激光测量标定机器人坐标系位姿变换的正交化解算

Laser measurement based evaluation for orthogonal transformation calibration of robot pose

针对大型装备智能制造中的机器人在线位姿激光跟踪测量与实时引导需求,提出了一种机器人坐标系与激光测量坐标系标定转换和解算方法。设计了基于距离原则的机器人末端光学工具中心点TCP(Tool Center Point)位置标定算法。通过运用空间点坐标重心化配置算法和基于罗德里格矩阵变换的最小二乘优化算法解算出了具有单位正交性的位姿变换旋转矩阵。进行了机器人坐标系位姿变换激光测量标定和优化对比实验,旋转矩阵初值和正交优化值进行点坐标转换后的综合RMSE分别为0.579 0mm和0.501 5mm。结果表明该方法能够有效改进姿态旋转矩阵正交性,并提高位姿变换解算精度。

Aiming at the requirements of online laser tracking measuring and real-time steering for robot position and orientation in the field of large-scale high-end equipment manufacturing,a type of transformation calibration method between the robot base and the laser measurement coordinate system was presented in this paper.The algorithm for positional evaluation of optical TCPs(tool center point)was devised based on the criterion of distance,and the computation of a coordinate system center of gravity configuration based on the theory of rigid body kinematics was applied.Subsequently,the least squares optimization estimation for rotation transformation of robot pose was achieved through a Lodrigues matrix transformation.An orthonormal matrix can be reached,which was the pose transformation for the base frame of the robot with respect to the laser measurement frame.The comparison experiment between initial calibration and optimal estimation of pose transformation was performed.Results verified the feasibility and effectiveness of the evaluation process,which significantly improved the unitary orthogonality of the rotation matrix,and the position transformation RMSE of a robot corresponding to the laser measurement frame,which can reach 0.579 0 and 0.501 5 mm,respectively.