Research And Implementation Of Surface Reconstruction And Spraying Trajectory Of Spray Painting Robot Based On 3D Vision

The spraying robot can realize the automatic planning of spraying trajectory,etc.It is widely used in the coating production lines of automobiles and other products at home and abroad.Its application can greatly improve the working environment and improve production efficiency.This paper focuses on the automatic path planning problem of spraying robots.Based on 3D vision technology,three difficult problems of background point removal in target point cloud acquisition,route optimization and time optimization in connection trajectory optimization are studied.The realization of point cloud surface reconstruction is also studied.The main contents of the paper are as follows:First,to solve the problem that the background points of complex surfaces are difficult to remove,a more flexible solution for acquiring point clouds of target components is proposed.Among them,for simple parts,the B-spline surface reconstruction method is used to obtain point clouds;for complex parts,the DBSCAN algorithm is used to classify the point clouds,and the largest class is selected to remove background points.The simulation shows that this scheme simplifies the point cloud acquisition process and has high flexibility,which can fully meet the needs of the spraying robot.Second,based on 3D vision,the surface reconstruction of complex components is realized by using the normal vector estimation algorithm,the Ball pivoting algorithm and the filtering technology.Compared with the traditional surface reconstruction method,a smoother surface can be obtained.In the subsequent surface recognition and slicing processing,the conventional surface recognition is improved,and the surface recognition algorithm is optimized in the case where the normal vector of the patch and the longest and narrow sides of the frame in the direction of the point cloud are approximately parallel,and the segmentation of adjacent patches is reduced.The simulation shows that the improved surface recognition method improves the performance of slice processing at the boundary based on the surface reconstructed in this paper.Thirdly,for the optimal path of patch connection trajectories,an improved genetic algorithm based on extended chromosomes is proposed.The information of endpoint pair selection is considered in the construction of chromosomes.Simulations show that the algorithm can obtain the shortest connection trajectories.Aiming at the time optimization problem of block connection trajectory,based on NURBS Cartesian space planning and B-spline joint space planning,an improved hybrid particle swarm algorithm based on bionic rollback technology is proposed.Simulation shows that the algorithm can obtain the connection trajectory with the shortest time,thus improving the spraying efficiency.Fourth,based on the above research,a spraying trajectory planning scheme based on block strategy and reference boundary registration trajectory processing is realized.In this scheme,the classification of the main components and the small components reduces the complexity of the trajectories generated by slices after surface recognition;the placement error is eliminated by using the registration of the reference boundary trajectory and the actual point cloud.The problem of poor matching between workpiece point cloud and spraying trajectory in actual spraying has been solved.Experiments show that the proposed scheme can obtain more regular patches,improve the efficiency of slicing processing,and can obtain spraying trajectories that match the actual point cloud more closely.