A vision-based automated robotic marking workstation

By integrating laser marking and pneumatic marking into a shared gripper at the robot's end effector, and combining 2D vision positioning and laser rust removal closed-loop preprocessing, the problems of poor flexibility and low automation level of metal workpiece marking solutions have been solved, achieving efficient and accurate marking and data traceability for various types of workpieces.

CN123103199APending Publication Date: 2026-11-17CHINA NUCLEAR IND FIFTH CONSTR CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202611383902.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-08
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing metal workpiece marking solutions suffer from problems such as poor flexibility and adaptability, low level of automation, insufficient positioning accuracy, single marking mode, limited working range, and lack of data traceability. Traditional manual marking relies on manual positioning and parameter adjustment, resulting in low efficiency and poor consistency. Fixed workstations and teaching-style marking are cumbersome to change types. Single laser or pneumatic modes cannot meet the needs of fine marking and deep imprinting. Rust removal and marking are independent of each other and cannot form a closed loop.

Method used

An automated robot marking workstation based on vision positioning is adopted, which integrates laser marking and pneumatic marking modes into a shared gripper at the robot end. Combined with 2D vision positioning and laser rust removal closed-loop pretreatment, it achieves fully automatic, high-precision, and flexible marking.

Benefits of technology

It has improved the flexibility and adaptability of various types of workpieces, significantly shortened the changeover time, improved production efficiency, and ensured stable and reliable marking accuracy and quality. It is applicable to a wide range of scenarios, has improved system collaboration and data traceability, and increased equipment utilization.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123103199A_ABST
    Figure CN123103199A_ABST
Patent Text Reader

Abstract

This invention discloses an automated robotic marking workstation based on vision positioning. The workstation includes a robot system, a ground rail walking system, a 2D vision positioning system, a dual-mode marking system, an electrical control system, a safety protection system, and a host computer data system. The ground rail walking system drives the industrial robot to move back and forth along the ground rail to cover long workpieces. The 2D vision positioning system identifies the workpiece edge or weld seam positioning reference through a 2D camera installed at the robot's end effector. Combined with the focal length of the laser marking lens, the spatial coordinates of the pneumatic marking head, and the reciprocating stroke, it automatically calculates the robot trajectory, marking position, and posture to achieve teachless adaptive positioning. The dual-mode marking system integrates the laser and pneumatic marking mechanisms on the same carrier through a shared gripper, enabling automatic switching between the two modes. The PLC coordinates the timing control, and the host computer collects and uploads the marking data to the manufacturing execution system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of industrial automation processing technology, and in particular to an automated robot marking workstation based on vision positioning. It is especially suitable for automated marking of metal workpieces such as pipes, profiles, and supports before and after welding. It belongs to the field of integrated equipment technology of robot control, machine vision positioning, laser marking and pneumatic marking, and industrial data interaction. Background Technology

[0002] In the manufacturing process of metal pipes, profiles, and welded workpieces, marking and labeling are crucial for quality traceability and material management. Currently, the industrial sector primarily employs the following technical solutions for marking and labeling these workpieces: The first method is manual marking with a handheld marking machine. This method relies on the operator to mark each workpiece individually with the handheld marking machine. The marking position is determined by visual inspection, and the marking parameters are adjusted manually based on experience. The marking quality is greatly affected by the operator's skill level and fatigue level, resulting in problems such as low positioning accuracy, poor consistency, high labor intensity, and low efficiency. Furthermore, it is impossible to generate marking data records, making it difficult to meet the requirements of mass production and quality traceability.

[0003] The second method involves using a dedicated fixture in conjunction with a fixed-station marking machine. This method uses a customized chemical fixture to fix the workpiece position, with the marking machine head position preset, allowing for targeted marking at a fixed station. Chinese patent document CN206084141U discloses a robotic intelligent positioning laser marking system, which uses a six-axis robot to hold a laser marking machine to mark the workpiece, achieving automation of the marking process to a certain extent. However, this solution has shortcomings: First, the robot's motion trajectory requires manual teaching and pre-setting; changes in workpiece model or size necessitate re-teaching and adjustment, resulting in cumbersome operation and long changeover times. Second, when there is a positional deviation in the workpiece loading, the system cannot autonomously sense and correct it, easily causing marking position offsets and incomplete markings. Third, with only a single laser marking mode, controlling the laser marking depth is difficult for workpieces requiring deep marking (such as post-weld finished parts), easily leading to problems such as excessively deep marking damaging the workpiece's structural strength or excessively shallow marking causing easy wear of the markings.

[0004] The third method is vision-guided automated marking. Chinese patent document CN118926710A discloses a method, system, and storage medium for vision-guided laser marking of the steel bottom of a thermos cup, which uses a vision system to identify the center of the steel bottom of the thermos cup to achieve adaptive positioning marking. Chinese patent document CN115818174A discloses an automated vision-controlled rotation marking device for a pneumatic exhaust pipe connector, which controls a pneumatic marking machine to mark after positioning by visual imaging. Although the above solutions introduce visual positioning, they all have limitations: First, visual positioning is designed for a single specific workpiece and lacks the ability to flexibly adapt to different workpiece types; second, they only use a single mode of laser or pneumatic marking, respectively, and cannot be flexibly switched; third, the vision system is only used for single positioning and does not form a closed-loop control with marking depth control and surface pretreatment.

[0005] In addition, regarding surface pretreatment before marking, Chinese patent document CN113083804A discloses a laser intelligent rust removal method, system and readable medium, which identifies rust marks and plans the laser rust removal path through visual images. However, this solution is an independent rust removal device that is not integrated with the marking workstation. There is no closed loop between rust removal and marking, and it does not involve the reuse of the same laser mechanism between rust removal and marking.

[0006] In summary, existing metal workpiece marking solutions suffer from several drawbacks, including poor flexibility, low automation, insufficient positioning accuracy, limited marking modes, restricted working range, and lack of data traceability. Traditional manual marking relies on manual positioning and parameter adjustment, resulting in low efficiency and poor consistency. Fixed-station and teach-type marking are cumbersome to change models and cannot autonomously correct feeding deviations. Single laser or pneumatic modes are insufficient to meet the needs of both fine marking and deep embossing. Surface pretreatment such as rust removal is independent of marking and cannot form a closed loop. Summary of the Invention

[0007] To address the shortcomings of existing technologies, the purpose of this invention is to provide an automated robot marking workstation based on vision positioning. By integrating laser marking and pneumatic marking modes into a shared gripper at the robot's end effector, and employing vision-based adaptive positioning without teaching and closed-loop laser rust removal pretreatment, this invention achieves fully automated, high-precision, and flexible marking of various specifications of metal workpieces before and after welding.

[0008] The above-mentioned objective of this invention is achieved through the following technical solutions: This invention provides an automated robot marking workstation based on vision positioning, including a robot system, a 2D vision positioning system, a dual-mode marking system, an electrical control system, a host computer data system, and a safety protection system; The robot system includes an industrial robot, and the end of the industrial robot is provided with an end flange; The dual-mode marking system includes a laser marking mechanism, a pneumatic marking mechanism, and a shared gripper. The laser marking mechanism includes a laser generator and a laser marking lens, and the pneumatic marking mechanism includes a pneumatic marking head. The common gripper is fixedly installed on the end flange. The laser marking lens, the pneumatic marking head, and the 2D camera are all installed on the common gripper. The optical axis of the laser marking lens and the marking direction of the pneumatic marking head are both parallel to the axial direction of the end flange. The optical axis of the 2D camera is parallel to the optical axis of the laser marking lens. The shared gripper integrates laser marking mode and pneumatic marking mode on the same carrier, and the industrial robot drives the automatic switching between the two marking modes. The 2D vision positioning system includes a 2D camera, a light source, and a vision positioning algorithm module. The vision positioning algorithm module is used to identify the workpiece edge or weld positioning reference based on the workpiece image acquired by the 2D camera, and automatically calculate the trajectory, marking position, and marking posture of the industrial robot during its movement based on the spatial coordinates of the marking area of ​​the workpiece, the focal length of the laser marking lens, the spatial coordinates and reciprocating stroke of the pneumatic marking head, so as to offset the workpiece loading position deviation. The laser marking mechanism is also used to perform laser rust removal on the surface of the workpiece to be marked before laser marking or pneumatic marking; The 2D vision positioning system also includes a surface cleanliness judgment module, which is used to identify whether there is rust on the surface to be marked based on the 2D camera's image capture results and output the judgment result. When the judgment result is that there is rust, the laser marking mechanism is controlled to perform laser rust removal on the surface to be marked. After rust removal, the 2D camera is used to take pictures again and the surface cleanliness is judged. This process is repeated until the surface to be marked meets the marking requirements. The electrical control system includes a PLC controller and a human-machine interface. The PLC controller is connected to the industrial robot, the 2D camera, the laser marking mechanism and the pneumatic marking mechanism respectively. The human-machine interface is used to input the basic parameters of the workpiece and select the marking mode. The host computer data system is communicatively connected to the electrical control system and is used to collect marking data and upload the marking data to the manufacturing execution system. The security protection system is located on the periphery of the workstation; When the PLC controller receives the arrival signal sent by the industrial robot, it triggers the laser marking mechanism or the pneumatic marking mechanism to perform the corresponding marking operation according to the marking mode selected by the human-machine interface, and controls the industrial robot to reset after the marking is completed, thus forming a closed-loop operation process.

[0009] According to one embodiment of the present invention, the visual positioning algorithm module includes an image acquisition unit, a positioning reference recognition unit, a coordinate calculation unit, and a trajectory correction unit; The image acquisition unit is used to control the 2D camera to acquire workpiece images at the zero-point imaging position; The positioning reference recognition unit is used to identify the positioning reference of the workpiece edge or weld in the workpiece image based on the grayscale gradient features. The coordinate calculation unit is used to calculate the spatial coordinates of the marking area of ​​the workpiece and the theoretical pose of the end effector of the industrial robot based on the identified positioning reference, combined with the calibration parameters of the 2D camera, the focal length of the laser marking lens, and the spatial coordinates and reciprocating stroke of the pneumatic marking head. The trajectory correction unit is used to correct the motion trajectory of the industrial robot based on the calculated spatial coordinates of the marking area and the theoretical pose, so as to offset the workpiece loading position deviation. The compensation range of the loading position deviation is ±10mm, and the marking position error is controlled within 0.1mm.

[0010] According to one embodiment of the present invention, it further includes a ground rail walking system, the ground rail walking system including a ground rail body extending along a first direction, a guide rail disposed on the ground rail body, and a slide seat slidably disposed on the guide rail, the base of the industrial robot being fixedly mounted on the slide seat, and the ground rail walking system driving the industrial robot to reciprocate along the ground rail body to extend the working stroke of the industrial robot to cover long workpieces. The ground rail walking system also includes a rack, a drive motor, and a reducer. The rack is fixedly mounted on the ground rail body along the extension direction of the ground rail body. The drive motor and the reducer are mounted on the slide. The output shaft of the drive motor is connected to the drive gear through the reducer. The drive gear meshes with the rack for transmission. The ground rail walking system also includes limit blocks and limit switches disposed at both ends of the ground rail body; The PLC controller is also connected to the ground track walking system via signals.

[0011] According to one embodiment of the present invention, the common gripper includes a gripper base, a laser lens mounting base, a pneumatic head mounting base, and a camera mounting base. The laser marking lens is mounted on the gripper base via the laser lens mounting base, the pneumatic marking head is mounted on the gripper base via the pneumatic head mounting base, and the 2D camera is mounted on the gripper base via the camera mounting base.

[0012] According to one embodiment of the present invention, the electrical control system further includes a marking timing control unit, which is used to: after the industrial robot reaches the designated marking position and sends a position signal, send a trigger signal to the laser marking mechanism or the pneumatic marking mechanism according to the marking mode selected by the human-machine interface; After the marking operation is completed, the robot receives a completion signal from the marking controller and controls the industrial robot to reset to its initial position.

[0013] According to one embodiment of the present invention, the host computer data system includes a data acquisition module, a data storage module, and a MES communication interface; The data acquisition module is used to collect workpiece number, marking mode, marking parameters, marking time and marking result data in real time; The data storage module is used to store the collected marking data locally; The MES communication interface is used to upload the marking data to the manufacturing execution system, so as to realize the full traceability of the marking data and production scheduling management.

[0014] According to one embodiment of the present invention, the safety protection system includes a safety fence, a safety light curtain, and an emergency stop button. The safety fence is installed around the workstation, the safety light curtain is installed at the entrance and exit opening of the safety fence, and the emergency stop button is signal-connected to the PLC controller. When the PLC controller receives a trigger signal from the safety light curtain or the emergency stop button, it controls the industrial robot to stop urgently.

[0015] According to one embodiment of the present invention, the electrical control system further includes a marking depth control unit, which is used to generate laser marking parameters or pneumatic marking parameters based on the workpiece material parameters and marking depth requirements entered by the human-machine interface. When using laser marking mode, control the laser power and scanning speed to keep the marking depth within 0.25mm; When using pneumatic marking mode, control the impact force and reciprocating frequency of the pneumatic marking head to keep the marking depth within 0.25mm.

[0016] The present invention also provides a method for marking using a vision-based automated robot marking workstation employing the above-described embodiments, comprising the following steps: Step a: The operator enters the basic parameters of the workpiece and selects the marking mode through the human-machine interface; Step b: The workpiece is loaded onto the positioning platform, and the industrial robot carrying the 2D camera moves to the zero-point imaging position to capture the workpiece image; Step c: The visual positioning algorithm module identifies the workpiece edge or weld positioning reference based on the workpiece image, and calculates the theoretical pose of the end effector of the industrial robot by combining the spatial coordinates of the marking area of ​​the workpiece, the focal length of the laser marking lens, the spatial coordinates and reciprocating stroke of the pneumatic marking head, and corrects the motion trajectory to offset the workpiece loading position deviation. Step d: Before marking, the surface cleanliness judgment module identifies whether there is rust on the surface to be marked based on the 2D camera image results; When the judgment result is that there is rust, the laser marking mechanism is controlled to perform laser rust removal on the surface to be marked. After rust removal, the 2D camera is used to take pictures again and the surface cleanliness is judged. This process is repeated until the surface to be marked meets the marking requirements. Step e: The industrial robot moves to the marking position along the corrected motion trajectory and sends an position signal to the PLC controller; The PLC controller triggers the laser marking mechanism or the pneumatic marking mechanism to perform the corresponding marking operation according to the selected marking mode. After the marking is completed, it receives the completion signal from the marking controller and controls the industrial robot to reset to the initial position. Step f: The host computer data system collects the workpiece number, marking mode, marking parameters, marking time and marking result data for this marking, and uploads them to the manufacturing execution system to complete the full traceability of the marking data.

[0017] In summary, compared with the prior art, the present invention has at least one of the following beneficial technical effects: Enhanced flexibility and adaptability: This invention eliminates the need for customized fixtures and uses 2D vision-based adaptive positioning without teaching, making it compatible with various workpiece types such as pipes, profiles, plates, and post-weld supports. When changing models, only the human-machine interface parameters need to be modified, eliminating the need for re-teaching, significantly reducing changeover time and adapting to the needs of multi-variety, small-batch production.

[0018] Dual-mode flexible marking: This invention integrates a laser marking mechanism and a pneumatic marking mechanism into a shared gripper at the end of a robot. The optical axes are arranged in parallel, and the robot drives the automatic switching between the two marking modes. Laser marking (fine, non-contact) or pneumatic marking (deep imprinting, wear-resistant) can be flexibly selected according to the workpiece material and marking requirements. The workpiece can be clamped once before or after welding to complete the matching marking.

[0019] Improved automation and intelligence: This invention achieves full automation of the entire process from "manual parameter input - autonomous positioning - surface pretreatment - automatic marking - automatic unloading", eliminating the need for manual teaching and positioning. A single operator can manage multiple devices simultaneously, reducing labor costs and improving production efficiency.

[0020] Stable and reliable marking accuracy and quality: This invention uses a visual positioning algorithm combined with the laser lens focal length and the reciprocating stroke of the pneumatic marking head to calculate the trajectory, which can offset a feeding deviation of ±10mm and a marking position error of no more than 0.1mm; the marking depth control unit precisely controls the marking depth within 0.25mm; through laser rust removal and surface cleanliness judgment closed-loop iteration, it ensures that the surface to be marked meets the marking requirements before marking, avoiding problems such as incomplete markings and poor adhesion.

[0021] Wide working range and comprehensive application scenarios: This invention extends the robot's working range through a ground rail walking system, enabling full-area marking of long pipes; it is also compatible with both pre-welding single-piece and post-welding finished / semi-finished product processing stages, allowing one device to replace multiple single-function marking devices, thus improving equipment utilization.

[0022] Improved system collaboration and data traceability: This invention uses a PLC controller to link various modules, ensuring precise signal transmission, stable equipment operation, and continuous batch processing; the host computer data system is interconnected with the manufacturing execution system, enabling full traceability of marking data and intelligent production scheduling. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the overall structure of the marking workstation provided in an embodiment of the present invention; Figure 2 This is a top view of the layout of the marking workstation provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the shared gripper and dual-mode marking mechanism provided in an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the principle of 2D visual positioning and coordinate calculation provided in an embodiment of the present invention; Figure 5 A closed-loop flowchart for laser rust removal and surface cleanliness assessment provided in this embodiment of the invention; Figure 6 This is a block diagram of the electrical control system architecture provided in an embodiment of the present invention; Figure 7 This is a schematic diagram illustrating the interaction between the host computer data system and the manufacturing execution system provided in an embodiment of the present invention; Figure 8 This is a flowchart illustrating the marking process logic provided in an embodiment of the present invention. Figure 9 This is a schematic diagram showing the integration of a pneumatic dot matrix marking head, a laser marking lens, and a 2D vision camera provided in an embodiment of the present invention.

[0024] Reference numerals: 1. Ground rail body; 2. Guide rail; 3. Slide; 4. Rack; 5. Drive motor; 6. Industrial robot; 7. End flange; 8. Shared gripper; 9. Laser marking lens; 10. Pneumatic marking head; 11. 2D camera; 12. Positioning platform; 13. Workpiece; 14. Safety fence; 15. Safety light curtain; 16. Electrical control cabinet; 17. Human-machine interface; 18. Host computer; 19. Zero-point imaging position; 20. Marking area; 21. Laser generator; 22. Light source; 23. Gripper base; 24. Laser lens mounting base; 25. Pneumatic head mounting base; 26. Camera mounting base; 27. Pneumatic marking head body; 28. Dot matrix marking needle; 29. ​​Reciprocating drive mechanism; 32. Marking controller; 33. Unloading mechanism; 34. Material frame; 35. Weld seam; 36. Workpiece edge. Detailed Implementation

[0025] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0026] In the description of this invention, the term "first direction" refers to the extension direction of the rail body 1 (i.e., the X direction), "second direction" refers to the direction perpendicular to the extension direction of the rail body 1 and located in the horizontal plane (i.e., the Y direction), and "third direction" refers to the vertical direction (i.e., the Z direction). The terms "connection," "installation," and "fixing," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral molding; it can be a mechanical connection, an electrical connection, or a signal connection; it can be a direct connection or an indirect connection through an intermediate medium.

[0027] Example 1 like Figure 1 and Figure 2 As shown, this embodiment provides an automated robot marking workstation based on vision positioning. The workstation is an integrated automated equipment, consisting of a robot system, a 2D vision positioning system, a dual-mode marking system, an electrical control system, a safety protection system, and a host computer data system. Preferably, the workstation also includes a ground rail walking system to extend the working stroke of the industrial robot to cover long-sized workpieces.

[0028] like Figure 1 As shown, Figure 1This is a schematic diagram of the overall structure of the marking workstation in this embodiment. The ground rail walking system includes a ground rail body 1 extending along a first direction (X direction). The ground rail body 1 is fixedly installed on the workshop floor. A guide rail 2 is provided on the ground rail body 1. The guide rail 2 extends along the first direction, and a slide block 3 is slidably provided on the guide rail 2.

[0029] The robot system includes an industrial robot 6, which is a six-axis articulated industrial robot. The base of the industrial robot 6 is fixed to the upper surface of the slide 3 by bolts. The industrial robot 6 moves back and forth along the ground rail body 1 with the slide 3, thereby extending the working stroke of the industrial robot 6 and enabling the industrial robot 6 to cover the marking area of ​​long workpieces (such as 12m long pipes).

[0030] like Figure 1 As shown, the ground rail walking system also includes a rack 4, a drive motor 5, and a reducer (not shown in the figure). The rack 4 is fixedly mounted on the ground rail body 1 along a first direction, and the rack 4 is arranged parallel to the guide rail 2. The rack 4 is fixedly installed on the side or upper surface of the ground rail body 1 by bolts. The drive motor 5 and the reducer are mounted on the slide 3. The output shaft of the drive motor 5 is connected to the drive gear through the reducer. The drive gear meshes with the rack 4, thereby driving the slide 3 and the industrial robot 6 to reciprocate along the ground rail body 1. The ground rail walking system also includes limit blocks and limit switches set at both ends of the ground rail body 1. The limit blocks are used to prevent the slide 3 from disengaging from the ground rail body 1, and the limit switches are used to detect whether the slide 3 has reached the extreme positions at both ends of the ground rail body 1 and output limit signals.

[0031] like Figure 1 and Figure 2 As shown, the 2D vision positioning system includes a 2D camera 11, a light source 22, and a vision positioning algorithm module. The 2D camera 11 is mounted on the end flange 7 of the industrial robot 6. Specifically, the 2D camera 11 is mounted on a common gripper 8 via a camera mounting bracket 26, and the common gripper 8 is fixedly mounted on the end flange 7 of the industrial robot 6. The light source 22 is a ring light source or a strip light source, and is installed around the 2D camera 11 to provide uniform illumination for the 2D camera 11.

[0032] like Figure 1 and Figure 3As shown, the dual-mode marking system includes a laser marking mechanism, a pneumatic marking mechanism, and a shared gripper 8. The laser marking mechanism includes a laser generator 21 and a laser marking lens 9. The laser generator 21 is located inside the electrical control cabinet 16 or independently on one side of the workstation. The laser generator 21 is connected to the laser marking lens 9 via optical fiber. The laser marking lens 9 includes a galvanometer 28, a reflector 29, and a focusing lens 30. The pneumatic marking mechanism includes a pneumatic marking head 10, which includes dot matrix marking needles 27 and a reciprocating drive mechanism 31. The reciprocating drive mechanism 31 drives the dot matrix marking needles 27 to reciprocate at high frequency along the marking direction, impacting the workpiece surface to form a dot matrix mark.

[0033] like Figure 3 As shown, Figure 3 This is a schematic diagram of the shared gripper 8 and the dual-mode marking mechanism in this embodiment. The shared gripper 8 includes a gripper base 23, a laser lens mounting base 24, a pneumatic head mounting base 25, and a camera mounting base 26. The gripper base 23 is a plate-shaped structural component made of aluminum alloy or steel plate, and is fixedly mounted on the end flange 7 of the industrial robot 6. The laser marking lens 9 is mounted on the gripper base 23 via the laser lens mounting base 24, the pneumatic marking head 10 is mounted on the gripper base 23 via the pneumatic head mounting base 25, and the 2D camera 11 is mounted on the gripper base 23 via the camera mounting base 26. The optical axis of the laser marking lens 9 and the marking direction of the pneumatic marking head 10 are both parallel to the axial direction of the end flange 7 of the industrial robot 6, and the optical axis of the 2D camera 11 is parallel to the optical axis of the laser marking lens 9. In this way, the laser marking lens 9, the pneumatic marking head 10 and the 2D camera 11 are integrated on the same common gripper 8. The common gripper 8 integrates the laser marking mode and the pneumatic marking mode on the same carrier. The two marking modes are automatically switched by the industrial robot 6 without the need to replace the end tool or recalibrate.

[0034] like Figure 2 As shown, Figure 2 This is a top view of the marking workstation in this embodiment. A positioning platform 12 is provided within the workstation's working area. The positioning platform 12 is located on one side of the ground rail body 1 (i.e., in the second direction, Y direction). The positioning platform 12 is used to support the workpiece 13 to be marked. The positioning platform 12 is either a fixed platform or a V-shaped bracket platform. After the workpiece 13 is loaded onto the positioning platform 12, precise positioning is not required, and deviations in the loading position are permissible.

[0035] like Figure 1 and Figure 2As shown, the safety protection system includes a safety fence 14, a safety light curtain 15, and an emergency stop button (not shown in the figure). The safety fence 14 is installed around the workstation. The safety fence 14 is a metal frame structure with a height of not less than 1.8m and has access openings. The safety light curtain 15 is installed at the access openings of the safety fence 14 to detect whether personnel or objects have entered the workstation. The emergency stop button is located on the outside of the safety fence 14 or on the electrical control cabinet 16. The emergency stop button is connected to the PLC controller. When the PLC controller receives a trigger signal from the safety light curtain 15 or the emergency stop button, it controls the industrial robot 6 and the ground track walking system to stop urgently.

[0036] like Figure 6 As shown, Figure 6 This is a block diagram of the electrical control system architecture in this embodiment. The electrical control system includes an electrical control cabinet 16, a PLC controller, a human-machine interface 17, and a marking controller 32. The electrical control cabinet 16 is located on one side of the workstation. The PLC controller, servo driver, pneumatic solenoid valve, electrical components, and laser generator 21 are all located inside or on the electrical control cabinet 16. The PLC controller is connected to the industrial robot 6 (via the robot control cabinet), the 2D camera 11, the laser marking mechanism, the pneumatic marking mechanism, and the ground rail walking system. The human-machine interface 17 is a touch screen, located on the electrical control cabinet 16 or outside the safety fence 14. The human-machine interface 17 is used to input basic workpiece parameters and select the marking mode. The marking controller 32 is connected to the laser marking mechanism and the pneumatic marking mechanism to control the execution of the marking operation.

[0037] like Figure 7 As shown, Figure 7 This is a schematic diagram illustrating the interaction between the host computer data system and the manufacturing execution system in this embodiment. The host computer data system includes a host computer 18, a data acquisition module, a data storage module, and an MES communication interface. The host computer 18 is an industrial computer, which is communicatively connected to the PLC controller in the electrical control cabinet 16. The host computer data system is also communicatively connected to the electrical control system, and is used to collect marking data and upload the marking data to the manufacturing execution system (MES).

[0038] like Figure 8 As shown, Figure 8 This is a flowchart illustrating the marking logic of the marking workstation in this embodiment. The marking operation process of this workstation is as follows: Step S1: Operator preparation. The operator checks the equipment origin and confirms that the industrial robot 6 and the slide 3 are in the initial origin position; enters the basic parameters of the workpiece through the human-machine interface 17, including workpiece type, workpiece size, material, marking area position, marking content, etc.; selects the marking mode (laser marking mode or pneumatic marking mode) through the human-machine interface 17.

[0039] Step S2: Switch to automatic mode and start the workstation. The operator switches the workstation to automatic mode via the human-machine interface 17 and clicks the start button. The workstation then begins to run automatically.

[0040] Step S3: Workpiece loading. Workpiece 13 is loaded onto positioning platform 12 manually, via conveyor belt, AGV trolley, or hoisting equipment. After workpiece 13 is loaded onto positioning platform 12, precise positioning is not required, and a loading position deviation within ±10mm is allowed.

[0041] Step S4: The robot moves to the zero-point imaging position. The industrial robot 6, carrying the 2D camera 11, moves along the ground rail system to the zero-point imaging position 19.

[0042] Step S5: The vision system acquires images. The 2D camera 11 acquires an image of the workpiece 13 at the zero-point imaging position 19. The light source 22 provides uniform illumination. The 2D camera 11 transmits the acquired image to the vision positioning algorithm module.

[0043] Step S6: Identify the positioning reference. The visual positioning algorithm module processes the acquired image, and through image preprocessing (grayscale conversion, filtering, binarization) and edge detection (Canny operator or Sobel operator), identifies the edge 36 or weld 35 of workpiece 13 as the positioning reference.

[0044] Step S7: Solve spatial coordinates. Based on the identified positioning reference, the visual positioning algorithm module combines the calibration parameters (intrinsic and extrinsic parameters) of the 2D camera 11, the focal length of the laser marking lens 9, the spatial coordinates and reciprocating stroke of the pneumatic marking head 10 to calculate the spatial coordinates of the marking area of ​​the workpiece 13 and the theoretical pose of the end effector of the industrial robot 6.

[0045] Step S8: Surface cleanliness judgment and laser rust removal. The surface cleanliness judgment module identifies whether there is rust on the surface to be marked based on the image captured by the 2D camera 11. When the judgment result indicates the presence of rust, the PLC controller controls the laser marking mechanism to perform laser rust removal on the surface to be marked. After rust removal, the industrial robot 6 drives the 2D camera 11 to take another image and perform surface cleanliness judgment. This process is repeated iteratively until the surface to be marked meets the marking requirements.

[0046] Step S9: The robot corrects its trajectory and moves to the marking position. The industrial robot 6 corrects its motion trajectory based on the calculated spatial coordinates of the marking area and the theoretical pose, and moves to the designated marking position. The end effector of the industrial robot 6 carries a shared gripper 8, which aligns the laser marking lens 9 or the pneumatic marking head 10 with the surface to be marked.

[0047] Step S10: The robot sends a position signal. After the industrial robot 6 reaches the designated marking position, it sends a position signal to the PLC controller through the robot control cabinet.

[0048] Step S11: PLC triggers marking. After receiving the arrival signal, the PLC controller sends a trigger signal to the marking controller 32 according to the marking mode selected by the human-machine interface 17. The marking controller 32 controls the laser marking mechanism or the pneumatic marking mechanism to perform the corresponding marking operation.

[0049] Step S12: Marking completion feedback. After the marking operation is completed, the marking controller 32 sends a completion signal to the PLC controller.

[0050] Step S13: Robot Reset. After receiving the completion signal, the PLC controller controls the industrial robot 6 to reset to its initial position (origin position).

[0051] Step S14: Workpiece unloading. Workpiece 13 is automatically unloaded into the material frame 34 via the unloading mechanism 33.

[0052] Step S15: Data Upload. The data acquisition module of the host computer 18 collects data such as the workpiece number, marking mode, marking parameters, marking time, and marking result of this marking, and uploads it to the Manufacturing Execution System through the MES communication interface.

[0053] Step S16: Batch judgment. The PLC controller determines whether all workpieces in this batch have been marked. If not, it returns to step S3 to mark the next workpiece; if the batch has been marked, it proceeds to step S17.

[0054] Step S17: Workstation shuts down and waits for the next batch of tasks.

[0055] In this embodiment, steps S1 to S17 constitute a specific implementation of a marking method using the marking workstation of this embodiment. This method corresponds to the marking method defined by the present invention, and the specific correspondence is as follows: Step S1 corresponds to step a above; Step S3 corresponds to workpiece loading in step b; Steps S4 and S5 correspond to the robot carrying a 2D camera moving to the zero-point imaging position to collect workpiece images in step b; Steps S6 and S7 correspond to step c; Step S8 corresponds to step d; Steps S9 and S10 correspond to the industrial robot moving along the corrected motion trajectory to the marking position and sending a position signal in step e; Steps S11, S12, and S13 correspond to triggering marking, receiving the completion signal, and resetting in step e; Step S15 corresponds to step f.

[0056] In this embodiment, the workstation adopts a parameterized, teachless, fully automatic control mode. The operator only needs to input the basic parameters of the workpiece and select the marking mode through the human-machine interface 17, and the system can run autonomously. There is no need for manual teaching of the trajectory or use of special tooling for positioning throughout the entire process, realizing full automation of the process of "manual parameter input - autonomous positioning - surface pretreatment - automatic marking - automatic unloading".

[0057] Example 2 This embodiment is basically the same as Embodiment 1, except that this embodiment further explains the visual positioning algorithm module and visual calibration method in the 2D visual positioning system.

[0058] Before the marking workstation is put into use, the 2D vision positioning system needs to be visually calibrated to establish the transformation relationship between the coordinate systems of the 2D camera 11, the laser marking lens 9, the pneumatic marking head 10, and the industrial robot's 6-base coordinate system. Since the laser marking lens 9, the pneumatic marking head 10, and the 2D camera 11 are all integrated on the same common gripper 8, the relative positions of each coordinate system are fixed after installation. Therefore, a single calibration can serve both laser marking and pneumatic marking modes simultaneously. The visual calibration includes the following steps: Step C1: Place the calibration plate (such as a checkerboard calibration plate or a dot array calibration plate) on the positioning platform 12, with the installation position of the calibration plate corresponding to the marking position of the workpiece 13 to be marked; Step C2: Control the industrial robot 6 to move the 2D camera 11 to multiple different poses. The 2D camera 11 acquires images of the calibration board. The intrinsic parameters (focal length, principal point coordinates, distortion coefficients) of the 2D camera 11 and the extrinsic parameters (hand-eye calibration) between the 2D camera 11 and the industrial robot 6 are calculated by Zhang Zhengyou calibration method or Tsai two-step calibration method. Step C3: Use the teach pendant of the industrial robot 6 or the host computer 18 to control the movement of the industrial robot 6, so that the focus of the laser marking lens 9 is aligned with multiple known coordinates on the calibration plate in sequence, and record the coordinates of each calibration point in the base coordinate system of the industrial robot 6, thereby establishing the transformation relationship between the coordinate system of the laser marking lens 9 and the base coordinate system of the industrial robot 6. Step C4: Control the movement of the industrial robot 6 so that the needle tip of the dot matrix marking needle 28 of the pneumatic marking head 10 is aligned with multiple known coordinates on the calibration plate in sequence. Record the coordinates of each calibration point in the base coordinate system of the industrial robot 6, thereby establishing the transformation relationship between the coordinate system of the pneumatic marking head 10 and the base coordinate system of the industrial robot 6. Step C5: Store the calibration results obtained in steps C2 to C4 in the visual positioning algorithm module to complete the visual calibration.

[0059] Once the visual calibration is complete, the workstation can begin marking operations. During the operation, the visual positioning algorithm module uses the calibration results to transform the positioning reference pixel coordinates identified by the 2D camera 11 into the industrial robot 6 base coordinate system, thereby calculating the spatial coordinates of the marking area of ​​the workpiece 13 and the theoretical pose of the end effector of the industrial robot 6.

[0060] like Figure 4 As shown, Figure 4 This is a schematic diagram illustrating the 2D visual positioning and coordinate calculation principle of this embodiment. The visual positioning algorithm module includes an image acquisition unit, a positioning reference recognition unit, a coordinate calculation unit, and a trajectory correction unit.

[0061] The image acquisition unit controls the 2D camera 11 to acquire images of the workpiece at the zero-point imaging position 19. The image acquisition unit controls the 2D camera 11 to take pictures through a trigger signal. The acquisition frame rate of the 2D camera 11 is not less than 30fps, and the exposure time is automatically adjusted according to the brightness of the light source 22 and the surface reflectivity of the workpiece 13.

[0062] The positioning reference recognition unit is used to identify the positioning reference of the workpiece edge 36 or weld 35 in the workpiece image based on gray-level gradient features. The positioning reference recognition unit first preprocesses the image, including gray-level processing, median filtering for noise reduction, and histogram equalization processing; then, it extracts the image edges using the Canny operator or the Sobel operator, identifies the workpiece edge 36 or weld 35 using gray-level gradient features, and fits the straight line equation or curve equation of the positioning reference.

[0063] The coordinate calculation unit is used to calculate the spatial coordinates of the marking area of ​​the workpiece 13 and the theoretical pose of the end effector of the industrial robot 6 based on the identified positioning reference, combined with the calibration parameters of the 2D camera 11, the focal length of the laser marking lens 9, and the spatial coordinates and reciprocating stroke of the pneumatic marking head 10. Since the laser marking lens 9 and the pneumatic marking head 10 are integrated on the same common gripper 8, the coordinate calculation unit needs to perform calculations for the two marking modes separately: in the laser marking mode, the relationship between the laser focusing plane and the marking position is determined based on the focal length of the laser marking lens 9; in the pneumatic marking mode, the contact position relationship between the marking tip and the workpiece surface is determined based on the spatial coordinates and reciprocating stroke of the pneumatic marking head 10.

[0064] The trajectory correction unit is used to correct the motion trajectory of the industrial robot 6 based on the calculated spatial coordinates of the marking area and the theoretical pose, in order to compensate for the loading position deviation of the workpiece 13. The trajectory correction unit compares the calculated theoretical pose with the theoretical design pose of the workpiece 13, calculates the position deviation and attitude deviation, and generates the corrected motion trajectory. The compensation range for the loading position deviation is ±10mm, and the marking position error is controlled within 0.1mm.

[0065] Example 3 This embodiment is basically the same as Embodiment 1, except that this embodiment adds laser rust removal and surface cleanliness judgment functions.

[0066] like Figure 5 As shown, Figure 5 This is a closed-loop flowchart for laser rust removal and surface cleanliness assessment in this embodiment. In this embodiment, the laser marking mechanism is also used to perform laser rust removal on the surface of the workpiece 13 to be marked before laser marking and pneumatic marking, removing rust, scale, oil, and other contaminants from the surface to be marked to ensure marking quality. Laser rust removal and laser marking share the same laser generator 21 and laser marking lens 9, eliminating the need for additional independent rust removal equipment.

[0067] The surface cleanliness assessment module is integrated into the 2D vision positioning system. This module identifies the presence of rust on the surface to be marked based on images captured by the 2D camera 11 and outputs the assessment result. Specifically, the module uses image processing algorithms to identify rust. It analyzes the grayscale distribution, color features, and texture features of the image of the surface to be marked to identify rust areas and determine whether the rust area percentage exceeds a preset threshold. If it exceeds the threshold, the assessment result is that rust is present; otherwise, the assessment result is that the surface meets the marking requirements.

[0068] The closed-loop process for laser rust removal and surface cleanliness assessment is as follows: Step T1: The 2D camera 11 captures an image of the surface to be marked at the zero-point imaging position 19; Step T2: The surface cleanliness judgment module identifies whether there is rust on the surface to be marked based on the image captured by the 2D camera 11; Step T3: When the judgment result is that rust exists, the PLC controller controls the laser marking mechanism to perform laser rust removal on the surface to be marked; during laser rust removal, the laser marking lens 9 focuses the laser beam on the surface to be marked, and the laser generator 21 outputs the laser at a low power density (such as 10W / cm² to 50W / cm²) to vaporize and peel off the rust and oxide scale on the surface to be marked without damaging the workpiece substrate; Step T4: After the laser rust removal is completed, the industrial robot 6 drives the 2D camera 11 to collect the image of the surface to be marked again, and returns to step T2 to repeat the surface cleanliness judgment. Step T5: When the judgment result is that it meets the marking requirements, the laser rust removal process ends and the subsequent marking operation begins.

[0069] Through the iterative closed-loop process of laser rust removal and surface cleanliness assessment, it is possible to ensure that the surface to be marked reaches sufficient cleanliness before marking, avoiding problems such as incomplete marking content and poor adhesion due to surface corrosion, thereby further improving marking quality. Laser rust removal and marking are integrated into the same laser marking mechanism and the same shared gripper, eliminating the need for manual intervention and achieving an automated closed-loop surface pretreatment before marking.

[0070] Example 4 This embodiment is basically the same as Embodiment 1, except that this embodiment further explains the marking timing control unit and marking depth control unit in the electrical control system.

[0071] The marking timing control unit is used to: after the industrial robot 6 reaches the designated marking position and sends a position signal, send a trigger signal to the laser marking mechanism or pneumatic marking mechanism according to the marking mode selected by the human-machine interface 17; and after the marking operation is completed, receive the completion signal fed back by the marking controller 32 and control the industrial robot 6 to reset to the initial position. The marking timing control unit is set in the PLC controller to ensure that the timing sequence of "position-trigger-mark-feedback-reset" is accurate and reliable, and to avoid false triggering or missed marking.

[0072] The marking depth control unit generates laser marking parameters or pneumatic marking parameters based on the workpiece material parameters and marking depth requirements entered into the human-machine interface 17. When using laser marking mode, the marking depth control unit controls the laser power and scanning speed according to the workpiece material (such as carbon steel, stainless steel, aluminum alloy, etc.) and marking depth requirements to keep the marking depth within 0.25mm. When using pneumatic marking mode, the marking depth control unit controls the impact force and reciprocating frequency of the pneumatic marking head 10 to keep the marking depth within 0.25mm. The marking depth control unit is located in the PLC controller or in the marking controller 32.

[0073] Example 5 This embodiment is basically the same as Embodiment 1, except that the structure of the common gripper 8 and the pneumatic marking head 10 of the dual-mode marking system is further explained in this embodiment.

[0074] like Figure 9 As shown, Figure 9This is a schematic diagram showing the integration of the pneumatic dot matrix marking head, laser marking lens, and 2D vision camera in this embodiment. The shared gripper 8 includes a gripper base 23, a laser lens mounting base 24, a pneumatic head mounting base 25, and a camera mounting base 26. The laser marking lens 9 is mounted on the gripper base 23 via the laser lens mounting base 24, the pneumatic marking head 10 is mounted on the gripper base 23 via the pneumatic head mounting base 25, and the 2D camera 11 is mounted on the gripper base 23 via the camera mounting base 26. The pneumatic marking head 10 includes dot matrix marking needles 27 and a reciprocating drive mechanism 31. The reciprocating drive mechanism 31 is a cylinder or an electromagnetic drive mechanism that drives the dot matrix marking needles 27 to reciprocate at high frequency along the marking direction, forming characters or graphic markings composed of dot matrix on the surface of the workpiece.

[0075] In this embodiment, the optical axis of the laser marking lens 9, the marking direction of the pneumatic marking head 10, and the optical axis of the 2D camera 11 are parallel to each other, and the focal plane of the laser marking lens 9 and the working plane of the needle tip of the pneumatic marking head 10 are located on the same plane or at a predetermined distance. This arrangement allows the visual calibration results to be applied to both laser marking and pneumatic marking modes simultaneously. That is, when the industrial robot 6 drives the shared gripper 8 to align with the surface to be marked in a certain posture, the 2D camera 11, the laser marking lens 9, and the pneumatic marking head 10 can simultaneously align with the same marking area, thereby achieving rapid and accurate switching between the two marking modes without the need for repeated calibration.

[0076] Example 6 This embodiment provides a specific application scenario for an automated robot marking workstation based on visual positioning.

[0077] In this embodiment, the workpiece 13 to be marked is a 12m long metal pipe. This pipe is a finished product after welding and needs to be permanently marked at a predetermined position on the outer wall of the pipe (including product number, specifications, material, production date, etc.). Since the pipe is 12m long, which exceeds the working range of the industrial robot 6's arm span, the workstation drives the industrial robot 6 to move back and forth along the ground rail body 1 through the ground rail walking system to cover the entire marking area of ​​the pipe.

[0078] During the marking operation, the pipe is hoisted onto the positioning platform 12. The positioning platform 12 is a V-shaped bracket type positioning platform, and the pipe is placed on multiple V-shaped brackets. After the pipe is loaded, the operator enters the pipe parameters (pipe diameter, length, material, etc.) and selects the marking mode through the human-machine interface 17. The workstation operates automatically: the industrial robot 6, carrying the 2D camera 11, moves to the zero-point imaging position 19, collects pipe images, and identifies the pipe weld 35 or the edge of the pipe end as the positioning reference; the visual positioning algorithm module, combined with the focal length of the laser marking lens 9 and the spatial coordinates and reciprocating stroke of the pneumatic marking head 10, calculates the spatial coordinates of the marking area and the theoretical pose of the end, and corrects the motion trajectory to compensate for the deviation of the pipe loading position; the surface cleanliness judgment module identifies rust on the surface to be marked and iteratively removes rust; the PLC controller triggers the laser marking mechanism or the pneumatic marking mechanism to mark the outer wall of the pipe according to the selected marking mode; after marking is completed, the industrial robot 6 resets, and the host computer 18 uploads the marking data to the manufacturing execution system to complete data traceability.

[0079] In this embodiment, the marking position error is no greater than 0.1mm, the marking depth is controlled within 0.25mm, and the marking is clear, uniform, and permanent, meeting the marking requirements of the finished pipe parts after welding.

[0080] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; 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 or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

[0081] Furthermore, those skilled in the art will understand that the visual positioning algorithm module, surface cleanliness judgment module, marking timing control unit, and marking depth control unit in the marking workstation of the present invention can be implemented by software programs, hardware circuits, or a combination of software and hardware, and their specific implementation methods do not constitute a limitation on the present invention.

[0082] Those skilled in the art should also understand that the industrial robot 6 used in this invention is not limited to a six-axis articulated industrial robot, but may also be a four-axis, five-axis or seven-axis robot; the laser generator 21 is not limited to a fiber laser, but may also be a semiconductor laser or a solid-state laser; the ground track walking system is not limited to rack and pinion transmission, but may also be a lead screw transmission, synchronous belt transmission or linear motor drive; all the above equivalent substitutions should be covered within the protection scope of this invention.

[0083] The embodiments described herein are preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape, and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. An automated robotic marking workstation based on vision positioning, characterized in that, This includes a robot system, a 2D vision positioning system, a dual-mode marking system, an electrical control system, a host computer data system, and a safety protection system. The robot system includes an industrial robot (6), and the end of the industrial robot (6) is provided with an end flange (7); The dual-mode marking system includes a laser marking mechanism, a pneumatic marking mechanism, and a common gripper (8). The laser marking mechanism includes a laser generator (21) and a laser marking lens (9). The pneumatic marking mechanism includes a pneumatic marking head (10). The common gripper (8) is fixedly installed on the end flange (7). The laser marking lens (9), the pneumatic marking head (10), and the 2D camera (11) are all installed on the common gripper (8). The optical axis of the laser marking lens (9) and the marking direction of the pneumatic marking head (10) are both parallel to the axial direction of the end flange (7). The optical axis of the 2D camera (11) is parallel to the optical axis of the laser marking lens (9). The common gripper (8) integrates the laser marking mode and the pneumatic marking mode on the same carrier and achieves automatic switching between the two marking modes through the industrial robot (6). The 2D vision positioning system includes a 2D camera (11), a light source (22), and a vision positioning algorithm module. The vision positioning algorithm module is used to identify the positioning reference of the edge or weld (35) of the workpiece (13) based on the image of the workpiece (13) collected by the 2D camera (11), and automatically calculate the trajectory, marking position and marking posture of the industrial robot (6) during the movement process based on the spatial coordinates of the marking area of ​​the workpiece (13), the focal length of the laser marking lens (9), the spatial coordinates and reciprocating stroke of the pneumatic marking head (10), so as to offset the deviation of the workpiece (13) loading position. The laser marking mechanism is also used to perform laser rust removal on the surface of the workpiece (13) to be marked before laser marking or pneumatic marking; the 2D vision positioning system also includes a surface cleanliness judgment module, which is used to identify whether there is rust on the surface to be marked based on the photo taken by the 2D camera (11) and output the judgment result; when the judgment result is that there is rust, the laser marking mechanism is controlled to perform laser rust removal on the surface to be marked, and after rust removal, the 2D camera (11) is used to take a photo again and perform surface cleanliness judgment, and the process is repeated until the surface to be marked meets the marking requirements; The electrical control system includes a PLC controller and a human-machine interface (17). The PLC controller is connected to the industrial robot (6), the 2D camera (11), the laser marking mechanism and the pneumatic marking mechanism respectively. The human-machine interface (17) is used to input the basic parameters of the workpiece (13) and select the marking mode. The host computer data system is communicatively connected to the electrical control system and is used to collect marking data and upload the marking data to the manufacturing execution system. The security protection system is located on the periphery of the workstation; After receiving the arrival signal sent by the industrial robot (6), the PLC controller triggers the laser marking mechanism or the pneumatic marking mechanism to perform the corresponding marking operation according to the marking mode selected by the human-machine interface (17), and controls the industrial robot (6) to reset after the marking is completed, forming a closed-loop operation process.

2. The automated robot marking workstation based on vision positioning according to claim 1, characterized in that, The visual positioning algorithm module includes an image acquisition unit, a positioning reference recognition unit, a coordinate calculation unit, and a trajectory correction unit. The image acquisition unit is used to control the 2D camera (11) to acquire the image of the workpiece (13) at the zero-point shooting position (19); The positioning reference identification unit is used to identify the positioning reference of the workpiece (13) edge or weld (35) in the workpiece (13) image based on the gray-scale gradient features; The coordinate calculation unit is used to calculate the spatial coordinates of the marking area of ​​the workpiece (13) and the theoretical pose of the end of the industrial robot (6) based on the identified positioning reference, combined with the calibration parameters of the 2D camera (11), the focal length of the laser marking lens (9) and the spatial coordinates and reciprocating stroke of the pneumatic marking head (10). The trajectory correction unit is used to correct the motion trajectory of the industrial robot (6) according to the calculated spatial coordinates of the marking area and the theoretical pose, so as to offset the workpiece (13) loading position deviation. The compensation range of the loading position deviation is ±10mm, and the marking position error is controlled within 0.1mm.

3. The automated robot marking workstation based on vision positioning according to claim 1, characterized in that, It also includes a ground rail walking system, which includes a ground rail body (1) extending along a first direction, a guide rail (2) disposed on the ground rail body (1), and a slide block (3) slidably disposed on the guide rail (2). The base of the industrial robot (6) is fixedly installed on the slide block (3). The ground rail walking system drives the industrial robot (6) to move back and forth along the ground rail body (1) to extend the working stroke of the industrial robot (6) to cover long workpieces (13). The ground rail walking system also includes a rack (4), a drive motor (5), and a reducer. The rack (4) is fixedly mounted on the ground rail body (1) along the extension direction of the ground rail body (1). The drive motor (5) and the reducer are mounted on the slide (3). The output shaft of the drive motor (5) is connected to the drive gear through the reducer. The drive gear meshes with the rack (4) for transmission. The ground rail walking system also includes limit blocks and limit switches at both ends of the ground rail body (1). The PLC controller is also connected to the ground track walking system via signals.

4. The automated robot marking workstation based on vision positioning according to claim 1, characterized in that, The common gripper (8) includes a gripper base (23), a laser lens mounting base (24), a pneumatic head mounting base (25), and a camera mounting base (26). The laser marking lens (9) is mounted on the gripper base (23) through the laser lens mounting base (24), the pneumatic marking head (10) is mounted on the gripper base (23) through the pneumatic head mounting base (25), and the 2D camera (11) is mounted on the gripper base (23) through the camera mounting base (26).

5. The automated robot marking workstation based on vision positioning according to claim 1, characterized in that, The electrical control system also includes a marking timing control unit, which is used to: after the industrial robot (6) reaches the designated marking position and sends a position signal, send a trigger signal to the laser marking mechanism or the pneumatic marking mechanism according to the marking mode selected by the human-machine interface (17); receive the completion signal fed back by the marking controller (32) after the marking operation is completed, and control the industrial robot (6) to reset to the initial position.

6. The automated robot marking workstation based on vision positioning according to claim 1, characterized in that, The host computer data system includes a data acquisition module, a data storage module, and an MES communication interface; The data acquisition module is used to collect workpiece (13) number, marking mode, marking parameters, marking time and marking result data in real time; The data storage module is used to store the collected marking data locally; The MES communication interface is used to upload the marking data to the manufacturing execution system, so as to realize the full traceability of the marking data and production scheduling management.

7. The automated robot marking workstation based on vision positioning according to claim 1, characterized in that, The safety protection system includes a safety fence (14), a safety light curtain (15), and an emergency stop button. The safety fence (14) is set around the workstation. The safety light curtain (15) is set at the entrance and exit of the safety fence (14). The emergency stop button is connected to the PLC controller. When the PLC controller receives the trigger signal of the safety light curtain (15) or the emergency stop button, it controls the industrial robot (6) to stop urgently.

8. The automated robot marking workstation based on vision positioning according to claim 1, characterized in that, The electrical control system also includes a marking depth control unit, which is used to generate laser marking parameters or pneumatic marking parameters based on the workpiece (13) material parameters and marking depth requirements entered by the human-machine interface (17). When using laser marking mode, the laser power and scanning speed are controlled to keep the marking depth within 0.25mm. When using pneumatic marking mode, the impact force and reciprocating frequency of the pneumatic marking head (10) are controlled to keep the marking depth within 0.25mm.

9. A method for marking using an automated robotic marking workstation based on vision positioning as described in any one of claims 1 to 8, characterized in that, Includes the following steps: Step a: The operator enters the basic parameters of the workpiece (13) and selects the marking mode through the human-machine interface (17); Step b: The workpiece (13) is loaded onto the positioning platform (12), and the industrial robot (6) carrying the 2D camera (11) moves to the zero-point shooting position (19) to collect the image of the workpiece (13); Step c: The visual positioning algorithm module identifies the positioning reference of the edge or weld (35) of the workpiece (13) based on the image of the workpiece (13), and calculates the theoretical pose of the end of the industrial robot (6) by combining the spatial coordinates of the marking area of ​​the workpiece (13), the focal length of the laser marking lens (9), the spatial coordinates and reciprocating stroke of the pneumatic marking head (10), and corrects the motion trajectory to offset the deviation of the workpiece (13) loading position. Step d: Before marking, the surface cleanliness judgment module identifies whether there is rust on the surface to be marked based on the photo taken by the 2D camera (11); when the judgment result is that there is rust, the laser marking mechanism is controlled to perform laser rust removal on the surface to be marked, and after rust removal, the 2D camera (11) is used to take a photo again and perform surface cleanliness judgment, and the process is repeated until the surface to be marked meets the marking requirements. Step e: The industrial robot (6) moves along the corrected motion trajectory to the marking position and sends a position signal to the PLC controller; the PLC controller triggers the laser marking mechanism or the pneumatic marking mechanism to perform the corresponding marking operation according to the selected marking mode, and receives the completion signal fed back by the marking controller (32) after marking is completed, and controls the industrial robot (6) to reset to the initial position; Step f: The host computer data system collects the workpiece (13) number, marking mode, marking parameters, marking time and marking result data for this marking, and uploads them to the manufacturing execution system to complete the full traceability of the marking data.

Citation Information

Patent Citations

  • Intelligent laser rust removal method and system and readable medium

    CN113083804A

  • Automatic visual rotation control marking device for pneumatic exhaust pipe connecting pipe

    CN115818174A

  • Vacuum cup steel bottom laser marking method and system based on visual guidance and storage medium

    CN118926710A

  • Intelligence location laser marking system of robot

    CN206084141U