Research On Shoe Mold Spraying Trajectory Generation Method Based On Point Cloud Fusion

The rising labor cost has been a difficult problem for labor-intensive industries.In order to reduce labor cost,more and more enterprises have introduced automation technology,and the footwear industry,as a traditional labor-intensive industry,has gradually started industrial upgrading.This paper takes the actual production demand as the starting point,takes the shoe mold as the research object,proposes a point cloud fusion method using multiple line structured light sensors to reconstruct the surface point cloud of the shoe mold,extracts the spray glue track line of the shoe mold by processing the 3D point cloud data,and conducts experimental verification,which mainly includes the following aspects:(1)Point cloud fusion based on multiple line laser sensors.In this paper,a point cloud fusion platform containing four sensors is designed according to the actual demand,and the algorithms of tilt correction and height correction are proposed.The four measurement coordinate systems are unified by using calibration blocks to realize the line fusion of point clouds,and then the y-axis coordinates of the contours are obtained by encoder counting.For the overlap phenomenon caused by the system error in the point cloud stitching process,this paper adopts equal interval processing for frame point clouds to eliminate the redundant point clouds generated in the fusion process,and eliminates the noise by using both statistical filtering and median filtering to improve the point cloud reconstruction accuracy.The reconstruction accuracy is within 0.2mm by verifying the reconstruction of the block point cloud.(2)Extraction of shoe mold spraying trajectory.For the spraying trajectory of the inner any surface,the scaled projection point cloud is used to extract the scaled 2D point cloud contour boundary using Alpha Shapes algorithm,and search the nearest neighbor points of the contour point set in the original projection point cloud,and then project the nearest neighbor points back to the original point cloud to get the spraying trajectory of the inner any surface.of the spray glue trajectory.For the point cloud data of the outer wall of the shoe mold,the point cloud slicing is used to obtain its spray glue trajectory,and finally the normal of the spray glue pre-point position is calculated by using Principal components analysis(PCA).(3)Point cloud based shoe mold spraying system design.Designed and developed a set of point cloud acquisition software system,the software can realize the line laser sensor measurement coordinate system correction and multi-sensor point cloud fusion,in order to verify the effect,the triangle iron as the calibrator of the robot arm for the eye outside the hand hand-eye calibration,and use the robot arm to go spray glue pre-spotting,after verification,the accuracy within 0.3mm,to meet the actual production needs.