中文摘要：

基于棋盘格的2维激光雷达和摄像机标定方法中,Francisco方法提出一个最小解决方案,通过3组＂线-面＂空间约束分别求解旋转矩阵和平移向量。在对偶空间中,Francisco方法巧妙地将旋转矩阵的求解转换为对P3P问题和三点匹配估计运动问题的求解。但是Francisco方法需要首先穷尽计算所有3个标定板组合,然后根据误差函数评价出一个最优解析解,因此计算复杂度较高。为此首先根据线-面约束方程组的系数矩阵的逆阵的二范数来选取一组抗扰动能力最强的组合,并依据这个组合计算解析解,其次改进了评价最优解析解的误差函数。与当前最先进的标定方法比较,实验结果定性和定量地显示该方法更快速更精确。

英文摘要：

In all the calibration methods of 2D laser radar and camera based on the checkerboard, the Francisco method proposes a minimal solution, which solves the relative rotation matrix and translation vector with the constraints of three planes and correspond- ing coplanar lines. In the dual space, the Francisco method transforms the solution problem of the rotation matrix into the solution problems of P3P and three point matching motion estimation, cleverly. However, the Francisco method has to calculate all the combinations of three checkerboards exhaustively, and then evaluate an optimum closed-form solution according to the error func- tion; therefore the computational complexity of the method is very high. To deal with this problem, firstly, we select a group of three checkerboards that has the strongest anti-disturbance capability based on the 2-norm of the inverse coefficient matrix of the lin- ear equations corresponding to the constraints of the coplanar lines and planes. Secondly, we compute a group of closed-form solu- tion of the calibration parameter according to the three checkerboards. Lastly, we improve the error function for evaluating the qual- ity of the closed-form solution. Compared with the most advanced calibration methods, experiment results show that the proposed method is faster and more accurate.