Research On Key Technologies Of Cylindrical Casting Grinding Robot Assisted By Vision

As one of the precision casting technologies,investment casting is often used to produce parts with high precision and complex structure.Due to the characteristics of the process,the casting and the mold head need to be connected through the casting sprue,and after the sprue removal process,there will still be some sprue residue on the surface of the workpiece.The traditional way to remove the residual marks is to grind the residual marks by hand with specific grinding equipment,but the grinding processing environment is extremely harsh,and manual grinding has the disadvantages of low production efficiency and inconsistent processing quality.In order to solve the above problems,according to the characteristics of the operating site environment and mechanical performance indexes,a vision-assisted grinding and cleaning solution for the residual marks of columnar castings is designed with industrial robots as the main body and vision detection technology as the auxiliary,which improves the automation of the investment casting production line and guarantees the casting processing quality.The specific research contents are as follows.Firstly,according to the characteristics of the residual marks on the surface of the cylindrical casting,the sanding requirements for the residual marks are determined.The influence of each grinding influencing factor on the grinding quality was analyzed,and the material characteristics of the sandpaper abrasives were determined.The grinding force characteristics of the grinding device carried by the robot arm during the grinding process were studied,and the extreme values of each grinding force component and the grinding power were solved as the basis for determining the model of the robot arm and the grinding device.Secondly,a set of fixture system for cylindrical castings is designed with the characteristics of the branch parts of cylindrical castings,which can be used in a certain size range.According to the force characteristics of the workpiece in the clamping process,a specific action sequence is arranged for each fixture,and the corresponding hydraulic system circuit is designed to realize the clamping or relaxing action of each fixture in sequence.PLC is used to program the fixture system,and the feasibility of the fixture system is verified through experiments.The fixture platform with the fixture and hydraulic device was designed and analyzed statically with ANSYS software.In order to realize the flexible grinding of the robotic arm,the end-effector of the robotic arm was designed in combination with the floating grinding spindle,and the modal analysis was performed in ANSYS software.The D-H method was used to establish the robotic arm spatial motion coordinate system and perform forward and inverse kinematic analysis,and the kinematic verification was performed in MATLAB and the robotic arm spatial cloud point diagram was obtained.The end trajectory planning of the robotic arm is carried out by using quintuple polynomial spline interpolation,and the angle curves of each joint are obtained.A robotic arm dynamics model was developed and the model was verified and analyzed.Finally,in order to enable the industrial camera to capture images of the workpiece from multiple angles,the vision acquisition module was mounted on another small robotic arm to determine a vision inspection-assisted residual mark grinding process under dual robotic arms,and the The robotic arm trajectory planning was verified in the ROS system.In order to realize the defect recognition detection and workpiece model matching function,the processing methods of image filtering and enhancement were studied,the Canny algorithm was improved to realize the contour detection of workpiece,and the Hu moment algorithm was used to realize the workpiece feature detection and model matching.Finally,the vision inspection experiment platform is built to detect the defects such as residual marks and flying edges of workpieces,and it is proved that the vision inspection system with the robotic arm meets the design requirements.