基于面结构光的T型焊接接头成形角度测量

针对钛合金带筋壁板T型焊接接头成形角度的测量需求，基于面结构光三维扫描仪设计了三维点云测量系统，以获取T型焊接接头区域三维点云。首先，对三维点云进行预处理，利用统计滤波去除离群点，通过直通滤波剪除场景点云和冗余点云，再用体素化栅格法简化点云；然后，使用区域增长算法分割出T型焊接接头的筋板点云和壁板点云；最后，基于随机抽样一致性（RANSAC）算法拟合出分割后的筋板和壁板平面点云，得到平面法向量，计算求得T型焊接接头的成形角度。将焊接前的测量角度反馈给装配机器人，用来修正筋板的装配精度；焊接后的测量角度则用来评判焊接成形质量。计算结果表明，运用所提的点云数据处理方法能快速精确地计算出T型焊接接头的成形角度。

Measurement of the forming angle of T-joints based on surface structured light

In response to the need to measure the forming angle of the T-joints of a titanium alloy ribbed wall plate, a three-dimensional point cloud measurement system based on a surface structured light three-dimensional scanner has been designed in order to provide the three-dimensional point cloud of the T-joint area. Firstly, the three-dimensional point cloud is preprocessed, the outliers are removed by statistical filtering, the field cloud and redundant point cloud are cut off by pass-through filtering, and the point cloud is simplified by means of the voxel grid method. Then, the stiffened plate point cloud and the wall plate point cloud of the T-joints are segmented by the regional growth algorithm. Finally, based on the random sample consensus (RANSAC) algorithm, the plane point cloud of the segmented stiffened plate and wall plate was fitted, the plane normal vector was obtained, and the forming angle of the T-shape welded joint was calculated. The measured angle before welding is fed back to the assembly robot, which is used to correct the assembly precision of the reinforcement plate. The measured angle after welding is used to judge the quality of the resulting weld. The calculation results show that the forming angle of T-joints can be calculated quickly and accurately using this point cloud data processing method.