A system and method for a collaborative robot to disorderly grab a plate member and nut projection weld

Abstract

The application belongs to the technical field of robots, and discloses a system and method for disordered grabbing of a plate member and nut projection welding of a collaborative robot, which comprises a workbench, a material box, a visual identification system and a robot linkage control system, the upper material box and the lower material box are arranged on the workbench, and the visual identification system is arranged opposite to the upper material box; the robot linkage control system comprises a collaborative robot, a programmable logic controller, a vertical spot welding machine and a nut conveyor, the programmable logic controller is arranged on a platform below the collaborative robot, the collaborative robot is provided with a grabbing execution mechanism and is arranged beside the vertical spot welding machine, the nut conveyor is arranged beside the vertical spot welding machine, and the visual identification system, the collaborative robot, the vertical spot welding machine and the nut conveyor are electrically connected with the programmable logic controller. The whole-link optimization of the application realizes the automation of the disordered grabbing of the plate member with the nut hole and the nut projection welding of the commercial vehicle welding line, and has the advantages of low cost, high flexibility and high stability.

Description

Technical Field

[0001] This invention belongs to the field of robotics technology, specifically relating to a system and method for a collaborative robot to randomly grasp plates and perform nut projection welding. Background Technology

[0002] In the commercial vehicle manufacturing sector, body-in-white welding is one of the core processes in vehicle production, and its level of automation directly determines the overall production efficiency, product quality, and production costs. Currently, the overall automation level of commercial vehicle body-in-white welding workshops has reached a high level, with most core processes in the production process being automated. Manual intervention is mainly concentrated in the sheet metal loading process and the welding operations of some special components.

[0003] Currently, mainstream OEMs' welding workshops have successfully implemented automated loading, unloading, and projection welding operations for medium and large-sized sheet metal parts based on vision-guided technology. This has effectively improved the processing efficiency and welding quality of medium and large-sized sheet metal parts and reduced the intensity of manual labor. However, for small and medium-sized sheet metal parts (such as those with nut holes), due to their inherent characteristics such as a wide variety of sheet metal types, diverse sizes and specifications, and significant differences in structural forms, existing automation technologies are difficult to adapt efficiently. Therefore, the loading, unloading, and projection welding operations for these types of sheet metal parts are still generally completed manually.

[0004] When manually handling small and medium-sized sheet metal parts with nuts and holes, manual picking, handling, and hand-held projection welding are required. This is not only labor-intensive and cumbersome, leading to low production efficiency and failing to meet the demands of large-scale, high-speed production in welding workshops, but also susceptible to subjective factors such as operator skill level, work condition, and fatigue. This results in uneven weld strength, incomplete welds, and missed welds, making it difficult to guarantee consistent weld quality and ultimately affecting the overall structural strength and safety of the vehicle body. Furthermore, hand-held projection welding requires close contact with the welding area, posing safety risks such as high temperatures, spatter, and electromagnetic radiation, potentially leading to accidents. Therefore, achieving automated clamping and welding of small and medium-sized sheet metal parts with nuts and holes is crucial for commercial vehicle body-in-white welding workshops to overcome automation bottlenecks and achieve reduced-management or even unmanned production. It is also an inevitable trend for industry technological upgrading.

[0005] With the continuous development of industrial automation technology, domestic and foreign vehicle manufacturers and related research institutions have begun to focus on the research and promotion of automated welding technology for small and medium-sized sheet metal parts, attempting to solve the aforementioned industry pain points through technological innovation. For example, Chinese utility model patent with publication number CN214161728U discloses an automatic nut welding system. This system achieves automatic nut welding operations by adding multiple sensors to form an automatic control system, thereby reducing manual intervention to a certain extent.

[0006] However, this existing technology still has many shortcomings in actual industrial applications, making it difficult to meet the production needs of commercial vehicle welding workshops for multi-variety, small-batch, and disordered incoming materials. Specific defects are as follows: First, the technology does not conduct in-depth research and description of the incoming material state, requiring pre-sorting and positioning of the materials. It has extremely poor adaptability to disordered stacking and randomly positioned incoming material environments, and cannot adapt to the current situation of disordered placement of small and medium-sized plates in actual production. Second, the technology uses a general-purpose robotic arm as the gripping device, without specific design for the size differences and structural characteristics of small and medium-sized plates with nut holes. This results in insufficient targeting of the gripping of plates, and the gripping process is prone to errors. Issues such as slippage and deviation indicate a lack of gripping stability; thirdly, the technology's visual recognition and path planning have significant limitations, relying on fixed conveying paths and positioning pins to achieve nut positioning, which cannot handle scenarios where the plate's posture is random and its position is not fixed, resulting in a narrow range of applications; fourthly, the system lacks linkage and flexibility, failing to form an integrated operation process from disordered gripping and precise positioning to automatic welding, and the robotic arm used is bulky and has low flexibility, making it difficult to adapt to the welding needs of different nut positions for various sizes of parts, and the operational flexibility and production efficiency still need to be improved, failing to fundamentally solve the industry pain points of automated welding of small and medium-sized plates.

[0007] In summary, current automated welding technology for small and medium-sized nut-hole plate parts in commercial vehicle body-in-white welding workshops still suffers from numerous shortcomings. Existing technologies cannot effectively address issues such as insufficient adaptability to disordered environments, lack of targeted and stable grasping, limitations in visual recognition and path planning, and insufficient system linkage and flexibility, thus hindering the automation upgrade process of welding workshops. Therefore, developing an automated welding method for small and medium-sized nut-hole plate parts that can adapt to disordered incoming material environments, provide stable and accurate grasping, offer flexible visual recognition and path planning, and have strong system linkage has become a pressing technical problem that needs to be solved by those skilled in the art. Summary of the Invention

[0008] The purpose of this invention is to provide a system and method for collaborative robots to randomly grasp plates and perform nut projection welding. Through the optimization of the entire process of vision, grasping, linkage and material frame, the system realizes the automation of random grasping and nut projection welding of plates with nut holes in commercial vehicle welding lines. It provides a feasible solution for reduced-management production in welding workshops, and has the advantages of low cost, high flexibility and strong stability.

[0009] The specific details of the plan are as follows:

[0010] A collaborative robot system for disordered grasping of sheet metal parts and projection welding of nuts includes a worktable, a material bin, a vision recognition system, and a robot linkage control system. The material bin includes an upper and lower bin, which are respectively set on the worktable for placing sheet metal parts before and after welding. The vision recognition system is positioned directly opposite the upper bin to acquire the three-dimensional pose information of the sheet metal parts to be welded in the upper bin. The robot linkage control system includes a collaborative robot, a programmable logic controller (PLC), a vertical spot welder, and a nut conveyor. The PLC is located on a platform below the collaborative robot. The end effector of the collaborative robot is equipped with a grasping actuator that can rotate to the welding positions of the material bin and the vertical spot welder to grasp and transfer the sheet metal parts to be welded. The vertical spot welder is used for projection welding of the sheet metal parts and nuts. The collaborative robot is positioned next to the vertical spot welder, and the nut conveyor is positioned next to the vertical spot welder to automatically configure the nuts for welding. The vision recognition system, collaborative robot, vertical spot welder, and nut conveyor are all electrically connected to the PLC.

[0011] In the collaborative robot system for disordered grasping of sheet metal and nut projection welding of the present invention, the vision recognition system includes a 3D camera and a camera mounting bracket. The camera mounting bracket is fixed next to the workbench, and the 3D camera is mounted on the camera mounting bracket, located above the loading bin. It is used to capture the three-dimensional pose information of the sheet metal to be welded in the loading bin. The contour features of the target sheet metal are extracted through point cloud region segmentation. The camera identifies the sheet metal model, and the programmable logic controller (PLC) calls the corresponding program. The orientation is confirmed; if the sheet metal is placed in the wrong orientation, manual processing is paused; if the sheet metal is placed in the right orientation, the three-dimensional coordinates of the grasping point are calculated, prioritizing the grasping of the higher-stacking sheet metal to avoid interference and collisions in subsequent grasping, and the coordinate data is sent to the PLC. The PLC of the robot linkage control system is equipped with a PN communication expansion module to enable communication between the 3D camera, 2D camera, and collaborative robot, ensuring real-time transmission of the pose data of the vision recognition (such as the three-dimensional coordinates of the grasping point). The vision recognition system, collaborative robot, vertical spot welding machine, and nut conveyor are all electrically connected to the PLC, and all transmit signals via communication. The visual recognition system also includes a 2D camera. If high precision is required for the board, secondary precise positioning is performed: the board is moved to the 2D camera area for image recognition, with a focus on identifying the nut holes. The collaborative robot's end effector is equipped with a gripping actuator for gripping the board, including electromagnets, suction cups, pneumatic magnets, grippers, or electro-permanent magnets. This invention can be matched with the corresponding gripping actuator based on the collaborative robot's effective load. For a collaborative robot with an effective load of 3-15kg: when the board is 0-3kg and the projection weld position is relatively regular, 1-2 electromagnets are used; for irregular positions, some contouring positioning blocks can be used for gripping; when the board is 3-8kg, suction cups can be used; when the board is 8-15kg, magnets or suction cups can be used when the boards are stacked, and grippers can be used when the boards are not stacked. This invention considers electromagnets to be the optimal gripping tool, addressing the following optimization issues: a 0.5mm thin film is wrapped around the magnet surface or a custom Φ3mm pin is installed at the bottom to eliminate the adhesion of workpieces caused by residual magnetism 1-2 seconds after demagnetization; for workpieces prone to eccentricity, dual magnets are used for symmetrical gripping to avoid unilateral gripping deviation. The collaborative robot of this invention has a customized program written according to the welding requirements of different workpieces. A PLC enables the coordinated control of the collaborative robot, vision system, and welding machine: after receiving workpiece pose data from the vision system, the robot automatically adjusts its gripping trajectory; upon reaching the positioning pin, it triggers the welding machine start signal; after projection welding is completed, the robot automatically moves the workpiece to the unloading area, while the vision system begins to identify the next batch of disordered workpieces, achieving seamless integration of disordered gripping, positioning, projection welding, and unloading. A nut conveyor is located next to the vertical spot welding machine for automatically dispensing welding nuts.A preferred embodiment of this invention involves specifying that the same batch of plates should be welded with nuts of the same specification when setting welding tasks. A nut conveyor supplies nuts to the vertical spot welding machine according to a pre-set program, thus eliminating the possibility of welding nuts of the wrong specification and saving on nut inspection. The safety system of this invention also includes safety light curtains and safety fences, which are respectively installed around the perimeter of the welding area. If personnel accidentally enter the work area, the machine will stop immediately to prevent accidents.

[0012] Furthermore, the visual recognition system includes a 3D camera and a camera mounting bracket. The camera mounting bracket is fixed next to the workbench, and the 3D camera is fixedly connected to the camera mounting bracket and located directly above the loading box, so as to take pictures to obtain the three-dimensional pose information of the plates to be welded in the loading box.

[0013] Furthermore, the 3D camera in the vision recognition system can identify the board model by comparing features in the feature database, and the programmable logic controller can automatically call the robot motion path and welding parameters according to the identified board model.

[0014] Furthermore, the visual recognition system also includes a 2D camera, which is fixed on the worktable to assist in precise positioning and to identify the nut holes on the board.

[0015] Furthermore, the grasping actuators include electromagnets, suction cups, pneumatic magnets, grippers, or electro-permanent magnets.

[0016] Furthermore, the height of the material bin is 200mm-300mm to effectively avoid interference during the gripping process.

[0017] For scenarios involving disordered stacking along the production line, this invention preferably uses a low-profile feeding box with a height between 200mm and 300mm. Compared to a high-profile feeding box (500mm in height), which has a 95% interference-free gripping rate, the low-profile feeding box has a 100% interference-free gripping rate. In actual production, the length and width of the feeding box can be increased to improve the number of boards that can be placed at one time.

[0018] Furthermore, it also includes a safety system, which includes safety light curtains and safety fences, installed around the perimeter of the work area to prevent accidental entry of personnel into the work area.

[0019] A method for a collaborative robot to randomly grasp and weld nuts onto sheet metal parts, applied to the system of the collaborative robot to randomly grasp and weld nuts onto sheet metal parts, includes the following steps:

[0020] S1. The worker first places the sheet metal into the material box on the workbench, resets the safety light curtain, starts the safety system, and prepares for the operation.

[0021] The S2 and 3D cameras take pictures of the boards in the loading box and identify them. The contour features of the target boards are extracted by dividing the point cloud region. The camera identifies the board model and the programmable logic controller calls the corresponding program. The orientation is confirmed. If the board is placed in the wrong orientation, manual processing is paused. If the board is placed in the right orientation, the three-dimensional coordinates of the gripping point are calculated. The boards with higher stacks are gripped first, and the coordinate data is sent to the programmable logic controller.

[0022] S3. The programmable logic controller sends grasping instructions to the collaborative robot, plans the path, determines whether the path is in a singularity area, and adjusts the material box or grasping path according to the position; the collaborative robot matches the corresponding grasping actuator according to the board load.

[0023] S4. Determine if the grabbing was successful. If the grabbing failed, proceed to step S2. If the grabbing was successful, determine the precision requirements of the board. If the board has high precision requirements, perform a second precise positioning. If the board has low precision requirements, proceed to step S5.

[0024] S5. The programmable logic controller (PLC) allows the collaborative robot to deliver the workpiece to the welding station of the vertical spot welder. Based on the process information, the PLC performs the corresponding projection welding operation. After the workpiece is in place, the PLC sends a signal to the vertical spot welder and the nut conveyor to execute the projection welding. During the projection welding process, current and pressure data are fed back in real time. If an abnormality occurs, the operation stops and an alarm is triggered. The worker checks whether the projection welding operation is completed. If it is not completed, it is re-welded. If the welding is completed, the collaborative robot places the welded workpiece into the unloading box.

[0025] S6. Repeat the task until the work is finished. If the material box is full, replace it.

[0026] Furthermore, in step S3, the corresponding gripping execution mechanism is matched according to the effective load of the collaborative robot. The effective load of the collaborative robot is 3-15kg: when the plate is 0-3kg and the projection weld position is relatively regular, 1-2 electromagnets are used. When it is irregular, some contour positioning blocks can be used for gripping. When the plate is 3-8kg, a suction cup can be used for gripping. When the plate is 8-15kg, when the plates are stacked, magnets or suction cups can be used for gripping. When the plates are not stacked, grippers can be used for gripping.

[0027] Furthermore, in step S4, the secondary precise positioning involves moving the plate to the 2D camera area for image recognition and precise positioning of the nut hole.

[0028] Compared with the prior art, the present invention has the following advantages:

[0029] This invention improves adaptability to disordered environments, enhances the targeting and stability of grasping, breaks through the limitations of visual recognition and path planning, and improves system linkage and flexibility. Through the optimization of the entire process of vision, grasping, linkage, and material frame, it realizes the automation of disordered grasping and nut projection welding of nut-hole plate parts in commercial vehicle welding lines, providing a feasible solution for reduced-management production in welding workshops, while also having the advantages of low cost, high flexibility, and strong stability. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the overall system of the collaborative robot of the present invention, which randomly grasps plates and welds nuts.

[0031] Figure 2 This is a flowchart illustrating the system method for the collaborative robot to randomly grasp plates and perform projection welding of nuts according to the present invention.

[0032] In the picture:

[0033] 1. Workbench; 2. Material bin; 2.1. Loading bin; 2.2. Unloading bin; 3. Vision recognition system; 3.1. 3D camera; 3.2. Camera mounting bracket; 3.3. 2D camera; 4. Robot linkage control system; 4.1. Collaborative robot; 4.1.1. Gripping actuator; 4.2. Programmable logic controller; 4.3. Vertical spot welding machine; 4.4. Nut conveyor; 5. Safety system; 5.1. Safety light curtain; 5.2. Safety fence. Detailed Implementation

[0034] To make the technical problems solved by the present invention, the technical solutions adopted, and the technical effects achieved clearer, the technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention. Furthermore, it should be noted that, for ease of description, only the parts related to the present invention are shown in the accompanying drawings, not all of them.

[0035] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. The terms "first position" and "second position" refer to two different positions.

[0036] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections or detachable connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal connections between two components. Those skilled in the art can understand the specific meaning of these terms in this invention based on the specific circumstances.

[0037] The following combination Figure 1 and Figure 2 The present invention will be described in conjunction with the embodiments:

[0038] A system for collaborative robots to randomly grasp plates and perform projection welding on nuts, see [link to system]. Figure 1 As shown, the system includes a workbench 1, a material bin 2, a vision recognition system 3, and a robot linkage control system 4. The material bin 2 includes a loading bin 2.1 and a unloading bin 2.2, which are respectively mounted on the workbench 1 to hold the plates before and after welding. The vision recognition system 3 is positioned directly opposite the loading bin 2.1 to acquire the three-dimensional pose information of the plates to be welded within the loading bin 2.1. The robot linkage control system 4 includes a collaborative robot 4.1, a programmable logic controller 4.2, a vertical spot welding machine 4.3, and a nut conveyor 4.4. The programmable logic controller 4.2 is mounted on the collaborative robot 4.1. In the platform below, the end effector of the collaborative robot 4.1 is equipped with a gripping actuator 4.11. The gripping actuator 4.11 of the collaborative robot 4.1 can rotate to the welding position of the material box and the vertical spot welding machine 4.3 for gripping and transferring the plates to be welded. The vertical spot welding machine 4.3 is used for projection welding of the plates and nuts. The collaborative robot 4.1 is set next to the vertical spot welding machine 4.3. The nut conveyor 4.4 is set next to the vertical spot welding machine 4.3 for automatically configuring the nuts for welding. The vision recognition system 3, the collaborative robot 4.1, the vertical spot welding machine 4.3, and the nut conveyor 4.4 are all electrically connected to the programmable logic controller 4.2.

[0039] The visual recognition system 3 includes a 3D camera 3.1 and a camera mounting bracket 3.2. The camera mounting bracket 3.2 is fixed next to the workbench 1, and the 3D camera 3.1 is fixedly connected to the camera mounting bracket 3.2 and located directly above the loading box 2.1. It is used to acquire the three-dimensional pose information of the plates to be welded in the loading box 2.1.

[0040] The 3D camera of the vision recognition system 3.1 can identify the board model by comparing features in the feature database, and the programmable logic controller 4.2 can automatically call the robot motion path and welding parameters according to the identified board model.

[0041] The visual recognition system 3 also includes a 2D camera 3.3, which is fixed on the worktable 1 to assist in precise positioning and to identify the nut hole positions of the plate.

[0042] The grasping actuator 4.11 includes an electromagnet, a suction cup, a pneumatically controlled magnet, a gripper, or an electro-permanent magnet.

[0043] The height of the material bin is 200mm-300mm, which effectively avoids interference during the gripping process.

[0044] It also includes a safety system 5, which includes a safety light curtain 5.1 and a safety fence 5.2. The safety light curtain 5.1 and the safety fence 5.2 are respectively installed around the gripping welding area to prevent personnel from accidentally entering the work area.

[0045] Example 2:

[0046] A method for collaborative robots to randomly grasp sheet metal parts and perform projection welding on nuts, see [link to relevant documentation]. Figure 2 As shown, the system applied to the collaborative robot for disordered grasping of plates and projection welding of nuts includes the following steps:

[0047] S1. The worker first puts the board into the loading box 2.1 of the workbench 1, resets the safety light curtain 5.1, starts the safety system 5, and prepares for the operation.

[0048] S2 and 3D camera 3.1 take pictures of the boards in the feeding box 2.1 for recognition. The contour features of the target board are extracted by dividing the point cloud region. The camera identifies the board model and the programmable logic controller 4.2 calls the corresponding program. The orientation is confirmed. If the board is placed in the wrong position, manual processing is paused. If the board is placed in the right position, the three-dimensional coordinates of the gripping point are calculated. The board with the higher stack is gripped first, and the coordinate data is sent to the programmable logic controller 4.2.

[0049] S3, the programmable logic controller 4.2 sends grasping instructions to the collaborative robot 4.1, plans the path, determines whether the path is in a singularity area, and adjusts the material box or grasping path according to the position; the collaborative robot 4.1 matches the corresponding grasping actuator 4.11 according to the board load;

[0050] S4. Determine if the grabbing was successful. If the grabbing failed, proceed to step S2. If the grabbing was successful, determine the precision requirements of the board. If the board has high precision requirements, perform a second precise positioning. If the board has low precision requirements, proceed to step S5.

[0051] S5, Programmable Logic Controller 4.2 allows the collaborative robot 4.1 to deliver the workpiece to the welding station of the vertical spot welding machine. Based on the process information, it performs the corresponding projection welding operation. After the workpiece is in place, the programmable logic controller 4.2 sends a signal to the vertical spot welding machine and the nut conveyor 4.4 to execute the projection welding. During the projection welding process, current and pressure data are fed back in real time. If an abnormality occurs, the operation stops and an alarm is triggered. The worker checks whether the projection welding operation is completed; any incomplete workpieces are re-welded. If the welding is complete, the collaborative robot 4.1 places the welded workpiece into the unloading bin 2.2.

[0052] S6. Repeat the next task. If the material box 2.2 is full, replace it until the work is finished.

[0053] In step S3, the corresponding gripping actuator 4.11 is matched according to the effective load of the collaborative robot 4.1. The effective load of the collaborative robot 4.1 is 3-15kg: when the plate is 0-3kg and the projection weld position is relatively regular, 1-2 electromagnets are used. When it is irregular, some contour positioning blocks can be used for gripping. When the plate is 3-8kg, a suction cup can be used for gripping. When the plate is 8-15kg, when the plates are stacked, magnets or suction cups can be used for gripping. When the plates are not stacked, grippers can be used for gripping.

[0054] In step S4, the secondary precise positioning involves moving the plate to the area of ​​the 2D camera 3.3 for image capture and recognition, and then precisely positioning the nut hole.

[0055] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A system for a collaborative robot to randomly grasp plates and perform projection welding on nuts, characterized in that, The system includes a workbench (1), a material bin (2), a vision recognition system (3), and a robot linkage control system (4). The material bin (2) includes a loading bin (2.1) and a unloading bin (2.2), which are respectively located on the workbench (1) for placing the plates before and after welding. The vision recognition system (3) is located directly opposite the loading bin (2.1) and is used to acquire the three-dimensional pose information of the plates to be welded in the loading bin (2.1). The robot linkage control system (4) includes a collaborative robot (4.1), a programmable logic controller (4.2), a vertical spot welding machine (4.3), and a nut conveyor (4.4). The programmable logic controller (4.2) is located on the collaborative robot (4.1). 4.1) In the platform below, the end of the collaborative robot (4.1) is provided with a gripping execution mechanism (4.11). The gripping execution mechanism (4.11) of the collaborative robot (4.1) can rotate to the welding position of the material box and the vertical spot welding machine (4.3) for gripping and transferring the plate to be welded. The vertical spot welding machine (4.3) is used for projection welding of the plate and the nut. The collaborative robot (4.1) is set next to the vertical spot welding machine (4.3). The nut conveyor (4.4) is set next to the vertical spot welding machine (4.3) for automatically configuring the nut for welding. The vision recognition system (3), the collaborative robot (4.1), the vertical spot welding machine (4.3), and the nut conveyor (4.4) are electrically connected to the programmable logic controller (4.2).

2. The system for collaborative robots to disorderly grasp plates and weld nuts according to claim 1, characterized in that, The visual recognition system (3) includes a 3D camera (3.1) and a camera mounting bracket (3.2). The camera mounting bracket (3.2) is fixed next to the workbench (1). The 3D camera (3.1) is fixedly connected to the camera mounting bracket (3.2) and located directly above the loading box (2.1) to obtain the three-dimensional pose information of the plate to be welded in the loading box (2.1).

3. The system for collaborative robots to disorderly grasp plates and weld nuts according to claim 2, characterized in that, The 3D camera (3.1) of the vision recognition system can identify the board model by comparing features in the feature database, and the programmable logic controller (4.2) can automatically call the robot motion path and welding parameters according to the identified board model.

4. The system for collaborative robots to randomly grasp plates and weld nuts according to claim 3, characterized in that, The visual recognition system (3) also includes a 2D camera (3.3), which is fixed on the worktable (1) to assist in precise positioning and to identify the nut hole position of the plate.

5. The system for collaborative robots to randomly grasp plates and weld nuts according to claim 1, characterized in that, The gripping actuator (4.11) includes an electromagnet, a suction cup, a pneumatically controlled magnet, a gripper, or an electro-permanent magnet.

6. The system for collaborative robots to disorderly grasp plates and weld nuts according to claim 1, characterized in that, The height of the material box is 200mm-300mm, which is used to effectively avoid interference during the gripping process.

7. The system for collaborative robots to disorderly grasp plates and weld nuts according to claim 1, characterized in that, It also includes a safety system (5), which includes a safety light curtain (5.1) and a safety fence (5.2). The safety light curtain (5.1) and the safety fence (5.2) are respectively installed around the work area to prevent personnel from accidentally entering the work area.

8. A method for a collaborative robot to randomly grasp plates and perform projection welding on nuts, characterized in that, The system for collaborative robots to randomly grasp plates and perform projection welding of nuts, as described in any one of claims 1-7, comprises the following steps: S1. The worker first puts the board into the workbench (1) and the loading box (2.1), resets the safety light curtain (5.1), starts the safety system (5), and prepares for the operation; S2 and 3D cameras (3.1) take pictures of the boards in the loading box (2.1) for identification. The contour features of the target boards are extracted by dividing the point cloud region. The camera identifies the board model and the programmable logic controller (4.2) calls the corresponding program. The orientation is confirmed. If the board is placed in the opposite direction, manual processing is paused. If the board is placed in the correct direction, the three-dimensional coordinates of the gripping point are calculated. The boards with higher stacks are gripped first, and the coordinate data is sent to the programmable logic controller (4.2). S3, the programmable logic controller (4.2) sends a grasping command to the collaborative robot (4.1), plans a path, determines whether the path is in a singularity area, and adjusts the bin or grasping path according to the position; the collaborative robot (4.1) matches the corresponding grasping actuator (4.11) according to the board load. S4. Determine if the grabbing was successful. If the grabbing failed, proceed to step S2. If the grabbing was successful, determine the precision requirements of the board. If the board has high precision requirements, perform a second precise positioning. If the board has low precision requirements, proceed to step S5. S5. The programmable logic controller (4.2) allows the collaborative robot (4.1) to deliver the board to the welding station of the vertical spot welding machine and perform the corresponding projection welding operation according to the process information. After the board is in place, the programmable logic controller (4.2) sends a signal to the vertical spot welding machine and the nut conveyor (4.4) to perform projection welding. During the projection welding process, the current and pressure data are fed back in real time. If an abnormality occurs, the operation will stop and an alarm will be triggered. The worker checks whether the projection welding operation is completed. If it is not completed, it will be repaired. If the welding is completed, the collaborative robot (4.1) places the welded board in the unloading box (2.2). S6. Repeat the next task. If the material box (2.2) is full, replace it until the work is finished.

9. The method for collaborative robots to disorderly grasp plates and perform projection welding of nuts according to claim 8, characterized in that, In step S3, the corresponding gripping execution mechanism (4.11) is matched according to the effective load of the collaborative robot (4.1). The effective load of the collaborative robot (4.1) is 3-15kg: when the plate is 0-3kg and the projection weld position is relatively regular, 1-2 electromagnets are used. When it is irregular, some contour positioning blocks can be used for gripping. When the plate is 3-8kg, a suction cup can be used for gripping. When the plate is 8-15kg, when the plates are stacked, magnets or suction cups can be used for gripping. When the plates are not stacked, grippers can be used for gripping.

10. The method for a collaborative robot to disorderly grasp plates and perform projection welding of nuts according to claim 8, characterized in that, In step S4, the secondary precise positioning involves moving the plate to the area of ​​the 2D camera (3.3) for photographic recognition and precise positioning of the nut hole.

Images