A method for intelligently analyzing truss grab information using three-dimensional layout software

Abstract

The present application relates to a method for intelligently analyzing truss grabbing information by means of three-dimensional layout software in the process of grabbing workpieces by a truss manipulator in a 3D printing automated production line, through layout software to collect image information of each side profile of the workpiece to obtain each planar image data around the workpiece, match the characteristic pattern of the ear end face to confirm the ear end face orientation, placement information and auxiliary information, and further convert into executable data information in the workbox coordinate system of the truss working environment, and provide the control system of the truss manipulator for grabbing, so as to realize automatic identification and grabbing of the grabbing position, can be used with existing truss equipment, improve the intelligence and accuracy of truss grabbing, and improve work efficiency and save labor cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of truss machinery, in particular to a method for intelligently analyzing truss grabbing information by using three-dimensional layout software, and is particularly suitable for sand cleaning and application hanging in the process of intelligent 3D printing casting production line. BACKGROUND

[0002] Casting is the basis of equipment manufacturing industry and is also the basic industry of national economy. From automobiles, machine tools, rail transit, building hardware to aviation, aerospace, national defense and people's daily life, etc., all need casting products. China's casting output is at the world leading level, and there are more than 20,000 casting factories in China, of which nearly 80% are using sand casting, and the trend is rising year by year. Every year, 40 million tons of sand cores need to be cleaned, detected, surface treated, transported and stacked, and the problems of manpower and efficiency are particularly prominent. In the field of robot and automatic industrial production, truss manipulator realizes the complete automation of sand core processing process, and plays an important role in sand core production line feeding, workpiece overturning, workpiece sequence changing and other stations.

[0003] In order to create an ecological development mode of casting industry and promote the development of casting industry to high-end, green, intelligent and service, the manipulator is introduced into the intelligent, automatic and modern intelligent casting forming factory to meet the process flow of large load sand core handling, cleaning and dipping of 3D printing equipment. Combined with actual working conditions, the clamping device of truss manipulator must have good bearing capacity, sufficient grabbing stroke, excellent sealing property and high quality repeated positioning and grabbing precision.

[0004] Generally, the data source of truss manipulator grabbing model position is that, in the layout process, the left and right grabbing positions and placement positions of the marked model are identified by human judgment, and then the software analyzes and outputs the absolute position of grabbing information in the work coordinate system. The truss manipulator completes the model grabbing according to the grabbing information. In this process, each part needs layout process personnel to manually identify whether it is necessary to automatically grab, and manually mark the grabbing information of each model. Not only the work is heavy, but also the error probability is high, which will delay the production efficiency. SUMMARY

[0005] The present application provides a method for intelligently analyzing truss grabbing information by using three-dimensional layout software to realize automatic identification of model grabbing information and calculation output, reduce the work load of process layout personnel and improve data accuracy.

[0006] The purpose of the present application is achieved by a method for intelligently analyzing truss grabbing information by using three-dimensional layout software, comprising the following steps:

[0007] A1: According to the gripper tool of different trusses, a matching three-dimensional model ear contour standard library is established;

[0008] A2: The periphery of the workpiece model is surrounded by the six faces of the axisymmetric bounding box, the periphery contour of each corresponding face of the three-dimensional model is intercepted, and six cross-sectional patterns are obtained;

[0009] A3: By two-dimensional graphic cutting method, the ear contour standard library established in step A1) is compared, the ear information of the three-dimensional model is judged, recognized and marked;

[0010] A4: The auxiliary information and placement information of the three-dimensional model are analyzed;

[0011] A5: The ear information, auxiliary information and placement information of the three-dimensional model in the workbox coordinate are calculated and output to the execution part of the truss robot.

[0012] The automatic recognition and grabbing method of the truss robot of the present application acquires the image data of each plane around the workpiece by collecting the image information of each side contour of the workpiece through the layout software, matches the feature pattern of the ear end face, confirms the ear end face orientation, placement information and auxiliary information, and further converts them into executable data information in the workbox coordinate system of the truss working environment, which is provided to the control system of the truss robot for grabbing, so as to realize the automatic recognition and grabbing of the grabbing position. It can be used with existing truss equipment, improve the intelligence and accuracy of truss grabbing, improve work efficiency and save labor cost.

[0013] Further, in step A1, the three-dimensional model refers to the three-dimensional model of the 3D printed workpiece.

[0014] In order to accurately identify and mark the end face information, in step A3, if the cross-sectional patterns of the two peripheral contours which are axisymmetric and parallel to each other are completely coincided with two cross-sectional patterns in the ear contour pattern of the standard library, the ear end face is marked, and the center point coordinates (Px, Py, Pz) and (Qx, Qy, Qz) of the ear end face are calculated. The ear information includes: the midpoint coordinates (Rx, Ry, Rz) of the connecting line of the two ear end face center points, the distance of the connecting line of the two ear end face center points, the distance of the center point of the ear end face to the ear end face contour along X, Y and Z directions, and the angle a of the connecting line of the ear end face center point to XOZ plane.

[0015] In order to preliminarily determine whether the three-dimensional model has the grabbing condition, in step A3, if there is no cross-sectional pattern completely coincided with the standard library in each group of cross-sectional patterns which are axisymmetric and parallel to each other in the six cross-sectional patterns, the three-dimensional model does not have the automatic grabbing condition of the truss robot.

[0016] To prevent the gripper from colliding with and damaging other parts of the workpiece during the gripping process, in step A4, the auxiliary information includes the distance from the midpoint of the line connecting the center points of the two gripper end faces to the maximum spatial point of the 3D model along the X, Y, and Z directions, and the distance from the midpoint of the line connecting the center points of the two gripper end faces to the minimum spatial point of the 3D model along the X, Y, and Z directions. The maximum spatial point of the 3D model is labeled (Mx, My, Mz), and the minimum spatial point of the 3D model is labeled (Sx, Sy, Sz).

[0017] To facilitate determining the placement orientation of the 3D model during sand removal or impregnation, the placement information includes the placement bottom surface and the auxiliary bottom surface. The placement bottom surface is the side surface with the largest cross-sectional area corresponding to the solid part among the four cross-sectional shapes of the 3D model excluding the end face of the gripper, and is identified as the placement bottom surface. The angle b of the horizontal downward rotation of the placement bottom surface of the 3D model around the line connecting the midpoints of the two gripper points is calculated. The auxiliary bottom surface is the side surface corresponding to the cross-sectional shape with the smallest area among the three cross-sectional shapes of the 3D model excluding the end face of the gripper and the placement bottom surface, and is identified as the auxiliary bottom surface. The angle c of the horizontal downward rotation of the auxiliary bottom surface of the 3D model around the line connecting the midpoints of the two gripper points is calculated.

[0018] To facilitate the conversion of the relevant information of the 3D model obtained by the layout software into parameters in the truss working coordinate system, in step A5, the grab information, placement information and auxiliary information obtained by the layout software in steps A3 and A4 are converted into grab information, placement information and auxiliary information for execution in the working box coordinate system. The offset of the coordinate origin in the layout software relative to the working box coordinate system is (dx, dy, dz).

[0019] Furthermore, the coordinates of the center point of the gripper end face of the workpiece corresponding to the center point of the gripper end face in the workbox coordinate system are: (Px+dx, Py+dy, Pz+dz) and (Qx+dx, Qy+dy, Qz+dz); the coordinates of the maximum point spatial coordinates of the three-dimensional model in the workbox coordinate system are: (Mx+dx, My+dy, Mz+dz), (Sx+dx, Sy+dy, Sz+dz). Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the structure of a truss gripper in one embodiment.

[0021] Figure 2 To and Figure 1 A schematic diagram of two types of grippers in a 3D model for gripper matching.

[0022] Figure 3 This is a schematic diagram of using a bounding box to enclose a 3D model.

[0023] Figure 4 for Figure 3The bounding box in the image captures six cross-sectional views of the 3D model.

[0024] Figure 5 This is a schematic diagram showing the offset of the origin coordinates of the layout software in the workbox coordinate system (top view direction).

[0025] This embodiment uses Figure 1 Using the truss gripper shown as an example, this invention details the method for intelligently analyzing truss gripping information using 3D mapping software. In this embodiment, the 3D mapping software used for 3D printing is IDreamSlices software. Figure 1 The truss shown has two grippers paired on the left and right sides, which are respectively connected to the two mechanical arms of the truss mechanism. The rotation center axes of the two grippers are set coaxially. Through the operation of the truss mechanism, the two grippers can rotate along the rotation center axis and move towards and away from each other along the rotation center axis, so as to adjust the distance and gripping angle of the grippers according to the position and size of the workpiece gripping ears.

[0026] Since the 3D-printed workpiece box is transferred from the 3D printing station to the subsequent station, the workpiece is removed from the workpiece box for sand removal, flipping, moving, or immersion flipping operations. When the worktable at the bottom of the workpiece box is raised to fully expose the workpiece, the workpiece is flipped and transferred by gripping the gripper part of the gantry gripper. In this embodiment, the method of intelligently analyzing the gantry gripping information using 3D mapping software specifically includes the following steps:

[0027] A1: Based on different truss grabber assembly 1, a standard library of matching 3D model grabber lug contours is established in 3D layout software. The grabber assembly 1 in this embodiment is as follows: Figure 1 As shown; with Figure 1 The structure of the gripper ear 2 in the 3D model corresponding to gripper 1 is as follows: Figure 2 The two structures shown in (a) and (b) are used in this embodiment for 3D printing of the 3D model 3 as follows. Figure 3 When the hand-held device grabs the workpiece corresponding to the 3D model 3, it moves from both sides to the center and then moves towards each other to engage and grab the workpiece. After that, it can perform operations such as lifting, flipping or moving.

[0028] Following the previous step, proceed to A2: (e.g.) Figure 3 As shown in the diagram, in 3D modeling software, the six faces of an axisymmetric bounding box are used to enclose the outer perimeter of the workpiece model. The outer contours of corresponding faces of the 3D model are then cropped to obtain six cross-sectional shapes, as shown below. Figure 4 As shown in a, b, c, d, e, f.

[0029] Following the previous step, proceed to A3: In the 3D modeling software, using a 2D graphic cutting method, compare the model with the standard library of claw contours established in step A1) to identify and label the claw information of the 3D model. During identification, if two cross-sectional graphics of each set of axisymmetric and parallel outer contours completely overlap with a certain claw contour graphic in the standard library, they are marked as claw end faces. In this embodiment, the identified claw information is... Figure 4 The cross-sectional shapes of a and b, which are the sides corresponding to the two cross-sectional shapes in the 3D model, are the gripper end faces. The coordinates of the center points of the gripper end faces (Px, Py, Pz) and (Qx, Qy, Qz) are then calculated. Further gripper information includes: the coordinates of the midpoint of the line connecting the center points of the two gripper end faces (Rx, Ry, Rz), the distance between the lines connecting the center points of the two gripper end faces, the distance from the center point of the gripper end face to the gripper end face contour along the X, Y, and Z directions, and the angle α between the line connecting the center points of the gripper end faces and the XOZ plane. In this step, if none of the axisymmetric and parallel cross-sectional shapes in the six cross-sectional shapes completely overlap with the standard library cross-section, then the 3D model does not meet the conditions for automatic gripping by the gantry robot, and the analysis and recognition process ends.

[0030] Proceed to A4: Analyze the auxiliary information and placement information of the three-dimensional model. The auxiliary information includes the distance from the midpoint of the line connecting the center points of the two claw end faces to the maximum spatial point of the three-dimensional model along the X, Y, and Z directions, and the distance from the midpoint of the line connecting the center points of the two claw end faces to the minimum spatial point of the three-dimensional model along the X, Y, and Z directions. The maximum spatial point of the three-dimensional model is labeled (Mx, My, Mz), and the minimum spatial point of the three-dimensional model is labeled (Sx, Sy, Sz). The placement information includes the placement base and the auxiliary base. The side with the largest solid area among the four cross-sectional shapes of the 3D model excluding the claw end face is designated as the placement base. The angle b of the horizontal downward rotation of the placement base with the 3D model about the line connecting the midpoints of the two claws is calculated. The auxiliary base is the side corresponding to the cross-sectional shape with the smallest area among the three cross-sectional shapes of the 3D model excluding the claw end face and the placement base. It is designated as the auxiliary base. The angle c of the horizontal downward rotation of the auxiliary base with the 3D model about the line connecting the midpoints of the two claws is calculated.

[0031] Following the previous step, proceed to step A5. Since the information obtained in the previous steps is data from the 3D model obtained in the 3D layout software, to facilitate adaptation to the working coordinate system of the truss mechanism, and for actual workpiece gripping in production, a conversion of the coordinate system data is required. The gripper information, placement information, and auxiliary information obtained in steps A3 and A4 using the layout software are converted into gripper information, placement information, and auxiliary information for execution in the workbox coordinate system. If the offset of the coordinate origin in the layout software relative to the workbox coordinate system is (dx, d... The coordinates of the center point of the transformed gripper end face in the workbox coordinate system are: (Px+dx, Py+dy, Pz+dz) and (Qx+dx, Qy+dy, Qz+dz); the coordinates of the midpoint of the line connecting the center points are: (Rx+dx, Ry+dy, Rz+dz); the coordinates of the maximum point spatial coordinates of the 3D model in the workbox coordinate system are: (Mx+dx, My+dy, Mz+dz), (Sx+dx, Sy+dy, Sz+dz).

[0032] The final output information, which facilitates the execution of the truss mechanism, is as follows:

[0033] .

[0034] The final output execution information of the truss mechanism in this embodiment is:

[0035] Note: The image file is located at C:\Users\Administrator\AppData\Roaming\Tencent\Users\304857003\TIM\WinTemp\RichOle\MKWP)TW]4903X3REU_0A8~A.png .

[0036] The automatic identification and grasping method of the gantry robot of the present invention acquires image information of the contours of each side of the workpiece through layout software to obtain four planar image data around the workpiece. The feature graphics of the grasping surface of the grasping ear are matched to confirm the orientation, specific position and auxiliary information of the grasping ear surface. This information is then further converted into the work box coordinate system of the gantry working environment and provided to the control system of the gantry robot for grasping. This realizes automatic identification and grasping of the grasping part. It can be used in conjunction with existing gantry equipment to improve the intelligence and accuracy of gantry grasping, increase work efficiency and save labor costs.

Claims

1. A method for intelligently analyzing truss capture information using 3D mapping software, characterized in that, Includes the following steps: A1: Establish a standard library of matching 3D model grab ear contours based on the grab hand assembly of different trusses; A2: Use the six faces of the axisymmetric bounding box to surround the workpiece model, and cut the outer contour of each corresponding face of the 3D model to obtain six cross-sectional graphics. A3: Using a two-dimensional graphic cutting method, compare the three-dimensional model with the standard library of ear-grabbing contours established in step A1). Compare the six cross-sectional graphics of the three-dimensional model. If two cross-sectional graphics of two axially symmetric and parallel outer contours completely overlap with the ear-grabbing contour graphics in the standard library, mark them as ear-grabbing end faces. Calculate the coordinates (Px, Py, Pz) and (Qx, Qy, Qz) of the center point of the ear-grabbing end face. Identify and mark the ear-grabbing information of the three-dimensional model. The ear-grabbing information includes: the coordinates (Rx, Ry, Rz) of the midpoint of the line connecting the center points of the two ear-grabbing end faces, the distance between the lines connecting the center points of the two ear-grabbing end faces, the distance from the center point of the ear-grabbing end face to the ear-grabbing end face contour along the X, Y, and Z directions, and the angle α between the line connecting the center points of the ear-grabbing end faces and the XOZ plane. A4: Analyze the auxiliary information and placement information of the 3D model; the auxiliary information includes the distance from the midpoint of the line connecting the center points of the two gripper end faces to the maximum spatial point of the 3D model along the X, Y, and Z directions, and the distance from the midpoint of the line connecting the center points of the two gripper end faces to the minimum spatial point of the 3D model along the X, Y, and Z directions. The maximum spatial point of the 3D model is labeled (Mx, My, Mz) and the minimum spatial point of the 3D model is labeled (Sx, Sy, Sz). A5: Calculate and output the gripper information, paving information, and placement information of the 3D model in the work box coordinates, and provide them to the execution components of the gantry robot.

2. The method for intelligently analyzing truss capture information using 3D mapping software according to claim 1, characterized in that, In step A1, the three-dimensional model refers to the three-dimensional model of the 3D printed workpiece.

3. The method for intelligently analyzing truss capture information using 3D layout software according to claim 1, characterized in that, In step A3, if among the six cross-sectional figures, there are no two cross-sectional figures in each group of axisymmetric and parallel figures that completely overlap with the standard library, then the 3D model does not meet the conditions for automatic grasping by the gantry robot.

4. The method for intelligently analyzing truss capture information using 3D mapping software according to claim 1, characterized in that, The placement information includes the placement base and the paving base. The placement base is the side with the largest cross-sectional area corresponding to the solid part among the four cross-sectional shapes of the 3D model excluding the claw end face, and is identified as the placement base. The angle b of rotating the 3D model counterclockwise around the line connecting the midpoints of the two claws to make the placement base horizontal and downward is calculated. The paving base is the side with the smallest area among the three cross-sectional shapes of the 3D model excluding the claw end face and the placement base, and is identified as the paving base. The angle c of rotating the 3D model counterclockwise around the line connecting the midpoints of the two claws to make the paving base horizontal and downward is calculated.

5. The method for intelligently analyzing truss capture information using three-dimensional layout software according to claim 4, characterized in that, In step A5, the grab information, placement information, and auxiliary information obtained from steps A3 and A4 using the layout software are converted into grab information, placement information, and auxiliary information for execution in the workbox coordinates. The offset of the coordinate origin in the layout software relative to the workbox coordinates is (dx, dy, dz).

6. The method for intelligently analyzing truss capture information using three-dimensional layout software according to claim 5, characterized in that, The coordinates of the center point of the gripper end face of the workpiece corresponding to the center point of the gripper end face in the workbox coordinate system are: (Px+dx, Py+dy, Pz+dz) and (Qx+dx, Qy+dy, Qz+dz); the coordinates of the maximum point spatial coordinates of the 3D model in the workbox coordinate system are: (Mx+dx, My+dy, Mz+dz), and the coordinates of the minimum point spatial coordinates in the workbox coordinate system are: (Sx+dx, Sy+dy, Sz+dz).

Images