Local Feature Recognition And Trajectory Planning Of Spraying Robot For Seating Furniture

The improvement of industrial robot’s intelligence in the field of production greatly expands the function of robot.Intelligent industrial robot can not only liberate people from the complex and harsh working environment,but also improve the production efficiency and product quality,which has successfully penetrated into all aspects of production and manufacturing.As an important part of industrial robot,spraying robot has been gradually applied in furniture spraying industry.Because there are many kinds of seat furniture and complex space structure,it is difficult to plan spraying independently.Manual teaching method is still used to use spraying robot,which will lead to low work efficiency and the quality of spraying is determined by the operation skills of teaching staff.In order to improve the intelligent level of spraying robot trajectory planning,this paper studies the automatic trajectory planning method of spraying robot based on feature recognition by using the image information of visual sensor.Firstly,the point cloud measurement scheme,point cloud processing and recognition algorithm of seat furniture are determined.The measurement principle and hand-eye calibration method of line laser contour sensor are studied,and the accurate3 D point cloud is obtained by scanning.Through and statistical filtering methods are used to remove irrelevant data and outliers,and a downsampling algorithm without losing edge feature points is proposed to simplify the point cloud.In order to solve the problem of poor effect of traditional point cloud registration and stitching methods in the case of low overlap rate of point clouds,combined with the measurement scheme,the cylinder feature recognition algorithm based on RANSAC is studied,and the multi view point cloud stitching of seat furniture is completed by using the algorithm,and the complete point cloud model of seat furniture is obtained.Secondly,the coating deposition models of air spraying under various working conditions are established,such as single spray point on plane,single spray point on free-form surface,single straight spray path and adjacent straight spray path.And the relationship between the coating thickness distribution and the moving speed of spray gun is studied.The relationship between the coating thickness distribution and the distance between adjacent paths is studied,too.Taking the surface spraying uniformity and spraying efficiency as the optimization objectives,the weighted evaluation function is designed to optimize the parameters of the distance between adjacent paths,and the optimized spraying parameters are obtained.In addition,the numerical simulation model of coating thickness of complex seat furniture workpiece is established.Then,the path planning methods of surface spraying and edge repainting are given.Aiming at surface spraying,an automatic segmentation method of spraying area based on the void ratio of bounding box block and the envelope number of point cloud is proposed.The method of combining point cloud slicing algorithm and flexible midline extraction algorithm is used to generate surface spraying path points,and the workpiece feature direction recognition algorithm is used to determine the spray gun attitude.Aiming at edge repainting,combined with edge extraction and sorting algorithm,a fast planning algorithm of spraying path is proposed,which can ensure 45 degree spray gun process attitude angle at any time along the edge of seat furniture repainting.Finally,the path planning experimental platform of spraying robot is established,and the whole process experiments are carried out on all spraying surfaces and spraying edges of seat furniture.The experiments prove that the proposed path planning method of surface spraying and edge patching can accurately and effectively complete the task of seat furniture spraying path planning,And the uniformity of coating thickness distribution and planning time can meet the needs of customers.