous coordinate transform principle and a solution to hand-eye matrix is proposed through measuring a fixed point at different robot hand postures.Considering the absolute accuracy of the robot,a scheme involving scale information and fixed point information is proposed to estimate the robot kinematic parameter errors and hand-eye errors.Test results of real experiments on an industrial robot are reported and analyzed.
出处
《控制理论与应用》
EI
CAS
CSCD
北大核心
2010年第5期663-667,共5页
Control Theory & Applications
基金
辽宁省科技厅资助项目(2007219003)
关键词
结构光视觉
运动学标定
手眼标定
structured-light vision
robot calibration
hand-eye calibration