An automatic sorting device for laser-cut parts

Abstract

The application relates to an automatic sorting device for laser cutting pieces, which comprises: a human-computer interaction processing module for acquiring first information for identifying laser cutting pieces; a numerical control task management module for establishing a navigation map of a region to which the laser cutting pieces belong and determining a search strategy; an image acquisition assembly for collecting images according to the search strategy and the navigation map; and a visual analysis processing module for identifying the images collected by the image acquisition assembly, obtaining position information of the laser cutting pieces transmitted to a cutting piece picking assembly, and enabling the cutting piece picking assembly to sort the laser cutting pieces based on the position information of the laser cutting pieces. The above device realizes intelligent automatic picking and can flexibly pick various types and different sizes of laser cutting pieces.

Description

[0001] This case belongs to the divisional application of application number: 202310127103.3 TECHNICAL FIELD

[0002] The present application relates to the field of visual recognition technology, and in particular to an automatic sorting device for laser cutting parts. BACKGROUND

[0003] At present, manually picking up laser cutting parts of metal plates is time-consuming and laborious. The large-arm spread picking robot picks up metal plates at a high cost, and often can only be applied to large workpieces. The arm span of a small picking robot itself cannot cover the discharging platform of the laser cutting machine, and auxiliary moving mechanisms need to be matched with the robot to move to pick up the whole platform, which is not conducive to fast movement and fast rhythm picking.

[0004] In addition, the picking system based on laser cutting sleeve positioning of workpiece position has difficulty in processing the abnormal conditions such as "workpiece deviation, missing material, picking belt plate, stacking, and individual piece micro-connection" of medium and small material metal plates (especially thin plates) on the discharging platform. Therefore, a reliable and relatively low-cost automatic sorting method is needed to intelligently process the above-mentioned abnormal conditions so as to be able to quickly sort medium and small material (medium and thin plates) metal plates. SUMMARY

[0005] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present application provides an automatic sorting method and an automatic sorting system for laser cutting parts, which realizes the purpose of low-cost fast analysis and fast rhythm picking.

[0006] In order to achieve the above-mentioned purpose, the main technical solutions adopted by the present application include:

[0007] In a first aspect, the present application provides an automatic sorting method for laser cutting parts, which comprises:

[0008] S10, acquiring first information for identifying laser cutting parts;

[0009] S20, establishing a navigation map of the region to which the laser cutting parts belong according to the first information, and determining a search strategy of the laser cutting parts based on the navigation map;

[0010] The navigation map comprises a visual map composed of a plurality of sequentially arranged visual cells; each visual cell has a unique serial number, and each visual cell is a region corresponding to the field of view of an image acquisition component; the search strategy comprises a step length and a search order determined according to the field of view cell;

[0011] S30, sending an image acquisition instruction to the image acquisition component according to the search strategy and the navigation map, and identifying position information of the laser cutting piece according to image recognition of the image acquisition component based on the image acquisition instruction, so that the cutting piece picking component sorts the laser cutting piece based on the position information of the laser cutting piece.

[0012] Optionally, the method further comprises: S30 comprises:

[0013] According to the search strategy and the navigation map, a first image acquisition instruction is sent to the image acquisition component, and a first real-time image returned by the image acquisition component is acquired;

[0014] And, based on the first information, it is identified whether there is a laser cutting piece in the first real-time image, and if there is, the position information of the laser cutting piece is identified, so that the cutting piece picking component sorts the laser cutting piece based on the position information of the laser cutting piece.

[0015] Otherwise, according to the search strategy and the navigation map, an Nth image acquisition instruction is sent to the image acquisition component, so that the image acquisition component acquires and feeds back an Nth real-time image based on the Nth image acquisition instruction, based on the first information, it is identified whether there is a laser cutting piece in the Nth real-time image, and if there is, the position information of the laser cutting piece is identified, so that the cutting piece picking component sorts the laser cutting piece based on the position information of the laser cutting piece.

[0016] Otherwise, repeat the above sending of the N+1th image acquisition instruction, N is greater than or equal to 1; if the navigation map is traversed and there is no laser cutting piece, an alarm signal is sent.

[0017] Optionally, S10 comprises:

[0018] S11, the automatic sorting numerical control system receives the user input / imported laser cutting piece specified format workpiece drawing (such as CAD workpiece drawing), and the type of the laser cutting piece;

[0019] S12, the automatic sorting numerical control system receives the nesting file and / or laser cutting coordinate information transmitted by the laser cutting system;

[0020] S13, the automatic sorting numerical control system acquires the basic information of the laser cutting piece as the first information according to the specified format workpiece drawing (such as CAD workpiece drawing), the type of the laser cutting piece, the nesting file and / or laser cutting coordinate information; the basic information includes one or more of the following: shape, coordinate origin, plate information of the plate to which the laser cutting piece belongs, inferred coordinate position, correspondence between the type of the laser cutting piece and the material receiving device, contour, model feature, cutting area information.

[0021] Optionally, S20 comprises:

[0022] S21, the automatic sorting numerical control system acquires a picking area of the laser cutting piece according to basic information of the laser cutting piece, the picking area being a certain area on a picking platform;

[0023] S22, a navigation map composed of multiple visual cells is established according to the picking area and a field of view size of the image acquisition component;

[0024] S23, a search strategy of the laser cutting piece is obtained according to position information of the visual cells in the navigation map and a size of the visual cells, a horizontal movement step and a vertical movement step, and a movement path.

[0025] Optionally, S23 comprises:

[0026] The size of the visual cell is 90cmx90cm, the horizontal movement step and the vertical movement step are both 30cm, and the diameter / length of the laser cutting piece is a workpiece less than or equal to 60cm;

[0027] The size of the visual cell is 100cmx100cm, the horizontal movement step and the vertical movement step are both 30cm, and the diameter / length of the laser cutting piece is a workpiece less than or equal to 70cm;

[0028] The size of the visual cell is 100cmx100cm, the horizontal movement step and the vertical movement step are both 20cm, and the diameter / length of the laser cutting piece is a workpiece less than or equal to 80cm.

[0029] Optionally, the image acquisition instruction comprises center coordinate information of a to-be-acquired image area;

[0030] S30 comprises:

[0031] The first real-time image is subjected to binarization processing, and the binarized image is subjected to brightness conversion and histogram equalization to obtain an intermediate result image;

[0032] The intermediate result image is subjected to feature extraction, and it is judged whether the extracted feature contour matches the laser cutting piece contour in the first information;

[0033] If the match is found, it is determined that the first real-time image contains the laser cutting piece, the type number of the laser cutting piece is determined, and the position information of the laser cutting piece as the grabbing center coordinate is acquired according to the information of the contour of the extracted feature.

[0034] Optionally, S30 further comprises:

[0035] The cutting piece pickup assembly comprises a suction disc array, and the automatic sorting numerical control system determines information of a suction nozzle array combination to be opened in the cutting piece pickup assembly according to information of a profile of the extracted feature, so as to control the suction disc array to sort the laser cutting piece based on the position information of the laser cutting piece.

[0036] Optionally, the method further comprises:

[0037] S40, for the real-time image of the existing laser cutting piece, judging whether the laser cutting piece meets the quality standard of the type to which the laser cutting piece belongs, and outputting the judgment result.

[0038] In a second aspect, the embodiments of the present application also provide an automatic sorting device for laser cutting pieces, which comprises:

[0039] A man-machine interaction processing module is configured to acquire first information for identifying the laser cutting piece;

[0040] A numerical control task management module is configured to establish a navigation map of a region to which the laser cutting piece belongs according to the first information, and determine a search strategy of the laser cutting piece based on the navigation map; and send an image acquisition instruction to the image acquisition assembly according to the search strategy and the navigation map.

[0041] The navigation map comprises a visual map composed of a plurality of sequentially arranged visual cells; each visual cell has a unique serial number, and each visual cell is a region corresponding to a field of view of an image acquisition assembly; and the search strategy comprises a step length and a search order determined according to the field of view.

[0042] A visual analysis processing module is configured to identify position information of the laser cutting piece according to an image recognized by the image acquisition assembly based on the image acquisition instruction, so as to make the cutting piece pickup assembly sort the laser cutting piece based on the position information of the laser cutting piece.

[0043] In an optional implementation, the numerical control task management module is configured to send a first image acquisition instruction to the image acquisition assembly according to the search strategy and the navigation map,

[0044] The visual analysis processing module is configured to acquire a first real-time image returned by the image acquisition assembly; identify whether the first real-time image contains a laser cutting piece based on the first information, and if so, identify position information of the laser cutting piece; and the numerical control task management module is further configured to make the cutting piece pickup assembly sort the laser cutting piece based on the position information of the laser cutting piece.

[0045] When the visual analysis processing module determines that the laser cutting piece does not exist in the first real-time image, the numerical control task management module is configured to send an Nth image acquisition instruction to the image acquisition assembly according to a search strategy and a navigation map, so that the image acquisition assembly acquires an Nth real-time image based on the Nth image acquisition instruction and feeds back the Nth real-time image; the visual analysis processing module is further configured to identify, based on first information, whether the laser cutting piece exists in the Nth real-time image, and if the laser cutting piece exists, identify position information of the laser cutting piece;

[0046] The visual analysis processing module and the numerical control task management module interactively implement traversal of the navigation map, and if the navigation map is traversed and no laser cutting piece exists, the numerical control task management module sends an alarm signal.

[0047] Optionally, the man-machine interaction processing module is specifically configured to receive a user input / imported workpiece map of a specified format of the laser cutting piece and a type of the laser cutting piece.

[0048] Receive a nesting file and / or laser cutting coordinate information transmitted by a laser cutting system.

[0049] According to the workpiece map of the specified format, the type of the laser cutting piece, the nesting file and / or the laser cutting coordinate information, basic information of the laser cutting piece as the first information is obtained; the basic information includes one or more of the following: shape, coordinate origin, plate information of a plate to which the laser cutting piece belongs, a presumed coordinate position, a correspondence relationship between the type of the laser cutting piece and a material receiving device, a contour, a model feature, and cutting region information.

[0050] Optionally, the numerical control task management module is specifically configured to obtain a material picking area of the laser cutting piece according to the basic information of the laser cutting piece, the material picking area being a certain area on a material picking platform.

[0051] According to the material picking area and a field of view size of the image acquisition assembly, a navigation map composed of multiple visual cells is established.

[0052] According to position information of the visual cells in the navigation map and a size of the visual cells, a horizontal movement step and a vertical movement step, and a movement path are determined, and a search strategy of the laser cutting piece is obtained.

[0053] Correspondingly, the size of the visual cell is 90cmx90cm, the horizontal movement step and the vertical movement step are both 30cm, and the diameter / length of the laser cutting piece is a workpiece less than or equal to 60cm.

[0054] The size of the visual cell is 100cmx100cm, the horizontal movement step and the vertical movement step are both 30cm, and the diameter / length of the laser cutting piece is a workpiece less than or equal to 70cm.

[0055] The size of the visual cell is 100cmx100cm, and the transverse moving step and the longitudinal moving step are both 20cm; the diameter / length of the laser cutting piece is less than or equal to 80cm.

[0056] Optionally, the image acquisition instruction comprises: center coordinate information of an image area to be acquired.

[0057] Optionally, the visual analysis processing module is specifically configured to perform binaryzation processing on the first real-time image, and perform brightness conversion and histogram equalization on the binaryzation image to obtain an intermediate result image.

[0058] The intermediate result image is subjected to feature extraction, and it is judged whether the extracted feature contour matches the contour of the laser cutting piece in the first information.

[0059] If the match is found, it is determined that the first real-time image contains the laser cutting piece, the type number of the laser cutting piece is determined, and the position information of the laser cutting piece as the grabbing center coordinate is obtained according to the information of the extracted feature contour.

[0060] Optionally, the cutting piece picking assembly comprises a suction disc array, and the numerical control task management module determines the information of the suction nozzle array combination needed to be turned on in the cutting piece picking assembly according to the information of the extracted feature contour; and the suction disc array is controlled to sort the laser cutting piece based on the position information of the laser cutting piece.

[0061] In a third aspect, the embodiment of the present application further provides an automatic sorting system of a laser cutting piece, which is used for automatically sorting the laser cutting piece on a picking platform, and comprises: an industrial computer integrated with an automatic sorting numerical control system, an automatic controller, an image acquisition assembly, and a first transmission structure; the first transmission structure is used for moving the image acquisition assembly to a specified position of the picking platform.

[0062] The industrial computer sends one or more instructions to the automatic controller according to the first information, so that the automatic controller controls the first transmission structure to move the image acquisition assembly according to the one or more instructions, to acquire images of the specified position in real time.

[0063] The automatic sorting numerical control system in the industrial computer determines the position information of the laser cutting piece according to the first information and the real-time acquired images; and the cutting piece picking assembly picks up the laser cutting piece according to the position information of the laser cutting piece.

[0064] The first information is user input information or information transmitted by a laser cutting system.

[0065] Optionally, the cutting piece picking assembly is installed on the first transmission structure.

[0066] Alternatively, the cutting piece pickup assembly is mounted on a second transmission structure, and the industrial computer controls the second transmission structure to move the cutting piece pickup assembly to pick up the laser cutting piece according to the position information of the laser cutting piece by means of the automatic controller.

[0067] Optionally, the automatic sorting numerical control system executes the automatic sorting method of the laser cutting piece according to any one of the first aspect.

[0068] Optionally, the end of the cutting piece pickup assembly comprises an open suction disc array structure of multiple suction nozzles. In this embodiment, the combination of each suction disc of the suction disc array structure to achieve suction can be realized by programming.

[0069] Optionally, the industrial computer and the automatic controller exchange information through a Modbus-TCP bus protocol.

[0070] The automatic controller and the first transmission structure exchange information through an EtherCAT bus protocol.

[0071] The automatic controller controls the cutting piece pickup assembly through a DMCNET bus mode.

[0072] Optionally, the first transmission structure comprises:

[0073] Y-axis guide rails located on both sides of the picking platform,

[0074] a cross beam spanning above the picking platform, X-axis guide rails arranged on the cross beam,

[0075] A Z-axis lead screw is mounted on the cross beam, and a cutting piece pickup assembly is fixed on the Z-axis lead screw.

[0076] The image acquisition assembly is fixed on the Z-axis lead screw.

[0077] The automatic controller drives the Y-axis guide rails, the X-axis guide rails and the Z-axis lead screw to move in their respective directions by means of the first transmission structure.

[0078] In a fourth aspect, the embodiments of the present application further provide a laser cutting device, comprising: a laser cutting system and an automatic sorting system of the laser cutting piece according to any one of the third aspect.

[0079] After the laser cutting piece is processed, the laser cutting system moves the platform carrying the laser cutting piece of the laser cutting system to the area of the automatic sorting system, and serves as a picking platform. The automatic sorting system of the laser cutting piece identifies and picks up the laser cutting piece on the picking platform according to the automatic sorting method of any one of the first aspect.

[0080] In addition, the embodiment of the present application also provides an operation method of the automatic sorting system of the laser cutting piece, the automatic sorting system being the automatic sorting system of the above embodiment; the operation method comprises:

[0081] The automatic sorting system acquires the basic information of the laser cutting piece, establishes a navigation map of the region to which the laser cutting piece belongs, and determines a search strategy of the laser cutting piece based on the navigation map;

[0082] The automatic sorting system acquires a coordinate position required for a camera of the image acquisition assembly to take a photograph directly above the current visual cell according to the coordinate position of the current visual cell;

[0083] The automatic sorting system issues a first action instruction to the automatic controller based on the coordinate position of the camera, and the automatic controller controls the transmission system to send the camera to the position required for photographing based on the first action instruction;

[0084] The automatic sorting system controls the camera to take a photograph according to the ambient light and the camera parameters, acquires a real-time image taken by the camera, and analyzes whether there is a laser cutting piece conforming to the workpiece model feature in the real-time image; the workpiece model feature is a feature in the basic information of the laser cutting piece;

[0085] If there is, the position information of the laser cutting piece and the corresponding suction nozzle array opening combination of the cutting piece pickup assembly and the type of the laser cutting piece are acquired;

[0086] The automatic sorting system issues a second action instruction to the automatic controller based on the position information of the laser cutting piece, the corresponding suction nozzle array opening combination of the cutting piece pickup assembly and the type of the laser cutting piece, and the automatic controller controls the transmission system to drop the cutting pickup assembly to a pre-pickup height to realize the pickup of the laser cutting piece based on the second action instruction.

[0087] Further, if there is no laser cutting piece, the search and picking task of the current visual cell ends, the next visual cell is prepared to be searched according to the search strategy, and the search of all visual cells in the navigation map ends, and then the picking is completed.

[0088] The method of the embodiment of the present application divides a larger region (a region required to be searched and picked) into multiple visual cells (camera fields of view for close-range observation) one by one to search and analyze by photographing, so that the image acquisition assembly has low cost and fast photographing, transmission and analysis, and the defects of the prior art, such as slow transmission of a large field of view high-definition photograph and inability to quickly analyze, can be avoided.

[0089] In addition, the automatic sorting method of the present application can realize intelligent automatic picking, and can flexibly pick various types and different small sizes of laser cutting pieces, and can effectively solve the abnormal situations such as missing material and picking stacking in the prior art. BRIEF DESCRIPTION OF DRAWINGS

[0090] Figure 1 It is a schematic diagram of an automatic sorting system of a laser cutting part provided in the embodiment of the present application;

[0091] Figure 2 It is a flowchart of an automatic sorting method executed by an automatic sorting numerical control system in an industrial computer in the embodiment of the present application;

[0092] Figure 3 It is a schematic diagram of a transmission structure in the embodiment of the present application;

[0093] Figure 4 It is a flowchart of an automatic picking method in the embodiment of the present application;

[0094] Figure 5A It is a schematic diagram of a visual cell in the embodiment of the present application;

[0095] Figure 5B It is a schematic diagram of a visual cell and a search strategy in the embodiment of the present application;

[0096] Figure 5C It is a process schematic diagram of a search process using a search strategy in the embodiment of the present application;

[0097] Figure 6 It is a process schematic diagram of a picking process executed by a PLC in the embodiment of the present application Figure 1 ;

[0098] Figure 7 It is a process schematic diagram of a picking process executed by a PLC in the embodiment of the present application Figure 2 .

[0099] BRIEF DESCRIPTION OF DRAWINGS

[0100] 1, picking platform; 2, left Y-axis guide rail; 3, right Y-axis guide rail; 4, cross beam; 5, X-axis guide rail; 6, picking head; 7, auxiliary light source; 8, camera; 9, lifting shaft. DETAILED DESCRIPTION

[0101] In order to better explain the present application, so as to be understood, the present application is described in detail by specific embodiments in combination with the drawings.

[0102] The present application provides an automatic sorting device and sorting method for fast analysis and fast rhythm picking. The technical solutions of the present disclosure are described in detail below.

[0103] The embodiment of the present application provides a laser cutting device, which can include: a laser cutting system and an automatic sorting system of a laser cutting part in the following embodiments;

[0104] The laser cutting system moves the platform carrying the laser cutting piece to the area of the automatic sorting system after the laser cutting piece is processed, and the platform is used as a picking platform. The automatic sorting system of the laser cutting piece identifies and picks the laser cutting piece according to the automatic sorting method described in any of the embodiments.

[0105] The laser cutting system can be an existing laser cutting system, and the embodiment does not make any improvement to the laser cutting system.

[0106] After the cutting processing of the laser cutting system is completed, the processed workpiece is moved to the backplane area, and the automatic sorting method of the embodiment is executed in the backplane area. In actual application, the cutting processing of the frontplane and the automatic sorting process of the backplane are not interfered with each other and can work in parallel.

[0107] In addition, the industrial computer in the automatic sorting system of the laser cutting piece and the industrial computer of the laser cutting system can be two independent devices or one device. In the following embodiments, two independent devices are described.

[0108] The “to-be-picked workpiece”, “processed workpiece”, “workpiece”, and “laser cutting piece” mentioned in any of the following embodiments have the same meaning.

[0109] The “user” and “operator” mentioned in the following embodiments have the same meaning. The servo drive assembly in the following embodiments is the transmission structure / transmission system.

[0110] Embodiment one

[0111] As shown in Figure 1 In this embodiment, the automatic sorting system of the laser cutting piece is used to pick the to-be-picked workpiece in the backplane area, i.e., the picking area. The automatic controller (such as a PLC, programmable logic controller) of the automatic sorting system in this embodiment is connected with a separate control device such as a numerical control device. The numerical control device integrates an automatic sorting numerical control system, which can calculate the “picking area” according to the “nesting / laser cutting coordinate information” and divide the “picking area” into multiple “vision cells” and automatically number them, thereby providing a basis for “establishing a navigation map and planning a search path”.

[0112] In the specific implementation process, after the laser cutting processing is completed, the “exchange platform” action is performed to move the to-be-picked workpiece to the backplane area. At this time, the frontplane cutting processing and the backplane workpiece sorting are implemented in parallel.

[0113] The automatic sorting system of the laser cutting piece comprises an industrial computer integrated with an automatic sorting numerical control system, an automatic controller (such as a PLC), an image acquisition assembly and a first servo drive assembly; the first servo drive assembly is used to drive the image acquisition assembly to a specified position;

[0114] The industrial computer sends one or more instructions to the automatic controller according to the first information, so that the automatic controller controls the first transmission structure (such as the first servo drive assembly) to move the image acquisition assembly according to the one or more instructions, to collect images of the specified area in real time;

[0115] The automatic sorting numerical control system integrated in the industrial computer determines the position information of the laser cutting piece according to the first information and the images collected in real time; and the cutting piece picking assembly picks up the laser cutting piece according to the position information of the laser cutting piece.

[0116] Figure 1 The image acquisition assembly shown in the specification can include a light source sensing structure and a low-cost small-view industrial camera.

[0117] The first information described above is user input and / or information transmitted by the laser cutting system, and the first information can be basic information of the laser cutting piece, which includes one or more of the following: shape, coordinate origin, plate information of the plate to which the laser cutting piece belongs, estimated coordinate position, correspondence between the type of the laser cutting piece and the material receiving device, contour, model feature, cutting area information, etc. In this embodiment, the information transmitted by the laser cutting system can be accurate information calculated during cutting of the laser cutting system.

[0118] Generally, the cutting piece picking assembly is installed on the first servo drive assembly, so that the automatic controller can realize integrated control and operation. In other embodiments, the cutting piece picking assembly can also be installed on a second servo drive assembly, and the industrial computer controls the second servo drive assembly to drive the cutting piece picking assembly to pick up the laser cutting piece according to the position information of the laser cutting piece. The above can be selected independently, and both can better realize intelligent picking of small workpieces. The cutting piece picking assembly of this embodiment can include a valve island, and an open suction cup array composed of a plurality of suction nozzles that can be opened and closed under the control of the valve island. The suction combination in the open suction cup array can be controlled by programming.

[0119] In a specific implementation, the end of the cutting piece picking assembly can be provided with an open suction cup array structure, and after the position information of the laser cutting piece is determined, the serial numbers of the suction cups used in the movable suction cup array structure can be determined, at which time the suction nozzles of the serial numbers can be controlled to open and move to the position information of the laser cutting piece to realize suction picking, and then move to the corresponding material receiving device.

[0120] In practical applications, the industrial control computer and the automatic controller in this embodiment exchange information via the Modbus-TCP bus protocol; the automatic controller and the first servo drive component exchange information via the EtherCAT bus protocol; and the automatic controller controls the cutting component via the DMCNET bus. The first servo drive component here can be a first transmission structure.

[0121] To better illustrate the structure of the aforementioned automated sorting system, the first servo drive component and its related cooperating components are described in detail, for example... Figure 3 As shown, the picking platform 1 of the automatic sorting system is equipped with guide rails on both sides (such as...). Figure 3 The left Y-axis guide rail 2 and the right Y-axis guide rail 3 are located in the middle, and guide rails (such as those on the top of the picking platform 1) are also provided above the platform. Figure 3 The X-axis guide rail 5 is shown as being mounted on the crossbeam 4. Specifically, Y-axis guide rails are mounted on both sides of the picking platform 1, a crossbeam 4 spans across the top of the picking platform 1, and X-axis guide rails are mounted on the crossbeam 4. A Z-axis lead screw is also mounted on the crossbeam. Figure 3 The image shows a lifting shaft 9 that achieves Z-axis lifting and lowering with the help of auxiliary components 6 of the X-axis and Z-axis (all of which are components of the Z-axis lead screw), and a cutting part picking component is fixed on the Z-axis lead screw; an image acquisition component (such as a camera 8) is fixed on the Z-axis lead screw; then the automatic controller drives the Y-axis guide rail, X-axis guide rail and Z-axis lead screw to move in their respective directions with the help of the first servo drive component.

[0122] The image acquisition device and the cutting and picking component can move freely in the three directions of the X, Y, and Z axes by moving the X-axis, Y-axis, and Z-axis guide rails / lead screws. In this embodiment, the image acquisition device is also equipped with an auxiliary light source 7 to provide illumination during the acquisition process by the camera 8.

[0123] In this embodiment, the "small workpiece" can be a workpiece whose outline can be completely covered by the visual area. At the same time, the field of view of the industrial camera can completely cover the industrial outline of the small workpiece. In this embodiment, the industrial camera is a low-cost camera with a small field of view, which can achieve fast information transmission and fast recognition, realizing low-cost portable material picking.

[0124] Example 2

[0125] The automatic sorting CNC system integrated in the industrial control computer can be a software part implemented by a computer program. It can implement the automatic sorting method of laser-cut parts in the following embodiments. In the implementation process, it cooperates with the automatic controller to effectively and reasonably acquire the required real-time images, and then determine the position information of the laser-cut parts. Based on the position information of the laser-cut parts, the offset coordinate information of the cutting and picking component is determined, so that the automatic controller can control the movement of the cutting and picking component to quickly pick up the laser-cut parts and place them in the corresponding receiving device.

[0126] As Figure 2 shown in the drawings, the automatic sorting method of the laser cutting piece of the present embodiment can include the following steps:

[0127] S10, obtaining first information for identifying the laser cutting piece; the first information includes: shape, coordinate origin, plate information of the plate to which the laser cutting piece belongs, cutting parameter information, presumed coordinate position, type of the laser cutting piece and correspondence relationship with the receiving device. Part of the information in the first information is input by the user, and part of the information is transmitted by the laser cutting system.

[0128] For example, S10 includes sub-steps S11 to S13:

[0129] S11, the automatic sorting numerical control system receives the CAD workpiece drawing of the laser cutting piece, the basic parameters of the plate to be processed, and the type of the laser cutting piece input / imported by the user;

[0130] S12, the automatic sorting numerical control system receives the nesting file and / or laser cutting coordinate information transmitted by the laser cutting system;

[0131] S13, the automatic sorting numerical control system obtains the basic information of the laser cutting piece as the first information according to the CAD workpiece drawing, the type of the laser cutting piece, the nesting file and / or the laser cutting coordinate information.

[0132] Of course, when the type of the laser cutting piece is multiple types, and the receiving device is multiple, the automatic sorting numerical control system is also used to receive the correspondence relationship between the type of the laser cutting piece and the receiving device input by the user; thereby, it is convenient for the last cutting pickup assembly to pick up the laser cutting piece and then move to the corresponding receiving device.

[0133] S20, establishing a navigation map of the area to which the laser cutting piece belongs according to the first information, and determining a search strategy of the laser cutting piece based on the navigation map;

[0134] The navigation map includes: a visual map composed of a plurality of sequentially arranged visual cells; each visual cell has a unique serial number, and each visual cell is an area corresponding to the field of view of an image acquisition assembly;

[0135] The search strategy includes determining the step size and the search order according to the field of view cell.

[0136] For example, S20 includes sub-steps S21 to S23:

[0137] S21, the automatic sorting numerical control system obtains a picking area of the laser cutting piece according to the cutting parameter information, and the picking area is a certain area on the picking platform;

[0138] S22, according to the pick-up area and the field of view size of the image acquisition component, a navigation map composed of multiple visual cells is established;

[0139] S23, according to the position information of the visual cells in the navigation map and the size of the visual cells, a lateral movement step and a longitudinal movement step are determined to obtain a search strategy of the laser cutting piece.

[0140] For example, the size of the visual cell is 90cmx90cm, the lateral movement step and the longitudinal movement step are both 30cm; the diameter / length of the laser cutting piece is less than or equal to 60cm;

[0141] The size of the visual cell is 100cmx100cm, the lateral movement step and the longitudinal movement step are both 30cm; the diameter / length of the laser cutting piece is less than or equal to 70cm;

[0142] The size of the visual cell is 100cmx100cm, the lateral movement step and the longitudinal movement step are both 20cm; the diameter / length of the laser cutting piece is less than or equal to 80cm.

[0143] The above is only an example, in other embodiments, the size of the visual cell, the lateral movement step, the longitudinal movement step, etc. can be adjusted according to the actual size of the workpiece, and the present embodiment does not limit it, and the workpiece size and the above-mentioned do not correspond to it, which can also be adjusted according to the need, and all are within the scope of the present application.

[0144] S30, according to the search strategy and the navigation map, an image acquisition instruction is sent to the image acquisition component, and the position information of the laser cutting piece is identified according to the image feedback by the image acquisition component based on the image acquisition instruction, so that the cutting piece picking component sorts the laser cutting piece based on the position information of the laser cutting piece.

[0145] For example, according to the search strategy and the navigation map, a first image acquisition instruction can be sent to the image acquisition component, and a first real-time image returned by the image acquisition component can be obtained;

[0146] And based on the first information, whether there is a laser cutting piece in the first real-time image is identified, if there is, the position information of the laser cutting piece is identified, so that the cutting piece picking component sorts the laser cutting piece based on the position information of the laser cutting piece;

[0147] Otherwise, the Nth image acquisition instruction is sent to the image acquisition component according to the search strategy and the navigation map, so that the image acquisition component acquires an Nth real-time image based on the Nth image acquisition instruction and feeds back, based on the first information, whether the laser cutting piece exists in the Nth real-time image, if the laser cutting piece exists, the position information of the laser cutting piece is identified, so that the cutting piece picking component sorts the laser cutting piece based on the position information of the laser cutting piece.

[0148] Otherwise, the Nth image acquisition instruction is sent to the image acquisition component according to the search strategy and the navigation map, so that the image acquisition component acquires an Nth real-time image based on the Nth image acquisition instruction and feeds back, based on the first information, whether the laser cutting piece exists in the Nth real-time image, if the laser cutting piece exists, the position information of the laser cutting piece is identified, so that the cutting piece picking component sorts the laser cutting piece based on the position information of the laser cutting piece.

[0149] In addition, based on the first information, whether the laser cutting piece exists in the first real-time image can specifically include:

[0150] The first real-time image is binarized, and the binarized image is subjected to brightness conversion and histogram equalization to obtain an intermediate result image;

[0151] The intermediate result image is subjected to feature extraction, and whether the profile of the extracted feature matches the profile of the laser cutting piece in the first information is judged;

[0152] If matched, it is determined that the first real-time image exists the laser cutting piece, the type number of the laser cutting piece is determined, and the position information of the laser cutting piece as the grabbing center coordinate is obtained according to the information of the profile of the extracted feature.

[0153] Further, when the cutting piece picking component includes a suction disc array, the information of the suction nozzle array combination needed to be turned on in the cutting piece picking component is determined according to the information of the profile of the extracted feature; so as to control the suction disc array to sort the laser cutting piece based on the position information of the laser cutting piece.

[0154] The existing large field of view (the photo can cover the picking table surface) camera shooting mode applied to the metal sheet fine slit cutting contour recognition is greatly difficult to identify due to the limitations of observation distance, resolution and field of view angle. Even if the high-definition camera is extremely expensive, it is also difficult to see all the details on the entire picking table surface with one or a few photos. In the embodiment, a low-cost small field of view industrial camera can identify each region, and the movable identification is not limited by the observation distance, resolution and field of view angle, which ensures the identification accuracy and fast transmission, and ensures the identification speed.

[0155] Therefore, the method of the above embodiment can realize the method of dividing a larger area (an area needing to search for picking) into multiple "visual cells" (camera fields of view for close-range observation) and searching and analyzing one by one, that is, the cost is reduced, and the workpiece identification and picking are realized quickly and effectively.

[0156] Embodiment three

[0157] The software part of the automatic sorting system mainly includes: an automatic sorting numerical control system running on an industrial computer, and a real-time control program running in a PLC. The automatic sorting numerical control system and the laser cutting system are independent systems.

[0158] As shown in Figure 4 The automatic sorting numerical control system of the embodiment can include: a man-machine interaction processing module, a numerical control task management module, and a visual analysis processing module. In addition, the automatic sorting numerical control system can also include a local parameter database for storing user input information or information transmitted by the laser cutting system, etc. The modules here correspond to the methods described above, and the scheme of the embodiment is described in more detail from the perspective of the modules.

[0159] For the man-machine interaction processing module, the man-machine interaction processing module of the embodiment can include: a state display and control main interface, an automatic back parameter / zero seeking page, a parameter configuration page, a workpiece type and material receiving device pairing page, and an alarm information prompt page, etc.

[0160] Among them, the state display and control main interface has X-axis, Y-axis, and Z-axis coordinate position data, a CAD workpiece drawing import button / workpiece addition button, a start and stop control button, etc. Through the buttons of the state display and control main interface, the user can load the material file and the processing coordinate system information, so that the visual analysis processing module calculates the placement range of the workpiece on the picking platform, so as to plan the workpiece search path by the visual analysis processing module. For example, the user can click the workpiece addition button to add the CAD workpiece drawing, and the numerical control task management module can match the material receiving device number according to the automatic type number. In addition, there are also some control buttons on the control cabinet panel, which belong to the man-machine interaction processing part, which are set according to actual needs, and the embodiment does not limit them.

[0161] After each power-on start, the automatic sorting numerical control system needs the operator to control the execution of the routine zero search, find the reference point, and establish the coordinate system. Before each formal picking, the operator needs to load the CAD graph of the workpiece (workpiece contour graph) of this processing, the "nesting / laser cutting coordinate" information, and the plate material and thickness parameters into the automatic sorting numerical control system in advance. The CAD graph of the workpiece (workpiece contour graph) of this processing, the "nesting / laser cutting coordinate" information, and the plate material and thickness parameters can also be transmitted to the automatic sorting numerical control system through the laser cutting system. In practice, the operator can also set the corresponding material receiving device (by material receiving device number) of each type of workpiece (by type number) that needs to be picked in the "workpiece type and material receiving device pairing page", that is, input the corresponding relationship between the type of laser cutting piece and the material receiving device, that is, set the corresponding material receiving device number of different workpieces. Each material receiving device number corresponds to the coordinate position of a material receiving position area.

[0162] As the numerical control task management module of the numerical control kernel, it is responsible for the overall planning and management of the picking task. Specifically, it includes the following: at the beginning of task creation, according to the "nesting / laser cutting coordinate" information, the placement range of the workpiece on the material inspection platform (i.e. the picking area) is obtained, the navigation map for visual search is established, and the workpiece search path is planned. During the execution of the picking task, communicate with the PLC, and according to the need, issue the current required action instruction or instruction group to the PLC; according to the need, issue the photograph search instruction to the image acquisition component, and obtain the workpiece grabbing point coordinates, workpiece type, suction nozzle number, etc. information by means of the photographed image.

[0163] The numerical control task management module obtains the placement range of the workpiece on the material inspection platform (i.e. the picking area) according to the nesting / laser cutting coordinate information, establishes the navigation map for visual search, and plans the workpiece search path.

[0164] The numerical control task management module loads the processing area information provided by the user or the laser cutting machine, calculates which area of the entire platform the cut workpiece mainly exists in, i.e. the picking area. Calculate which "visual cell" on the picking platform is occupied by this area.

[0165] Combined Figures 5A to 5C The visual cell and the navigation map are described.

[0166] As shown in Figure 5A The numerical control task management module divides the entire picking platform into X1Y1-X8Y6 a total of 48 "visual cells"; according to the information transmitted by the laser cutting system (such as nesting / laser cutting coordinates), it is known that the placement range of the workpiece is Figure 5AThe middle black area), it is calculated that 20 "visual cells" of X4Y3-X8Y6 are needed to completely cover the workpiece placement area. That is, the 20 "visual cells" of X4Y3-X8Y6 need to search for workpieces. The numerical control task management module automatically numbers the "visual cell" that needs to search for the picking in the total picking path (such as Figure 5A N1-N20) as shown, so that the actual picking process always searches and picks one by one according to the number of "visual cells".

[0167] After the human-computer interaction processing module receives the correspondence between the workpiece type number and the receiving device number and the above-mentioned "visual cell" is determined, the search strategy can be determined according to the workpiece type.

[0168] In actual picking, there are workpieces "crossing" between two or even multiple adjacent "visual cells", such as Figure 5B As shown, the workpiece crosses between N2, N3, N6, and N7 "visual cells". Therefore, the actual search process is always "interlaced overlapping" scanning cells, that is, the horizontal and vertical movement steps in the search strategy can be smaller than the width and length of the visual cell. In actual application, the workpiece can be scanned by traversal, and there are multiple scanning processes in practice.

[0169] According to the above principle, assuming that the image range of each "visual cell" is 90cmx90cm, then moving search scanning with a step distance of 30cm each time can ensure that a workpiece of a maximum of 60cmx60cm is searched.

[0170] Assuming that the image range of each "visual cell" is 100cmx100cm, then moving search scanning with a step distance of 30cm each time can ensure that a workpiece of a maximum of 70cmx70cm is searched.

[0171] Assuming that the image range of each "visual cell" is 100cmx100cm, then moving search scanning with a step distance of 20cm each time can ensure that a workpiece of a maximum of 80cmx80cm is searched.

[0172] That is, the picking of the largest workpiece is related to the image range of the "visual cell" and the minimum movement step. In the human-computer interaction processing module, the search strategy can be manually freely configured, or it can be determined by the numerical control task management module with the help of workpiece contour information and the size of the visual cell. The above-mentioned search strategy and the configuration method of the visual cell in the navigation map enhance the flexibility of the system picking size.

[0173] In the case of no other system alarm signal, the numerical control task management module can start picking at any time in response to user operation.

[0174] For the visual analysis processing module, the visual analysis processing module in this embodiment can be referred to as a vision module. The hardware part corresponding to the vision module is mainly an intelligent light source system and an industrial camera used for machine vision.

[0175] The intelligent light source system is responsible for automatically detecting the intensity of ambient light and adjusting the intensity of the auxiliary light source according to the intensity of the ambient light. The industrial camera is responsible for automatically adjusting the focal length, shooting a digital image, and transmitting the digital image to the numerical control task management module.

[0176] The intelligent light source system and the industrial camera described above can form an image acquisition assembly, which can be installed on the Z-axis lead screw or driven by a separate servo drive assembly. Generally, to save costs and simplify the structure, it can be installed in the fixed area of the Z-axis lead screw and driven by the first servo drive assembly.

[0177] The vision module is mainly used to control shooting, receive image information, and analyze and process it. It uses machine vision technology to search for and determine the position information of the workpiece in the image, and then calculates the "offset coordinates" required for the cutting piece picking assembly to pick up the workpiece and the "suction nozzle array opening combination" required for the workpiece.

[0178] In the processing of the vision module described above, digital image processing is performed first, such as image binarization, brightness transformation, gray histogram, and histogram equalization processing. Then, feature extraction based on convolution operation is performed to extract edge, corner, and region features, analyze whether the contour features in the image match the established (workpiece contour map) mathematical model, and determine whether it is a workpiece contour according to the "matching degree".

[0179] When it is determined to be a workpiece contour, i.e., the workpiece is recognized, the color difference features inside and outside the contour are used to distinguish the to-be-picked workpiece from the remaining hollow area of the picked workpiece, and the specific coordinate position of the workpiece is determined. Finally, according to the characteristics of the workpiece and the suction nozzle array, the "suction nozzle array opening combination" required for the workpiece and the offset coordinates are determined, and the analysis results are fed back to the numerical control task management module.

[0180] Specifically, the vision module analyzes and processes the CAD workpiece drawing, the suction disc array feature parameters, the suction nozzle switching signal, and the image information. The output results include: whether there is a workpiece, if there is a workpiece, the output results also include: the workpiece type number, the workpiece grabbing point center coordinates, the suction nozzle array opening combination mode, etc.

[0181] In other embodiments, the vision module can also be used to estimate the mass of the workpiece according to the CAD workpiece drawing (contour map), the workpiece material, and the thickness parameters.

[0182] The automatic sorting numerical control system adopts the strategy of dividing "vision cell", and can "see clearly" the metal sheet slit cutting contour using a general industrial camera, thereby greatly reducing the hardware cost and recognition success rate of the vision scheme.

[0183] In addition, the image data of a relatively short distance and a small field of view is beneficial to rapid transmission and rapid analysis of the vision image, thereby facilitating fast-paced picking.

[0184] Due to the strategy of dividing "vision cell" and loading and analyzing "material set / laser cutting coordinate information", the numerical control system can automatically construct a "navigation map", plan a search strategy, and establish a feasible numerical control basis.

[0185] The receiving device of the automatic sorting system is open, and the position and size of the receiving device can be set relatively freely by the user, thereby improving the applicability of the device to different users and different laser cutting machines. For example, the "receiving device" set by the user can be a bucket, an inclined table, or the starting end of a conveyor belt, thereby improving the flexibility of arrangement.

[0186] Embodiment Four

[0187] The automatic sorting system of the laser cutting machine based on an actual structure, for example, an industrial computer, a PLC, an image acquisition assembly, a first servo drive assembly, and a workpiece picking assembly integrated with an automatic sorting numerical control system,

[0188] The first servo drive assembly is used to drive the image acquisition assembly to a specified position and drive the workpiece picking assembly to a picking position to realize picking of the workpiece. The industrial computer and the PLC exchange information through a Modbus-TCP bus protocol; the PLC and the first servo drive assembly exchange information through an EtherCAT bus protocol; and the PLC controls the workpiece picking assembly through a DMCNET bus mode.

[0189] The specific structure of the first servo drive assembly and the cooperating assembly are shown in Figure 3 The embodiment provides an operation method of the automatic sorting system, as shown in Figure 6 .

[0190] After the industrial computer is powered on, the automatic sorting system needs to be initialized, and after the initialization, the first step and the second step are executed. The initialization of the embodiment can specifically include that an operator controls to perform a conventional zero search, finds a reference point, and establishes a coordinate system.

[0191] In the implementation process, if the receiving workpiece picking device is multiple, such as according to the type number, the type of the workpiece to be picked can be set in the initialization, and the number of the receiving device corresponding to each type and other information, that is, the corresponding relationship between the type of the workpiece to be picked and the number of the receiving device is input in advance. Generally, each receiving device number corresponds to a receiving position area coordinate position, that is, the corresponding relationship at this place can be the corresponding relationship between the type and the belonging coordinate position, and then the PLC places the corresponding receiving position according to the corresponding relationship after picking the workpiece to be picked according to the picking instruction.

[0192] First task planning stage

[0193] 1.1 Load workpiece graph and establish workpiece model.

[0194] The user / laser cutting machine system provides the automatic sorting system with a "workpiece contour graph" (CAD workpiece graph) and "nesting / laser cutting coordinate" information. The vision module of the automatic sorting system establishes a "workpiece model" for visual recognition according to the features of the "workpiece contour graph".

[0195] 1.2 Workpiece and receiving device pairing.

[0196] The user needs to match each "workpiece number" (corresponding to "workpiece model") with "receiving device number" (corresponding to "receiving device coordinate position"). That is, determine which "receiving device" the "workpiece" needs to be placed in.

[0197] 1.3 Establish navigation map, plan search strategy and search path.

[0198] The automatic sorting system establishes a "navigation map" for visual search according to the "nesting / laser cutting coordinate" information. That is, the entire area that needs to be searched for materials is divided into N "visual cells" (camera shooting field of view area) that need to be visually searched and automatically numbered, ready to be searched and picked one by one.

[0199] Second step: specific picking method

[0200] The specific method for searching and picking N "visual cells" one by one:

[0201] 2.1 Calculate the coordinate position of the camera when shooting

[0202] First, the automatic sorting system calculates the coordinate position of the camera needed to shoot directly above the "visual cell" according to the coordinate position of the current "visual cell".

[0203] 2.2 Execute camera positioning

[0204] The automatic sorting system sends action instructions to the PLC, and the PLC and the transmission system, i.e., the first servo drive assembly, are responsible for sending the camera to the position required for shooting (e.g., lifting the mechanical hand to the translation height).

[0205] 2.3 Shoot, analyze, and feedback the search results.

[0206] After the camera is in place, the automatic sorting system is responsible for light source control, fine focus adjustment, shooting control, obtaining the image information shot by the camera, and analyzing whether there is a "workpiece" that meets the "workpiece model" characteristics in the image.

[0207] If there is a "workpiece" (not more than one optional), calculate the optimal workpiece grabbing position and the suction nozzle array opening combination required for grabbing the workpiece, and feed the grabbing position, suction nozzle array opening combination, and workpiece type information back to the automatic sorting system.

[0208] If there is no "workpiece", feed the "no workpiece" information back to the automatic sorting system, and the "search and pick sub-task" of the "vision cell" is completed, as shown in the flow of the pick sub-task. Figure 7

[0209] 2.4 Execute "mechanical hand to position".

[0210] If there is a workpiece, the automatic sorting system sends action instructions to the PLC, and the PLC and the first servo drive assembly are responsible for executing the "mechanical hand to position", including sending the cutting and picking assembly above the picking point in the X-axis and Y-axis directions, and dropping the cutting and picking assembly to the pre-picking height in the Z-axis direction. This process can be executed in X-axis, Y-axis, and Z-axis linkage.

[0211] 2.5 Execute the "lowering-suction-pred-lifting" action group

[0212] 2.5.1 Lowering: The PLC is responsible for controlling the cutting and picking assembly to slowly lower from the "pre-picking height" to the "picking height" in the Z-axis direction.

[0213] 2.5.2 Suction: The automatic sorting system controls the suction of the suction nozzle according to the "suction nozzle array opening combination" provided by the vision module, so that the cutting and picking assembly can suck the plate. (During the suction process, the suction pressure can be detected to determine whether the workpiece is "normally sucked").

[0214] 2.5.3 Pre-lifting: The automatic sorting system sends action instructions to the PLC, and the PLC is responsible for controlling the cutting and picking assembly to slowly lift from the "picking height" to the "pre-lifting height" in the Z-axis direction.

[0215] 2.6 Execute "putting to position".

[0216] ​PLC is responsible for sending the workpiece to the corresponding "material receiving device" above and stopping at the material placing height. In this process, the "workpiece" moves from the "picking point position" to the "material receiving device" directly above in the X-axis and Y-axis directions, and first lifts up (up to the translation height if the stroke is far enough) and then falls to the "material placing height" in the Z-axis direction.

[0217] 2.7 Execute "release workpiece", end the picking task of the workpiece.

[0218] The automatic sorting system controls the suction nozzle to stop inhaling, thereby releasing the workpiece.

[0219] 2.8 Repeat the above search for picking.

[0220] According to the "shooting offset position", return to step 2.2, and cycle the above until the visual search result is "no workpiece", then the "search for picking sub-task" of the "visual cell" ends, and the "search for picking sub-task" number is increased by 1, ready for the search task of the next "visual cell".

[0221] If the visual search result is "no workpiece", and the "visual cell" is the last one (the Nth), then the whole picking task is completed.

[0222] In addition, as Figure 7 mentioned above, the PLC performs "dropping detection" in real time during the execution of the above "material placing in position action group". If the PLC detects that a "dropping" event has occurred, it automatically stops the current action, marks the dropping position, notifies the data task management module of the automatic picking system, and triggers the "dropping picking" interpolation task.

[0223] It should be noted that in the description of the present specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, different embodiments or examples described in the present specification and the features of different embodiments or examples can be combined and combined by those skilled in the art without contradiction.

[0224] Although the preferred embodiments of the present application have been described, those skilled in the art can make further changes and modifications to these embodiments after learning the basic creative concept. Therefore, the claims should be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0225] It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application cover modifications and variations of this application provided they come within the scope of the appended claims and their equivalents.

Claims

1. An automatic sorting device for laser-cut parts, characterized in that, The automated sorting device is used for the intelligent picking of small workpieces and thin plates, including: The human-computer interaction processing module is used to acquire and identify the first information of the laser-cut part; the first information is the information calculated by the laser cutting system and transmitted by the laser cutting system. The CNC task management module is used to establish a navigation map of the area to which the laser-cut part belongs based on the first information, and to determine the search strategy for the laser-cut part based on the navigation map; according to the search strategy and the navigation map, it sends an image acquisition command to the image acquisition component so that the image acquisition component can acquire images based on the image acquisition command. The navigation map includes: a visual map composed of multiple sequentially arranged visual cells; each visual cell has a unique serial number, and each visual cell is the area of ​​the field of view of the corresponding image acquisition component; the search strategy includes determining the search step size and search order based on the visual cells; the small workpiece is a workpiece whose outline can be completely covered by the visual cells, and the field of view of the image acquisition component is a small field of view that completely covers the industrial outline of the small workpiece. The visual analysis and processing module is used to recognize the images acquired by the image acquisition component and obtain the position information of the laser-cut parts transmitted to the cutting part picking component, so that the cutting part picking component can sort the laser-cut parts based on the position information of the laser-cut parts.

2. The apparatus according to claim 1, characterized in that, The CNC task management module is used to send a first image acquisition command to the image acquisition component according to the search strategy and navigation map, and to obtain the first real-time image returned by the image acquisition component; The visual analysis and processing module is used to identify whether there is a laser-cut part in the first real-time image based on the first information. If there is, the module identifies the position information of the laser-cut part so that the cutting part picking component can sort the laser-cut part based on the position information of the laser-cut part. When no laser-cut part is present in the first real-time image, the CNC task management module sends the Nth image acquisition command to the image acquisition component according to the search strategy and navigation map, so that the image acquisition component can acquire the Nth real-time image based on the Nth image acquisition command. The visual analysis and processing module is used to identify whether there is a laser-cut part in the Nth real-time image based on the first information. If there is, the position information of the laser-cut part is identified so that the cutting part picking component can sort the laser-cut part based on the position information of the laser-cut part. Otherwise, the CNC task management module is also used to repeat the above-mentioned sending of the (N+1)th image acquisition command, where N is greater than or equal to 1; if no laser-cut parts are found after traversing the navigation map, an alarm signal is issued.

3. The apparatus according to claim 1, characterized in that, The human-computer interaction processing module is specifically used to receive the specified format workpiece drawing and the type of laser-cut part input / imported by the user; and to receive the nesting file and / or laser cutting coordinate information transmitted by the laser cutting system; and to obtain the basic information of the laser-cut part as the first information based on the specified format workpiece drawing, the type of laser-cut part, the nesting file and / or laser cutting coordinate information. The basic information includes one or more of the following: shape, coordinate origin, sheet material information of the laser-cut part, estimated coordinate position, correspondence between the type of laser-cut part and the receiving device, outline, model features, and cutting area information.

4. The apparatus according to claim 1, characterized in that, The CNC task management module is specifically used to obtain the picking area of ​​the laser-cut part based on the basic information of the laser-cut part, wherein the picking area is a certain area on the picking platform; Based on the picking area and the field of view of the image acquisition component, a navigation map composed of multiple visual cells is established; based on the location information and size of the visual cells in the navigation map, the lateral movement step size, longitudinal movement step size, and movement path are determined to obtain the search strategy for the laser-cut parts.

5. The apparatus according to claim 4, characterized in that, The visual cell measures 90cm x 90cm, with both the lateral and longitudinal movement steps being 30cm. The laser-cut parts have a diameter / length of 60cm or less. The visual cell is 100cm x 100cm in size, with a horizontal and vertical movement step of 30cm each; the laser-cut part has a diameter / length of 70cm or less. The visual cell is 100cm x 100cm in size, with a horizontal and vertical movement step of 20cm each; the laser-cut part is a workpiece with a diameter / length of 80cm or less.

6. The apparatus according to claim 2, characterized in that, Each image acquisition command includes: the center coordinates of the image area to be acquired; The visual analysis and processing module is specifically used for: The first real-time image is binarized, and the binarized image is then subjected to brightness conversion and histogram equalization to obtain an intermediate result image. Feature extraction is performed on the intermediate result image, and it is determined whether the contour of the extracted features matches the contour of the laser-cut part in the first information. If a match is found, it is determined that a laser-cut part exists in the first real-time image, the type number of the laser-cut part is determined, and the position information of the laser-cut part, which serves as the center coordinate of the grasping, is obtained based on the contour information of the extracted features.

7. The apparatus according to claim 6, characterized in that, The cutting part picking assembly includes: when the suction cup array is in operation, determining the information of the combination of suction nozzle arrays that need to be opened in the cutting part picking assembly based on the information of the extracted feature contour; so as to control the suction cup array to sort the laser-cut parts based on the position information of the laser-cut parts.

8. The apparatus according to claim 2, characterized in that, Also includes: The quality standard judgment module is used to judge whether the laser-cut parts meet the quality standards of the type of laser-cut parts based on real-time images of the laser-cut parts, and output the judgment result.

9. An automatic sorting system for laser-cut parts, characterized in that, The automatic sorting system is used to automatically sort laser-cut parts on the picking platform. The automatic sorting system includes: an industrial control computer with an integrated automatic sorting CNC system, an automatic controller, an image acquisition component, and a first transmission structure; the first transmission structure is used to move the image acquisition component to a designated position on the picking platform. The industrial control computer sends one or more instructions to the automatic controller according to the first information, so that the automatic controller controls the first transmission structure to move the image acquisition component according to the one or more instructions, so as to acquire images at a specified position in real time. The automatic sorting CNC system in the industrial computer determines the position information of the laser-cut part based on the first information and the real-time acquired image, so that the cutting part picking component can pick up the laser-cut part according to the position information of the laser-cut part. The first information is either user-input information or information transmitted by the laser cutting system; the automatic sorting CNC system executes the automatic sorting method for laser-cut parts as described in any one of claims 1 to 8.

10. The automatic sorting system for laser-cut parts according to claim 9, characterized in that, The end of the cutting component pickup assembly includes an open suction cup array structure with multiple suction nozzles; And / or, the industrial computer and the automatic controller exchange information via the Modbus-TCP bus protocol; The automatic controller and the first transmission structure exchange information via the EtherCAT bus protocol; The automatic controller controls the cutting part picking component via the DMCNET bus.

Images