基于多线激光雷达的圆柱面检测

Cylindrical Surface Detection Based on Multi-line LIDAR

针对在水面场景中使用多线激光雷达检测圆柱面时,出现点云稀疏、局部特征不明显的问题,提出了基于多线激光雷达的圆柱面检测方法。该方法通过利用多线激光雷达扫描圆柱面时形成的对称椭圆弧点阵特征,建立不等式提取属于圆柱面的点云,并且利用改进的椭圆拟合方法对经过拟合预处理的圆柱面点阵断面坐标进行二维平面椭圆拟合,然后利用拟合结果计算出圆柱面的参数。仿真与实验验证结果表明,对于在需要使用多线激光雷达检测圆柱面的场景中,该方法具有有效性与正确性。

Aiming at the problem of sparse point clouds and inapparent local features,the multi-line LiDAR was used to detect cylindrical surface in the water surface scenes,a detection method for cylindrical surface was proposed.In this method,the feature with a symmetrical elliptical on the cylindrical surface formed by the single-line laser beam was used to establish the inequality equation to detect the lattice on the cylindrical surface;and the two-dimensional plane ellipse was fitted on the lattice cross-section coordinated belonging to the cylindrical surface after fitting preprocessing by the improved ellipse fitting method;then the parameters of the cylindrical surface were calculated using the fitting results.The simulation and experimental results showed that the method was effective and correct for scenes where multi-line LiDAR was required to detect cylindrical surfaces.