To improve the uniformity of coating thickness and spraying efficiency, new point cloud modeling and slicing algorithm are proposed to deal with free-form surfaces for the spray painting robot in this paper. In the process of point cloud modeling, the edge preservation algorithm is firstly presented to avoid damaging the edge characteristic of the point cloud model. For the spraying gun, the coating deposition model on the free-form surface is determined on the basis of the elliptic doubleβdistribution model. Then, the grid projection algorithm is proposed to obtain grid points between adjacent slices on the free-form surface. Based on this, the analytical solution for calculating the coating thickness at each grid point is obtained. The cross-section contour points are obtained by intercepting the point cloud model with several parallel slices, which is important for the trajectory planning of the spray painting robot. Finally, the uniformity of coating thickness is optimized in terms of the moving speed of the spraying gun and the slice thickness. The simulation and numerical experiment results show that the uniformity of coating thickness and spraying efficiency are improved using the proposed point cloud modeling and slicing algorithm.

Accurate rail location is a crucial part of safety monitoring in railway operational systems. Light detection and ranging can be used to obtain point clouds that contain three-dimensional (3D) information about the railway environment, especially in darkness and poor weather conditions. In this paper, a real-time rail recognition method based on 3D point clouds is proposed to solve the challenges, such as the disorder, uneven density and large volume of the point clouds. A voxel downsampling method is first presented to balance the density of the railway point clouds, and pyramid partitioning is designed to divide the 3D scanning area into voxels with different volumes. A feature encoding module is then developed to find the nearest neighbor points and to aggregate their local geometric features for the central point. Finally, a multiscale neural network is proposed to generate the prediction results for each voxel and the rail location. Experiments are conducted on nine sequences of 3D point cloud railway data. The results show that this method has good performance in detecting straight and curved rails and other complex rail topologies.