Cutting line positioning method, cutting method and intelligent cutting device

Abstract

The application provides a cutting-off line positioning method, a cutting method and an intelligent cutting device, and relates to the technical field of intelligent cutting and sorting of steel plates. The cutting-off line positioning method comprises the following steps: obtaining a plurality of theoretical cutting-off line position data based on steel plate theoretical model data containing a plurality of workpiece contour cutting line models; performing overall visual detection on the steel plate to obtain actual edge contour data of the steel plate; obtaining 3D position data of each corner of the steel plate based on the actual edge contour data of the steel plate; matching the steel plate theoretical model data and the 3D position data of each corner of the steel plate to obtain an actual posture of the steel plate; performing local accurate visual detection on a plurality of preset positions of a plurality of corresponding actual cutting-off lines one by one based on the actual posture of the steel plate and the plurality of theoretical cutting-off line position data; and determining the positions of the plurality of actual cutting-off lines based on the results of the plurality of local accurate visual detections.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent cutting and sorting of steel plates, in particular to a cutting-off line positioning method, a cutting method and an intelligent cutting device. BACKGROUND

[0002] In modern industrial manufacturing, welding cutting is widely used in various industries, especially in the field of plate cutting. Various engineering machinery and other equipment are generally produced by cutting plate according to cutting lines and then processed by subsequent processes to produce specific products.

[0003] Under normal circumstances, a plurality of different workpiece shapes are first cut on the steel plate, and then the cut workpieces are removed from the steel plate. Since the workpieces are prone to sticking to the steel plate or deforming during cutting, it is difficult to remove the workpieces from the steel plate. Therefore, workers need to cut a cutting-off line on the steel plate to allow the workpieces to be removed from the steel plate smoothly. Moreover, the workpieces are removed from the steel plate by manual sorting to prevent damage to the workpieces. This is inefficient and time-consuming. SUMMARY

[0004] To solve the above problems, the present application provides a cutting-off line positioning method, comprising:

[0005] Based on the theoretical model data of the steel plate containing a plurality of workpiece contour cutting line models, a plurality of theoretical cutting-off line position data are obtained;

[0006] The steel plate is subjected to overall visual inspection to obtain actual edge contour data of the steel plate;

[0007] Based on the actual edge contour data of the steel plate, 3D position data of each corner of the steel plate are obtained;

[0008] The theoretical model data of the steel plate and the 3D position data of each corner of the steel plate are matched to obtain the actual attitude of the steel plate;

[0009] Based on the actual attitude of the steel plate and the plurality of theoretical cutting-off line position data, the pre-set positions of a plurality of corresponding actual cutting-off lines are subjected to local accurate visual inspection one by one;

[0010] Based on the results of the plurality of local accurate visual inspections, the positions of the plurality of actual cutting-off lines are determined.

[0011] In addition, the present application also provides a cutting method for an intelligent cutting device, wherein the intelligent cutting device comprises a cutting gun, a mechanical arm, a grabbing device and a visual recognition device, the cutting gun, the visual recognition device and the grabbing device are all arranged at the end of the mechanical arm,

[0012] The grabbing device comprises two electromagnets for adsorbing the cut workpieces, and the distance between the two electromagnets can be automatically adjusted,

[0013] The cutting method comprises:

[0014] Based on a plurality of different workpiece contour cutting line models, the grabbing positions of the two electromagnets for a plurality of different workpieces are determined;

[0015] The intelligent cutting equipment is used to cut a plurality of different workpiece contour actual cutting slots on the steel plate;

[0016] The intelligent cutting equipment is used to perform the cutting line positioning method and cut a plurality of actual cutting lines.

[0017] Based on the determined grabbing positions of a plurality of different workpieces, the distance between the two electromagnets is adjusted, and the grabbing device is used to grab and sort a plurality of different workpieces.

[0018] In addition, the present application also provides an intelligent cutting equipment for performing the cutting method,

[0019] The grabbing device of the intelligent cutting equipment is provided with a torque sensor for detecting the grabbing force of the grabbing device on the workpiece.

[0020] During the grabbing process of the grabbing device on the workpiece, the intelligent cutting equipment judges whether to continue to make the grabbing device grab the workpiece based on whether the grabbing force detected by the torque sensor exceeds a set value.

[0021] The technical effects of the present application at least include:

[0022] In the present application, first, based on the steel plate theoretical model data containing a plurality of workpiece contour cutting line models, a plurality of theoretical cutting line position data are obtained. Thus, the accuracy of the subsequent preset positions of a plurality of corresponding actual cutting lines is ensured. Then, the steel plate is subjected to overall visual detection to obtain the actual edge contour data of the steel plate, and the 3D position data of each corner of the steel plate is obtained accordingly, and then the steel plate theoretical model data and the 3D position data of each corner of the steel plate are matched to ensure the accuracy of the obtained actual posture of the steel plate. Then, based on the accurately obtained actual posture of the steel plate and a plurality of theoretical cutting line position data, the preset positions of the accurately cut actual cutting lines are subjected to local accurate visual detection, so as to accurately position the cutting line in combination with the visual features of the preset positions of the actual cutting lines on the steel plate accurately detected by visual detection. Then, the automatic cutting robot cuts the corresponding cutting line on the steel plate according to the accurately positioned cutting line.

[0023] Thus, compared with the prior art, the cutting line positioning method, the cutting method and the intelligent cutting device are combined to realize full automation of cutting line positioning, cutting and grabbing, replace manual operation, greatly improve efficiency, reduce labor cost and thus greatly reduce cost. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 The schematic flow chart of the main steps of the cutting line positioning method described in the specific embodiment of the present application;

[0025] Figure 2 The schematic flow chart of obtaining multiple theoretical cutting line position data for the specific embodiment of the present application;

[0026] Figure 3 The image data schematic diagram of obtaining multiple theoretical cutting lines in the yellow coil marked white line, the green line is the workpiece contour cutting line, and the blue line is the steel plate edge contour line for the specific embodiment of the present application;

[0027] Figure 4 The schematic flow chart of the specific embodiment of the present application for overall visual inspection of the steel plate;

[0028] Figure 5 The schematic flow chart of matching the theoretical contour data of the steel plate and the 3D position data of each corner of the steel plate for the specific embodiment of the present application;

[0029] Figure 6 The projection schematic diagram of the steel plate for the specific embodiment of the present application;

[0030] Figure 7 The steel plate region schematic diagram for the specific embodiment of the present application;

[0031] Figure 8 The steel plate edge contour line schematic diagram for the specific embodiment of the present application;

[0032] Figure 9 The relative initial state schematic diagram before matching the theoretical contour data of the steel plate and the 3D position data of each corner of the steel plate for the specific embodiment of the present application, wherein the white line is the 3D position data of each corner of the steel plate, and the red line is the theoretical contour data of the steel plate;

[0033] Figure 10A schematic diagram of the matching state of the theoretical profile data of the steel plate and the 3D position data of each corner of the steel plate in the specific embodiment of the present application, wherein the white line is the 3D position data of each corner of the steel plate, and the red line is the theoretical profile data of the steel plate;

[0034] Figure 11 A schematic flow chart of determining the positions of a plurality of the actual cutting lines in the specific embodiment of the present application;

[0035] Figure 12 A schematic flow chart of respectively performing high-exposure visual detection, laser visual detection and normal-exposure visual detection on the area of the preset position in the specific embodiment of the present application;

[0036] Figure 13 A schematic diagram of laser image data obtained by laser visual detection in the specific embodiment of the present application;

[0037] Figure 14 A schematic diagram of extracting laser center points by laser visual detection in the specific embodiment of the present application;

[0038] Figure 15 A schematic diagram of extracting candidate cutting lines from the laser image data in the specific embodiment of the present application;

[0039] Figure 16 A schematic diagram of overexposure image data obtained by high-exposure visual detection in the specific embodiment of the present application;

[0040] Figure 17 A schematic diagram of extracting one cutting line by high-exposure visual detection in the specific embodiment of the present application, wherein the blue line is one cutting line;

[0041] Figure 18 A schematic diagram of extracting another cutting line by high-exposure visual detection in the specific embodiment of the present application, wherein the blue line is another cutting line;

[0042] Figure 19 A schematic diagram of the preset position of the actual cutting line in the specific embodiment of the present application, wherein the two blue points represent the outer edge points of the actual cutting, the two red points represent the middle points of the slit, and the two green points represent the inner edge points of the actual cutting;

[0043] Figure 20 A schematic flow chart of the cutting method in the specific embodiment of the present application;

[0044] Figure 21 A schematic flow chart of determining the grasping positions of a plurality of different workpieces by the electromagnets in the specific embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to make the above objectives, characteristics and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application are described in detail below with reference to the drawings. In the following description, a large number of specific details are set forth in order to facilitate a thorough understanding of embodiments of the present application. It should be understood that the specific embodiments described herein are merely intended to explain the present application and are not intended to limit the present application. Embodiments of the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without departing from the scope of the present application. Therefore, the present application is not limited by the specific embodiments disclosed below.

[0046] It can be understood that the terms "first", "second" and the like used in the present application can be used herein to describe various technical terms, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. However, unless otherwise specified, these technical terms are not limited by these terms. These terms are only used to distinguish one technical term from another. For example, without departing from the scope of the present application, the first receiving device and the second receiving device are different receiving devices, the first surface and the second surface are different surfaces, and the first plane, the second plane, the third plane and the fourth plane are different planes. In the description of embodiments of the present application, the meaning of "a plurality of" or "several" is at least two, for example, two, three, etc., unless otherwise explicitly specified.

[0047] In the description of embodiments of the present application, unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "setting", "fixing" and the like should be understood in a broad sense, for example, it can be fixedly connected, or it can be detachably connected, or it can be integrated; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the internal communication of two components or the interaction relationship between two components, unless otherwise explicitly limited. For those skilled in the art, the specific meaning of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0048] Unless otherwise defined, all technical and scientific terms used in the present application have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used in the present application in the specification are only for the purpose of describing specific embodiments and are not intended to limit the present application.

[0049] Referring to Figures 1 to 21 The present embodiment provides a cutting line positioning method based on a theoretical model and visual detection, comprising:

[0050] Based on the steel plate theoretical model data containing a plurality of workpiece contour cutting line models, a plurality of theoretical cutting line position data are obtained;

[0051] performing overall visual inspection on the steel plate to obtain actual edge profile data of the steel plate;

[0052] based on the actual edge profile data of the steel plate, obtaining 3D position data of each corner of the steel plate;

[0053] matching the steel plate theoretical model data and the 3D position data of each corner of the steel plate to obtain an actual posture of the steel plate;

[0054] based on the actual posture of the steel plate and the plurality of theoretical cutting line position data, performing local accurate visual inspection on a plurality of corresponding preset positions of actual cutting lines one by one;

[0055] based on the results of the plurality of local accurate visual inspections, determining the positions of the plurality of actual cutting lines.

[0056] It should be noted that based on the steel plate theoretical model data containing a plurality of workpiece profile cutting line models, a plurality of theoretical cutting line position data can be obtained by automatic calculation by a computer, or a plurality of theoretical cutting lines can be manually set by a computer according to the mutual positional relationship of the plurality of workpiece profile cutting line models, thereby obtaining the plurality of theoretical cutting line position data.

[0057] The theoretical cutting line position data in the present embodiment can refer to the plane coordinates of the two ends of the theoretical cutting line.

[0058] In the present embodiment, first, based on the steel plate theoretical model data containing a plurality of workpiece profile cutting line models, a plurality of theoretical cutting line position data is obtained, thereby ensuring the accuracy of the subsequent determination of the plurality of corresponding preset positions of actual cutting lines. Then, overall visual inspection is performed on the steel plate to obtain actual edge profile data of the steel plate, and based thereon, 3D position data of each corner of the steel plate is obtained, and then the steel plate theoretical model data and the 3D position data of each corner of the steel plate are matched, thereby ensuring the accuracy of the obtained actual posture of the steel plate. Then, based on the accurately obtained actual posture of the steel plate and the plurality of theoretical cutting line position data, local accurate visual inspection is performed on the preset positions of the accurately cut actual cutting lines, thereby combining the visual features at the preset positions of the actual cutting lines on the steel plate accurately detected by the accurate visual inspection to accurately position the cutting lines. Then, the automatic cutting robot cuts the corresponding cutting lines on the steel plate according to the accurately positioned cutting lines.

[0059] In this way, compared with the manual cutting method of cutting the cutting lines according to the operation experience in the prior art, the cutting efficiency of the cutting lines can be greatly improved, the labor cost can be reduced, and the production cost can be greatly reduced.

[0060] Reference is made to Figures 1 to 10, further, obtaining actual edge contour data of the steel plate through overall visual inspection of the steel plate comprises:

[0061] visually detecting the outer contour edge point cloud data of the steel plate;

[0062] fitting the outer contour edge point cloud data of the steel plate into a steel plate plane;

[0063] rotating the steel plate plane to be horizontal;

[0064] projecting the outer contour edge point cloud data of the steel plate to a two-dimensional plane in combination with the steel plate plane rotated to be horizontal, and performing morphological processing to obtain complete area contour data of the steel plate;

[0065] extracting actual edge contour lines from the complete area contour data of the steel plate;

[0066] calculating actual position contour edge lines and actual intersection points of the steel plate according to the theoretical contour data of the steel plate and the actual edge contour lines. Since the actual intersection points converge together, the corresponding actual corner points are formed, so the actual intersection points here can also be called actual corner points.

[0067] It should be noted that the actual edge contour data can be the coordinates of the actual edge contour in the plane.

[0068] Referring to Figures 1 to 10 , further, obtaining each corner 3D position data of the steel plate based on the actual edge contour data of the steel plate comprises:

[0069] restoring the actual position contour edge lines and the actual intersection points of the steel plate to the corresponding positions of the three-dimensional point cloud to obtain each corner 3D point cloud corner contour data of the steel plate;

[0070] matching the theoretical contour data of the steel plate and each corner 3D position data of the steel plate to obtain the actual pose of the steel plate comprises:

[0071] obtaining the theoretical centroid of the steel plate based on the theoretical contour data of the steel plate;

[0072] obtaining the actual centroid of the steel plate based on each corner 3D point cloud corner contour data of the steel plate;

[0073] making the theoretical centroid of the steel plate coincide with the actual centroid of the steel plate, and rotating the theoretical contour of the steel plate to coincide with each corner 3D point cloud corner contour of the steel plate, and obtaining a conversion matrix in combination with an ICP algorithm; the ICP algorithm here can also be called an iterative closest point algorithm.

[0074] Based on the conversion matrix, the actual posture of the steel plate is obtained.

[0075] This calculation method is simple, which can simplify the calculation process while ensuring the accuracy of obtaining the actual posture of the steel plate.

[0076] Referring to Figure 2 Further, based on the steel plate theoretical model data containing a plurality of workpiece contour cutting line models, a plurality of theoretical cutting line position data are obtained, including:

[0077] Step 1, by a plurality of straight line segments, all the convex hull points of one of the workpiece contour cutting lines are connected to the nearest points of other workpiece contour cutting lines around the workpiece contour cutting line, or to the nearest points of the steel plate edge contour line around the workpiece contour cutting line.

[0078] Step 2, delete the straight line segment passing through any one of the workpiece contour cutting lines; thereby filtering out redundant invalid straight line segments.

[0079] Step 3, a set number of straight line segment position data with relatively far distance between each other and relatively short length in the plurality of straight line segments are set as the theoretical cutting line position data; the straight line segment position with relatively far distance between each other in the plurality of straight line segments forms a diagonal arrangement relative to the shape surrounded by the workpiece contour cutting line. The cutting line arranged in this way can ensure that the workpiece cut along the workpiece contour cutting line can be smoothly removed from the steel plate, providing more reliable guarantee for subsequent automatic sorting of the workpiece. Moreover, the set number of straight line segments with relatively short length can reduce the length of the actual cutting line as much as possible, saving cutting time.

[0080] Step 4, sequentially execute steps 1 to 3 on a plurality of workpiece contour cutting line models;

[0081] Step 5, only keep the one with the shortest length in the intersecting plurality of theoretical cutting line position data or the approximately plurality of theoretical cutting line position data as the corresponding theoretical cutting line position data.

[0082] Considering that executing steps 1 to 3 on each of the workpiece contour cutting line models will inevitably cause the theoretical cutting line to overlap, which is not conducive to subsequent local accurate visual detection. Therefore, only the one with the shortest length in the intersecting plurality of theoretical cutting line position data or the approximately plurality of theoretical cutting line position data is kept as the corresponding theoretical cutting line position data. Thus, the generated theoretical cutting line is further filtered, ensuring that the finally obtained theoretical cutting line is more reasonable.

[0083] It should be noted that the number of the straight line segments can be 2, and the number of the straight line segments as the cutting lines can be less, thereby saving time.

[0084] It should be noted that the plurality of the theoretical cutting line position data can be a nesting file (.nc format file).

[0085] Referring to Figure 2 Further, after the step 1, the plurality of the straight line segments are connected to the nearest points of the other workpiece contour cutting lines around the workpiece contour cutting line or the nearest points of the steel plate edge contour line around the workpiece contour cutting line, the step 1 further includes:

[0086] The convex points are offset by a set distance along the workpiece contour cutting line, so that the generated straight line segments are perpendicular to the workpiece contour cutting lines adjacent to the two ends of the straight line segments.

[0087] For example, the convex points can be offset by 20 cm along the workpiece contour cutting line.

[0088] The generated straight line segments are perpendicular to the workpiece contour cutting lines adjacent to the two ends of the straight line segments, so that the local accurate visual detection is more convenient to recognize and position in the subsequent local accurate visual detection, and the local accurate visual detection is ensured to be accurate.

[0089] Further, before the step of performing the overall visual detection on the steel plate to obtain the actual edge contour data of the steel plate, the step further includes:

[0090] Based on the workpiece contour cutting line model, a plurality of actual cutting slots of the workpiece contour are cut on the steel plate;

[0091] Based on the actual attitude of the steel plate and the plurality of the theoretical cutting line position data, the pre-set positions of a plurality of corresponding actual cutting lines are sequentially subjected to the local accurate visual detection, and the step includes:

[0092] The area of the pre-set position is subjected to the high-exposure visual detection, the laser visual detection and the normal-exposure visual detection, respectively, to obtain the corresponding over-exposure image data, the laser image data and the normal image data;

[0093] The over-exposure image data, the laser image data and the normal image data all contain the image data of the actual cutting joints at the two ends of the pre-set position, and the laser in the laser visual detection extends along the direction of the theoretical cutting line and covers the surface of the actual cutting joint at the two ends of the pre-set position.

[0094] The determining of the position of the actual cutting line based on the results of the multiple local accurate visual detections comprises:

[0095] The actual cutting line data corresponding to the actual cutting line at both ends of the preset position is extracted from the overexposed image data.

[0096] The laser center point in the laser image data is extracted, and candidate cutting line data is obtained based on the laser center point.

[0097] The simulated cutting line data is obtained based on the normal image data and the theoretical cutting line position data, that is, the simulated cutting line is drawn according to the theoretical cutting line position in the normal image.

[0098] The position data of the actual cutting line is determined based on the cutting line data and the candidate cutting line data.

[0099] The position data of the actual cutting line is interactively confirmed with the simulated cutting line data, and if the position data of the actual cutting line deviates from the simulated cutting line data within a first set range, the position of the actual cutting line is determined by the position data of the actual cutting line, thereby ensuring the accuracy of the positioning of the actual cutting line.

[0100] In this way, before the actual cutting line is cut, the accurate positioning of the actual cutting line at the preset position is realized by the combination of high-exposure visual detection, laser visual detection and normal-exposure visual detection.

[0101] Referring to Figures 11 to 19 Further, the determining of the position data of the actual cutting line based on the cutting line data and the candidate cutting line data comprises:

[0102] The position data of the outer edge point of the actual cutting line at both ends of the preset position, the position data of the inner edge point and the position data of the gap intermediate point of the actual cutting line are obtained based on the cutting line data.

[0103] If the position data of the outer edge point and the position data of the gap intermediate point deviate within a second set range, it is determined that the plate surfaces on both sides of the actual cutting line are substantially flush, and the position data of the actual cutting line is determined based on the inner edge point, and if not within the second set range, it is determined that the plate surfaces on both sides of the actual cutting line are too high, and the actual cutting line at the preset position is skipped; preventing the head of the cutting gun from colliding with the local position of the high protrusion, thereby protecting the cutting gun.

[0104] It should be noted that the position data in the embodiment can refer to position coordinates.

[0105] If the deviation of the position data of the inner edge point and the position data of the gap middle point is within a third set range, it is determined that the actual cutting line of the preset position is adopted by the piercing cutting, and if it is not within the third set range, it is determined that the actual cutting line of the preset position is not adopted by the piercing cutting. In this way, the workpiece is prevented from being cut when the cutting gun emits light.

[0106] Thus, the detailed process of the cutting line positioning method in the embodiment can be:

[0107] Step 1, connecting all the convex points of a workpiece contour cutting line with the nearest points of other workpiece contour cutting lines around the workpiece contour cutting line or with the nearest points of the steel plate edge contour line around the workpiece contour cutting line through a plurality of straight line segments;

[0108] Step 2, deleting the straight line segment passing through any one of the workpiece contour cutting lines;

[0109] Step 3, taking the position data of a set number of straight line segments with a relatively far distance from each other and a relatively short length in the plurality of straight line segments as the theoretical cutting line position data;

[0110] Step 4, sequentially executing steps 1 to 3 on a plurality of workpiece contour cutting line models;

[0111] Step 5, only keeping the one with the shortest length in the intersecting plurality of theoretical cutting line position data or the approximately plurality of theoretical cutting line position data as the corresponding theoretical cutting line position data;

[0112] Step 6, visually detecting the steel plate's outer contour edge point cloud data;

[0113] Step 7, fitting the steel plate's outer contour edge point cloud data into a steel plate plane;

[0114] Step 8, rotating the steel plate plane to be horizontal;

[0115] Step 9, combining the steel plate plane rotated to be horizontal, projecting the steel plate's outer contour edge point cloud data to a two-dimensional plane, and performing morphological processing to obtain the complete area contour data of the steel plate;

[0116] Step 10, extracting the actual edge contour line from the complete area contour data of the steel plate;

[0117] Step 11, calculating the actual position contour edge line and the actual intersection point of the steel plate according to the theoretical contour data of the steel plate and the actual edge contour line.

[0118] Step 12, restoring the actual position profile edge line and the actual intersection of the steel plate to the corresponding position of the three-dimensional point cloud to obtain the 3D point cloud corner profile data of each corner of the steel plate;

[0119] Step 13, obtaining the theoretical centroid of the steel plate based on the theoretical profile data of the steel plate;

[0120] Step 14, obtaining the actual centroid of the steel plate based on the 3D point cloud corner profile data of each corner of the steel plate;

[0121] Step 15, making the theoretical centroid of the steel plate coincide with the actual centroid of the steel plate, and rotating the theoretical profile of the steel plate to the shape coincidence relative to the 3D point cloud corner profile of each corner of the steel plate, and obtaining the conversion matrix by combining the ICP algorithm;

[0122] Step 16, obtaining the actual pose of the steel plate based on the conversion matrix.

[0123] Step 17, respectively performing high-exposure visual detection, laser visual detection and normal-exposure visual detection on the area of the preset position to obtain corresponding overexposure image data, laser image data and normal image data;

[0124] Step 18, extracting the actual cutting seam corresponding to the cutting seam line data at both ends of the preset position in the overexposure image data;

[0125] Step 19, extracting the laser center point in the laser image data, and obtaining candidate cutting line data based on the laser center point;

[0126] Step 20, obtaining the simulated cutting line data based on the normal image data and the theoretical cutting line position data;

[0127] Step 21, obtaining the position data of the outer edge point, the position data of the inner edge point and the position data of the gap intermediate point of the actual cutting seam at both ends of the preset position based on the cutting seam line data;

[0128] Step 22, if the deviation of the position data of the outer edge point and the position data of the gap intermediate point is within the second set range, it is determined that the plate surface height on both sides of the actual cutting seam is basically flat, and the position data of the actual cutting line is determined based on the inner edge point, if it is not within the second set range, it is determined that the plate surface height on both sides of the actual cutting seam is too large, and the actual cutting line of the preset position is skipped.

[0129] Step 23, if the deviation of the position data of the inner edge point and the position data of the slit middle point is within a third set range, it is determined that the actual cutting line of the preset position is adopted by the piercing cutting, if not within the third set range, it is determined that the actual cutting line of the preset position is not adopted by the piercing cutting;

[0130] Step 24, the position data of the actual cutting line is interactively confirmed with the simulation cutting line data, if the position data of the actual cutting line deviates from the simulation cutting line data within a first set range, the position of the actual cutting line is determined by the position data of the actual cutting line.

[0131] Referring to Figure 20 and Figure 21 In addition, the embodiment also provides a cutting method for an intelligent cutting device, the intelligent cutting device comprising: a cutting gun, a mechanical arm, a grabbing device and a visual recognition device, the cutting gun, the visual recognition device and the grabbing device are all arranged at the end of the mechanical arm,

[0132] The grabbing device comprises two electromagnets, the electromagnets are used for adsorbing and grabbing the cut workpieces, the distance between the two electromagnets can be automatically adjusted,

[0133] The cutting method comprises:

[0134] Based on a plurality of different workpiece contour cutting line models, the grabbing positions of the two electromagnets for grabbing a plurality of different workpieces are determined;

[0135] Using the intelligent cutting device, a plurality of actual cutting slots of different workpiece contours are cut on the steel plate;

[0136] Using the intelligent cutting device, the cutting line positioning method is executed, and a plurality of actual cutting lines are cut;

[0137] Based on the determined grabbing positions of a plurality of different workpieces, the distance between the two electromagnets is adjusted, and the grabbing device is used to grab and sort a plurality of different workpieces.

[0138] It should be noted that the grabbing and sorting of a plurality of different workpieces by the grabbing device herein can mean sorting different workpieces to corresponding different positions for classification.

[0139] Further, based on a plurality of different workpiece contour cutting line models, the grabbing positions of the two electromagnets for grabbing a plurality of different workpieces are determined, which comprises:

[0140] Based on the workpiece contour cutting line model, the maximum inscribed rectangle of the theoretical contour of the workpiece and the barycentric position of the workpiece are calculated and obtained;

[0141] The two electromagnets are arranged to have the same direction of extension as the length direction of the maximum inscribed rectangle.

[0142] The two electromagnets are arranged to have the same direction of extension as the length direction of the maximum inscribed rectangle.

[0143] In addition, the embodiment further provides an intelligent cutting device for executing the cutting method.

[0144] The intelligent cutting device is provided with a torque sensor at the grabbing device, which is used to detect the lifting force of the grabbing device on the workpiece.

[0145] During the process of grabbing the workpiece by the grabbing device, the intelligent cutting device judges whether to continue to make the grabbing device grab the workpiece based on whether the torque sensor detects that the lifting force exceeds a set value, so as to prevent the workpiece from being stuck by the slit edge of the steel plate and make the mechanical arm unable to grab, thereby causing the mechanical arm to be overloaded.

[0146] In this way, the above-mentioned cutting line positioning method, cutting method and intelligent cutting device are combined to realize full automation of cutting line positioning, cutting and sorting, replace manual operation in the prior art, greatly improve the efficiency, reduce the labor cost, and thus greatly reduce the cost.

[0147] Although the present disclosure is as above, the protection scope of the present disclosure is not limited to this. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present disclosure, and these changes and modifications shall fall within the protection scope of the present disclosure.

Claims

1. A method of positioning a wire severing device, comprising: The method comprises the following steps: Based on the steel plate theoretical model data containing a plurality of workpiece contour cutting line models, a plurality of theoretical cutting line position data are obtained, which comprises the following steps: Step 1, connect all the convex hull points of one of the workpiece contour cutting lines to the nearest points of other workpiece contour cutting lines around the workpiece contour cutting line or to the nearest points of the edge contour line of the steel plate around the workpiece contour cutting line through a plurality of straight line segments; Step 2, delete the straight line segment passing through any one of the workpiece contour cutting lines; Step 3, set a certain number of straight line segment position data with relatively far distance and relatively short length among a plurality of straight line segments as the theoretical cutting line position data; Step 4, sequentially execute steps 1 to 3 on a plurality of workpiece contour cutting line models; Step 5, only keep the one with the shortest length among the intersecting or similar theoretical cutting line position data as the corresponding theoretical cutting line position data; Overall visual inspection is performed on the steel plate to obtain actual edge contour data of the steel plate; Based on the actual edge contour data of the steel plate, each corner 3D position data of the steel plate is obtained; The steel plate theoretical model data and each corner 3D position data of the steel plate are matched to obtain the actual attitude of the steel plate; Based on the actual attitude of the steel plate and a plurality of theoretical cutting line position data, a plurality of pre-set positions of actual cutting lines are locally and accurately detected one by one; Based on the results of a plurality of local accurate visual detections, the positions of a plurality of actual cutting lines are determined.

2. The severing thread positioning method of claim 1, wherein The overall visual inspection on the steel plate to obtain the actual edge contour data of the steel plate comprises the following steps: Visual inspection is performed on the outer contour edge point cloud data of the steel plate; The outer contour edge point cloud data of the steel plate is fitted into a steel plate plane; The steel plate plane is rotated to be horizontal; The outer contour edge point cloud data of the steel plate is projected onto a two-dimensional plane in combination with the steel plate plane rotated to be horizontal, and morphological processing is performed to obtain complete area contour data of the steel plate; Actual edge contour lines are extracted from the complete area contour data of the steel plate; According to the theoretical contour data of the steel plate and the actual edge contour lines, actual position contour edge lines and actual intersection points of the steel plate are calculated and obtained.

3. The severing thread positioning method of claim 2, wherein Based on the actual edge contour data of the steel plate, each corner 3D position data of the steel plate is obtained, which comprises the following steps: The actual position contour edge lines and actual intersection points of the steel plate are restored to the corresponding positions of the three-dimensional point cloud to obtain each corner 3D point cloud corner contour data of the steel plate; The steel plate theoretical model data and each corner 3D position data of the steel plate are matched to obtain the actual attitude of the steel plate, which comprises the following steps: Based on the theoretical contour data of the steel plate, a theoretical centroid of the steel plate is obtained; Based on each corner 3D point cloud corner contour data of the steel plate, an actual centroid of the steel plate is obtained; aligning a theoretical center of mass of the steel plate with an actual center of mass of the steel plate and rotating a theoretical profile of the steel plate to a shape of a corner profile of each corner 3D point cloud of the steel plate to coincide, and obtaining a conversion matrix by combining an ICP algorithm; obtaining an actual pose of the steel plate based on the conversion matrix.

4. The severing thread positioning method of claim 1, wherein Step 1, connecting all the convex points of a workpiece profile cutting line to the nearest points of other workpiece profile cutting lines around the workpiece profile cutting line or to the nearest points of a steel plate edge profile line around the workpiece profile cutting line by a plurality of straight line segments, and then including: offsetting the convex points along the workpiece profile cutting line by a set distance to make the generated straight line segments perpendicular to the workpiece profile cutting line adjacent to both ends of the straight line segments.

5. The cutting line positioning method according to any one of claims 1 to 3, characterized in that, before the overall visual inspection of the steel plate to obtain the actual edge profile data of the steel plate, including: cutting a plurality of actual cutting seams of workpiece profiles on the steel plate based on the workpiece profile cutting line model; based on the actual pose of the steel plate and a plurality of theoretical cutting line position data, performing local accurate visual inspection on a plurality of preset positions of the corresponding actual cutting lines one by one, including: performing high-exposure visual inspection, laser visual inspection and normal-exposure visual inspection on the area of the preset position respectively to obtain corresponding overexposure image data, laser image data and normal image data; wherein the overexposure image data, laser image data and normal image data all contain image data of the actual cutting seam at both ends of the preset position, and the laser in the laser visual inspection extends along the direction of the theoretical cutting line and covers the surface of the actual cutting seam at both ends of the preset position; based on the results of a plurality of local accurate visual inspections, determining the positions of a plurality of actual cutting lines, including: extracting cutting seam line data corresponding to the actual cutting seam at both ends of the preset position in the overexposure image data; extracting laser center points in the laser image data and obtaining candidate cutting line data based on the laser center points; obtaining simulated cutting line data based on the normal image data and the theoretical cutting line position data; determining the position data of the actual cutting line based on the cutting seam line data and the candidate cutting line data; interactively confirming the position data of the actual cutting line with the simulated cutting line data, and if the position data of the actual cutting line deviates from the simulated cutting line data within a first set range, determining the position of the actual cutting line with the position data of the actual cutting line.

6. The severing thread positioning method of claim 5, wherein, based on the cutting seam line data and the candidate cutting line data, determining the position data of the actual cutting line, including: based on the cutting seam line data, obtaining position data of an outside edge point, position data of an inside edge point and position data of a gap middle point of the actual cutting seam at both ends of the preset position; If the deviation between the position data of the outer edge point and the position data of the slit middle point is within a second set range, it is determined that the plate surface heights on both sides of the actual cutting line are substantially flat, and the position data of the actual cutting line at the preset position is determined based on the position data of the inner edge point; if the deviation is not within the second set range, it is determined that the plate surface heights on both sides of the actual cutting line are too large, and the actual cutting line at the preset position is skipped; If the deviation between the position data of the inner edge point and the position data of the slit middle point is within a third set range, it is determined that the actual cutting line at the preset position is cut by perforation; if the deviation is not within the third set range, it is determined that the actual cutting line at the preset position is not cut by perforation.

7. A method of cutting, characterized by The intelligent cutting device comprises a cutting gun, a mechanical arm, a grabbing device and a visual recognition device, the cutting gun, the visual recognition device and the grabbing device are arranged at the end of the mechanical arm, The grabbing device comprises two electromagnets for adsorbing and grabbing a cut workpiece, and the distance between the two electromagnets can be automatically adjusted, The cutting method comprises: Based on a plurality of different workpiece contour cutting line models, the grabbing positions of the two electromagnets for grabbing a plurality of different workpieces are determined; Using the intelligent cutting device, a plurality of actual cutting slots of different workpiece contours are cut on a steel plate; Using the intelligent cutting device, the cutting line positioning method of any one of claims 1 to 6 is executed, and a plurality of actual cutting lines are cut; Based on the determined grabbing positions of a plurality of different workpieces, the distance between the two electromagnets is adjusted, and a plurality of different workpieces are grabbed and sorted using the grabbing device.

8. The cutting method according to claim 7, wherein, Based on a plurality of different workpiece contour cutting line models, the grabbing positions of the two electromagnets for grabbing a plurality of different workpieces are determined, which comprises: Based on the workpiece contour cutting line model, the maximum inscribed rectangle of the theoretical contour of the workpiece and the center of gravity position of the workpiece are calculated and obtained; The grabbing angles of the two electromagnets are determined so that the extension directions of the two electromagnets are consistent with the length direction of the maximum inscribed rectangle; Starting from the center of gravity position of the workpiece, along the long side of the maximum inscribed rectangle, the long sides of the maximum inscribed rectangle are traversed respectively to the two ends of the long sides, and when the distance between the two electromagnets reaches the maximum value, the grabbing positions of the two electromagnets for grabbing the workpiece are determined.

9. A smart cutting device, characterized in that, The intelligent cutting device is used to execute the cutting method of claim 7 or 8, A torque sensor is arranged at the grabbing device of the intelligent cutting device, and the torque sensor is used to detect the lifting force of the grabbing device on the workpiece; During the process of grabbing the workpiece by the grabbing device, the intelligent cutting device determines whether to continue to make the grabbing device grab the workpiece based on whether the lifting force detected by the torque sensor exceeds a set value.

Images