正交约束下的机器人线结构光手眼标定

Hand-eye calibration of robot with linear structured light under orthogonal constraints

为了将传感器局部坐标系下的测量数据转化为机器人坐标系下的全局三维数据,需要确定机器人法兰盘与传感器之间的关系,即手眼标定。为实现机器人与线结构光传感器的现场手眼标定,提出一种正交约束条件下的新方法,该方法标定过程无需人工参与,仅需一块具有空间正交约束关系的平面靶标。标定过程中任意变换机器人位置与姿态,控制激光线与靶标两直角边分别相交,并记录传感器测量坐标值以及对应的机器人姿态数据,再基于空间正交约束建立非线性优化模型,最后采用内点法迭代求得最优解。实验结果表明:该标定方法快速、稳定且精度较高,系统整体测量精度优于0.3 mm,可满足常规测量需求。

In order to transform the measurement data in the local coordinate system of line structured light sensor into global 3D data in the coordinate system of the robot,the relationship between the robot flange and the sensor needs to be determined,which is called the hand-eye calibration.A new algorithm based on spatial orthogonal constraints is proposed for fast hand-eye calibration of line structured light sensor,which requires no human involvement in the calibration process,and only a plane target with a spatially orthogonal constraints relationship.During the calibration process,the laser line is controlled to intersect with both sides of the target,and the intersection coordinates and the corresponding robot positions and postures are recorded.Then,a nonlinear optimization model is built based on the space orthogonal constraints and finally the interior point method is applied to iteratively obtain the optimal solution.The experimental results show that the calibration method is fast,stable and highly accurate,and the overall measurement accuracy of the system is better than 0.3 mm,which can meet the needs of conventional measurement.