A visual positioning method and system for a tire mold automatic drilling robot

Abstract

This invention proposes a visual positioning method and system for an automated drilling robot for tire molds. The method involves calibrating the camera's intrinsic and extrinsic parameters and acquiring the workpiece coordinates of the target surface markings and key points of the tire mold. Based on image feature templates, the method identifies the pixel coordinates of the target surface markings and the actual physical dimensions corresponding to each pixel. Based on the camera's intrinsic parameters, the workpiece coordinates of the target surface markings, and the pixel coordinates, the method solves for the initial transformation relationship between the workpiece and the robot coordinate system. Based on this initial transformation relationship, the method obtains the pixel coordinates of the key points. Based on the pixel coordinates of the key points, the image pose, and the camera's intrinsic parameters, the method calculates the actual coordinates of the key points in the robot coordinate system. Finally, a singular value decomposition algorithm is used to match the 3D coordinates of the key points with the robot coordinate system coordinates, thus solving for the precise transformation relationship between the workpiece and the robot coordinate system. By using template matching and machine vision to identify the target surface and key points, and by calculating the precise transformation relationship between the workpiece and the robot coordinate system, efficient positioning of the exhaust holes in tire molds is achieved.

Description

Technical Field

[0001] This invention relates to the field of robot control technology, and in particular to a visual positioning method and system for an automatic drilling robot for tire molds. Background Technology

[0002] Tire molds are used for the vulcanization and molding of various tires. During the tire vulcanization process, air holes need to be created on the tire surface of the mold to allow air to escape from the mold cavity. These air holes are numerous, irregularly distributed, and their axes must be perpendicular to the tire surface, with a wide range of directional variations. Currently, the processing of air holes in tire molds is mostly done manually, which is highly unpredictable and makes it difficult to guarantee consistency. The high-intensity work also involves mechanical vibration and noise pollution, and long-term exposure to this work can easily lead to occupational health risks, including but not limited to hearing damage and musculoskeletal strain.

[0003] Patent application number 202410961287.8 discloses a robot system and its control method for automatic drilling of tire molds, which uses a contact probe to locate the vent hole by detecting the outer contour of the tire mold. Patent application number 202310685202.3 discloses a fully automatic drilling machine for tire molds, which uses a 3D scanner to locate the vent hole.

[0004] Both of the above positioning methods save manpower, reduce the labor intensity of operators, and improve work efficiency. However, the contact probe uses a contact-based point-by-point measurement method to achieve positioning, which has disadvantages such as contact, low speed, and high cost; the 3D scanner requires dedicated CAM software, and the point cloud data generated by its scanning needs to be solved by algorithms for each exhaust hole one by one, which involves a large amount of calculation, high hardware performance requirements, and is greatly affected by noise, resulting in high equipment cost. Summary of the Invention

[0005] To address the aforementioned issues, this invention proposes a visual positioning method and system for an automated drilling robot for tire molds. By using template matching and machine vision to identify the target surface and key points respectively, the impact of noise is reduced. Furthermore, by calculating the precise transformation relationship between the workpiece and the robot coordinate system, the positioning of the exhaust hole of the tire mold is achieved, thereby saving time and improving positioning efficiency.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides a visual positioning method for an automated drilling robot for tire molds, comprising:

[0008] Calibrate the camera's intrinsic and extrinsic parameters, create image feature templates, and train the YOLO model;

[0009] Extract the workpiece coordinates of the target surface markings and key points on the tire mold;

[0010] Based on image feature templates, identify the target surface of the tire mold and obtain the pixel coordinates of the marking and the actual physical size corresponding to a single pixel;

[0011] Based on the camera intrinsic parameters, the workpiece coordinates and pixel coordinates marked on the target surface, the initial transformation relationship of the workpiece relative to the robot coordinate system is solved;

[0012] Based on the initial transformation relationship and the workpiece coordinates of key points, calculate the photographic pose of key points in the robot coordinate system, generate camera movement commands, drive the camera to the photographic pose to take pictures, and obtain the pixel coordinates of key points based on the pre-trained YOLO model;

[0013] Calculate the actual coordinates of the key points relative to the robot coordinate system based on the pixel coordinates of the key points, the photo pose, and the camera intrinsic parameters.

[0014] By matching the workpiece coordinates with the robot coordinate system coordinates of key points using the singular value decomposition algorithm, the precise transformation relationship between the workpiece and the robot coordinate system can be solved.

[0015] Preferably, the target markings are nine perfectly circular, uniformly spaced and sized circles arranged in a 3×3 pattern, pre-made at the center of the lower edge of the tire mold;

[0016] The image feature template is used to identify the target surface markings. The template includes nine perfectly circular shapes of the same size and spacing, arranged in a 3×3 configuration.

[0017] Preferably, the key points include the center point of the vent hole to be drilled and eight boundary points, with the nine key points arranged in a 3×3 configuration.

[0018] Preferably, the actual physical size corresponding to the single pixel is specifically as follows:

[0019]

[0020] Where D and d are the actual distance and pixel distance of any perfect circle in the target surface marking, respectively.

[0021] Preferably, the initial transformation relationship of the workpiece relative to the robot coordinate system is solved using the SQPnP algorithm based on the camera intrinsic parameters, the workpiece coordinates and pixel coordinates marked on the target surface.

[0022] Preferably, the step of calculating the photographic pose of key points in the robot coordinate system based on the initial transformation relationship and the workpiece coordinates of key points specifically includes:

[0023] Calculate the initial position of the key points in the robot coordinate system based on the initial transformation relationship and the workpiece coordinates of the key points;

[0024] Shift the initial position backward along the key point normal vector The distance is used to determine the shooting pose; among which For camera focal length, The x-coordinate of a single pixel. This represents the actual physical size corresponding to a single pixel.

[0025] Preferably, the step of calculating the actual coordinates of the key points relative to the robot coordinate system based on the key point pixel coordinates, the shooting pose, and the camera intrinsic parameters specifically includes:

[0026] ,

[0027]

[0028] in, The distance the camera pose is shifted backward, and the actual coordinates of the key points relative to the robot's coordinate system are: The pose for taking the photo is ), These are the pixel coordinates of the key points; ) represents the coordinates of the image center, and represents the camera intrinsic parameters; This represents the actual physical size corresponding to a single pixel.

[0029] Secondly, the present invention provides a vision positioning system for an automated drilling robot for tire molds, comprising:

[0030] The calibration module is used to calibrate the camera's intrinsic and extrinsic parameters, create image feature templates, and train the YOLO model.

[0031] The workpiece coordinate extraction module is used to extract the workpiece coordinates of the target surface markings and key points on the tire mold;

[0032] The template matching module is used to identify the target surface of the tire mold based on image feature templates, and obtain the pixel coordinates of the marking and the actual physical size corresponding to a single pixel;

[0033] The initial solution module is used to solve the initial transformation relationship of the workpiece relative to the robot coordinate system based on the camera intrinsic parameters, the workpiece coordinates and pixel coordinates marked on the target surface;

[0034] The dynamic positioning module is used to calculate the photographic pose of key points in the robot coordinate system based on the initial transformation relationship and the workpiece coordinates of key points, generate camera movement commands, drive the camera to the photographic pose to take pictures, and obtain the pixel coordinates of key points based on the pre-trained YOLO model;

[0035] The coordinate calculation module is used to calculate the actual coordinates of the key points relative to the robot coordinate system based on the pixel coordinates of the key points, the shooting pose, and the camera intrinsic parameters.

[0036] The precise solution module is used to match the workpiece coordinates and robot coordinates of key points using the singular value decomposition algorithm, and solve the precise transformation relationship between the workpiece and the robot coordinate system.

[0037] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the visual positioning method for an automated drilling robot for tire molds as described in the first aspect.

[0038] Fourthly, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the visual positioning method for an automatic drilling robot for tire molds as described in the first aspect.

[0039] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0040] This invention provides a foundation for precise positioning by calibrating the camera, creating templates, and training a YOLO model. Specifically, using a monocular camera to locate the exhaust holes in a tire mold reduces costs; the YOLO deep learning model effectively eliminates noise and improves recognition accuracy for key point identification. Furthermore, extracting target surface markers and key point coordinates, combined with image feature templates for target surface recognition, enables rapid acquisition of key information. Solving the initial transformation relationship, calculating the image pose and obtaining the pixel coordinates of key points, and then obtaining the actual coordinates, finally using a singular value decomposition algorithm to obtain the precise transformation relationship, achieves high-precision positioning, and improves positioning efficiency.

[0041] Compared to manual drilling, this method eliminates the arbitrariness of manual operation, ensures processing consistency, and avoids health risks. Compared to contact probes and 3D scanner positioning methods, it does not require contact with the mold, has a fast measurement speed, low cost, low computational load, low hardware performance requirements, and is less affected by noise. It effectively solves the shortcomings of existing positioning methods, provides efficient and accurate positioning support for automatic drilling of tire molds, and promotes the development of the tire mold processing industry towards automation and intelligence.

[0042] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0043] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute a limitation thereof.

[0044] Figure 1The main flowchart of a visual positioning method for an automatic drilling robot for tire molds provided in an embodiment of the present invention;

[0045] Figure 2 A schematic diagram of a tire mold provided in an embodiment of the present invention;

[0046] Figure 3 A schematic diagram of the tire mold target surface provided in an embodiment of the present invention;

[0047] Figure 4 This is a schematic diagram illustrating the identification of a tire mold target surface based on an image feature template, provided in an embodiment of the present invention.

[0048] Figure 5 The actual coordinates of key points relative to the robot coordinate system provided in the embodiments of the present invention Calculation diagram. Detailed Implementation

[0049] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0050] Example 1

[0051] like Figure 1 As shown, this embodiment discloses a visual positioning method for an automated drilling robot for tire molds, including the following steps:

[0052] S1: Calibrate the camera's intrinsic and extrinsic parameters, create image feature templates, and train the YOLO model;

[0053] S2: Extract the workpiece coordinates of the target surface markings and key points on the tire mold;

[0054] S3: Based on image feature templates, identify the target surface of the tire mold and obtain the pixel coordinates of the marking and the actual physical size corresponding to a single pixel;

[0055] S4: Based on the camera intrinsic parameters, the workpiece coordinates and pixel coordinates marked on the target surface, solve the initial transformation relationship of the workpiece relative to the robot coordinate system;

[0056] S5: Calculate the photographic pose of key points in the robot coordinate system based on the initial transformation relationship and key point workpiece coordinates, generate camera movement commands; drive the camera to the photographic pose to take pictures, and obtain the pixel coordinates of key points based on the pre-trained YOLO model;

[0057] S6: Calculate the actual coordinates of the key points relative to the robot coordinate system based on the key point pixel coordinates, the shooting pose, and the camera intrinsic parameters;

[0058] S7: Match the workpiece coordinates and robot coordinates of key points using the singular value decomposition algorithm to solve the precise transformation relationship between the workpiece and the robot coordinate system.

[0059] Next, combined Figure 1 This embodiment provides a detailed description of a visual positioning method for an automatic drilling robot for tire molds.

[0060] S1. Calibrate the camera's intrinsic and extrinsic parameters, create image feature templates, and train the YOLO model.

[0061] Specifically, a monocular camera was used, and its intrinsic and extrinsic parameters were calibrated using the Zhang Zhengyou calibration method. The calibration results include the camera's focal length. F Size of a single pixel ( ), Image center coordinates ( The relationship between the camera and the robot's end flange conversion. Parameters such as these.

[0062] Furthermore, image feature templates are created and the YOLO model is trained.

[0063] The image feature template is made based on the shape of the target surface coordinates and is used to identify the target surface at the center of the lower edge of the tire. The target surface is CNC machined, which has high precision and is suitable for low-cost and high-efficiency template matching.

[0064] The YOLO model is used to identify key points. Its training includes image acquisition of tire mold vent holes, vent hole annotation, and YOLO model weight training. The tire mold vent hole image acquisition is achieved by taking pictures of vent holes of different models and regions using a calibrated camera. The vent hole annotation is done using Labelme software. In this embodiment, the preferred YOLO model is YOLOv8l-seg. Since the vent hole machining area is cast and has an irregular shape, a more suitable YOLO deep learning model is selected.

[0065] In this embodiment, the camera's intrinsic and extrinsic parameters are calibrated using the Zhang Zhengyou calibration method, enabling precise acquisition of various camera parameters and providing accurate foundational data for subsequent image-based measurement and positioning. The generated image feature templates can quickly identify the tire mold target surface, improving target surface recognition efficiency and accuracy. Training the YOLO model effectively identifies key points, reducing recognition errors caused by the complex shape of the exhaust vents, laying the foundation for accurately determining the exhaust vent location, and comprehensively improving the performance and reliability of the visual positioning system.

[0066] S2. Extract workpiece coordinates from the target surface markings and key points on the tire mold. .

[0067] like Figure 2 As shown, the tire mold includes a target surface located at the center of its lower edge, an vent, and key points. The target surface coordinates and key point coordinates are also shown. The coordinates include the coordinates of nine circular circles on the target surface, and the coordinates of nine key points consisting of the center point of the vent hole to be drilled and eight boundary points, arranged in a 3×3 format. The coordinates were extracted using CAM software.

[0068] like Figure 3 As shown, the target surface is composed of nine circles arranged in a 3×3 pattern. In this embodiment, the horizontal spacing is 12.5 mm and the vertical spacing is 10 mm.

[0069] In this embodiment, the workpiece coordinates of the target surface markings and key points on the tire mold are accurately extracted, providing crucial data support for subsequent calculations of the mold's position and orientation in the robot coordinate system. This coordinate information forms the basis for establishing the motion relationship between the mold and the robot. By obtaining accurate coordinates, the robot can operate the mold more precisely. Furthermore, the 3×3 arrangement of coordinate points facilitates subsequent calculations and analysis, improving the accuracy and stability of positioning and ensuring that the robot can quickly locate the target position during automatic drilling.

[0070] In S3, based on image feature templates, the target surface of the tire mold is identified, and the pixel coordinates of the marking and the actual physical size corresponding to a single pixel are obtained.

[0071] The target surface shape is as follows Figure 3 As shown, target surface image acquisition is achieved by taking pictures with a camera.

[0072] The image feature template matching is performed using a shape-based template matching method, resulting in nine circular identifier pixel coordinates after template matching. The actual physical size corresponding to a single pixel .

[0073] Specifically, such as Figure 4 As shown, based on image feature templates, the target surface of the tire mold is identified, and after matching the circles, the pixel coordinates of the circle's center are obtained. Two identified circles can be randomly selected... and The pixel distance between the two circles The actual distance between the two circles ,but :

[0074]

[0075] Where D and d are the actual distance and pixel distance of any perfect circle in the target surface marking, respectively.

[0076] In this embodiment, the target surface is identified based on image feature templates, enabling rapid and accurate location of the target surface and acquisition of the marked pixel coordinates. The actual physical size corresponding to a single pixel is calculated, establishing a relationship between image pixels and actual physical dimensions, converting image information into a meaningful physical quantity. This allows for the accurate calculation of the target surface's position and size in actual space based on pixel coordinates, providing crucial data for determining the transformation relationship between the workpiece and the robot coordinate system. This significantly improves positioning accuracy and reduces the impact of positioning errors on subsequent drilling operations.

[0077] In S4, based on the camera intrinsic parameters, the workpiece coordinates marked on the target surface, and the pixel coordinates, the initial transformation relationship of the workpiece relative to the robot coordinate system is solved. 1;

[0078] Specifically, the robot is an industrial six-degree-of-freedom robot; the solution method is a pose estimation algorithm including PnP, DLT, EPnP, and SQPnP, preferably using the SQPnP algorithm for solution.

[0079] It should be understood that the process of solving the problem using algorithms is something that can be implemented by those skilled in the art.

[0080] In S5, the keypoints' pose in the robot coordinate system is calculated based on the initial transformation relationship and the workpiece coordinates, generating camera movement commands; the camera is driven to the desired pose for capture, and the pixel coordinates of the keypoints are obtained based on a pre-trained YOLO model. Specifically, this includes:

[0081] S501: Calculate the initial position of key points in the robot coordinate system based on the initial transformation relationship and key point workpiece coordinates. );

[0082] S502: Along the key point normal vector ( )Will Move back The distance is used to determine the shooting position. Generate camera movement commands;

[0083] S503: Based on camera movement commands, it drives the camera to the shooting pose and obtains the pixel coordinates of key points based on a pre-trained YOLO model.

[0084] In S501, firstly, the key point selection conforms to the principle of spatial analysis, and nine key points are selected at the edge and center of the entire exhaust port area.

[0085] Specifically, the calculation method is as follows:

[0086] 1

[0087] in, The transformation relationship from the robot's end flange to the robot's coordinate system is known when the robot leaves the factory. The conversion relationship from camera to end flange (obtained from camera calibration). This represents the transformation relationship from the object to the camera coordinate system; This refers to the initial transformation relation solved in S4. 1; These are the workpiece coordinates for key points.

[0088] Furthermore, the S502 calculates the camera pose. );

[0089] Specifically, the photographing pose The calculation method is to calculate along the normal vector of the exhaust hole. Move back The distance.

[0090] A suitable camera pose allows the camera to clearly capture key points, avoiding image blur or missing information due to improper viewing angles, and ensuring accurate recognition by the YOLO model. This embodiment determines the pose by accurately calculating the backward movement distance, which reduces coordinate transformation errors and improves the accuracy of the transformation from pixel coordinates to actual physical coordinates. This provides reliable data for solving the precise transformation relationship between the workpiece and the robot coordinate system, ensuring the robot accurately positions the vent hole and improving the accuracy and consistency of drilling.

[0091] Furthermore, in the S503, the robot drives the camera to... For key point P n Individual photo recognition;

[0092] Specifically, the key point recognition is performed using the YOLO model, and the recognition result outputs the pixel coordinates of the key points. .

[0093] It should be understood that step S5 can only identify one key point at a time, so it needs to be performed 9 times to obtain the pixel coordinates of all key points.

[0094] In S6, the coordinates of the key points in the robot coordinate system are calculated. );

[0095] Specifically, combined Figure 5 The coordinates of the key points in the robot coordinate system The calculation method is as follows:

[0096] ,in ;

[0097] In this embodiment, the actual coordinates are calculated based on the key point pixel coordinates, the image pose, and camera intrinsic parameters, transforming the pixel information in the image into actual spatial coordinates in the robot coordinate system. Through a precise coordinate transformation formula, and by fully utilizing the previously acquired parameters, an accurate mapping from image space to actual physical space is achieved. The actual coordinates of the key points obtained in this step accurately reflect the position of the vent in the robot's workspace, providing crucial data for ultimately determining the precise transformation relationship between the mold and the robot coordinate system, ensuring the accuracy and reliability of the positioning.

[0098] In S7, the precise transformation relationship between the workpiece and the robot coordinate system is calculated through key points. 2;

[0099] Specifically, the precise transformation relationship 2 is to set the key workpiece coordinates P n Its corresponding point The process involves forming point pairs and solving them using the singular value decomposition algorithm, a process that can be implemented by those skilled in the art.

[0100] In this embodiment, a singular value decomposition algorithm is used to match the coordinates of key points and solve for the precise transformation relationship. This fully utilizes the large amount of previously acquired data, eliminates error accumulation, and yields a more accurate position and orientation relationship between the mold and the robot coordinate system. This precise transformation relationship enables the robot to accurately locate the position of the vent hole on the tire mold, greatly improving positioning accuracy. Based on this precise transformation relationship, the robot can perform drilling operations more accurately, improving drilling accuracy and consistency, reducing scrap rate, and enhancing the quality and efficiency of tire mold processing.

[0101] This specific embodiment addresses the shortcomings of existing tire mold vent hole positioning methods by proposing a highly efficient and low-cost visual positioning method. Compared to positioning methods using contact probes, it avoids the drawbacks of contact measurement. By determining the workpiece pose through visual recognition and calculation, it avoids direct contact with the mold, reducing equipment wear, increasing measurement speed, and lowering costs.

[0102] Compared to methods using 3D scanners, this embodiment eliminates the need for dedicated CAM software. It utilizes a YOLO model to select key points, combining image feature templates and camera parameters to solve for transformation relationships. This approach reduces computational load, lowers hardware performance requirements, and is less affected by noise, significantly lowering equipment costs. Furthermore, this embodiment achieves precise positioning. By employing multiple calculations and singular value decomposition algorithms to solve for accurate transformation relationships, it ensures positioning accuracy, providing a reliable positioning foundation for automated drilling of tire molds and effectively improving the overall efficiency and quality of the drilling process.

[0103] Example 2

[0104] This embodiment provides a vision positioning system for an automated drilling robot for tire molds, including:

[0105] The calibration module is used to calibrate the camera's intrinsic and extrinsic parameters, create image feature templates, and train the YOLO model.

[0106] The workpiece coordinate extraction module is used to extract the workpiece coordinates of the target surface markings and key points on the tire mold;

[0107] The template matching module is used to identify the target surface of the tire mold based on image feature templates, and obtain the pixel coordinates of the marking and the actual physical size corresponding to a single pixel;

[0108] The initial solution module is used to solve the initial transformation relationship of the workpiece relative to the robot coordinate system based on the camera intrinsic parameters, the workpiece coordinates and pixel coordinates marked on the target surface;

[0109] The dynamic positioning module is used to calculate the photographic pose of key points in the robot coordinate system based on the initial transformation relationship and the workpiece coordinates of key points, generate camera movement commands, drive the camera to the photographic pose to take pictures, and obtain the pixel coordinates of key points based on the pre-trained YOLO model;

[0110] The coordinate calculation module is used to calculate the actual coordinates of the key points relative to the robot coordinate system based on the pixel coordinates of the key points, the shooting pose, and the camera intrinsic parameters.

[0111] The precise solution module is used to match the workpiece coordinates and robot coordinates of key points using the singular value decomposition algorithm, and solve the precise transformation relationship between the workpiece and the robot coordinate system.

[0112] Example 3

[0113] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the visual positioning method for an automatic drilling robot for tire molds as described in Embodiment 1 above.

[0114] Example 4

[0115] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the visual positioning method for an automatic drilling robot for tire molds as described in Embodiment 1 above.

[0116] The steps or modules involved in Embodiments 2 to 4 above correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0117] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A visual positioning method for an automated drilling robot for tire molds, characterized in that, include: Calibrate the camera's intrinsic and extrinsic parameters, create image feature templates, and train the YOLO model; Extract the workpiece coordinates of the target surface markings and key points on the tire mold; Based on image feature templates, identify the target surface of the tire mold and obtain the pixel coordinates of the marking and the actual physical size corresponding to a single pixel; Based on the camera intrinsic parameters, the workpiece coordinates and pixel coordinates marked on the target surface, the initial transformation relationship of the workpiece relative to the robot coordinate system is solved; Based on the initial transformation relationship and the workpiece coordinates of key points, calculate the photographic pose of key points in the robot coordinate system, generate camera movement commands, drive the camera to the photographic pose to take pictures, and obtain the pixel coordinates of key points based on the pre-trained YOLO model; Based on the keypoint pixel coordinates, the image pose, and camera intrinsic parameters, calculate the actual coordinates of the keypoints relative to the robot coordinate system, specifically including: ， in, The distance the camera pose is shifted backward, and the actual coordinates of the key points relative to the robot's coordinate system are: The pose for taking the photo is ), These are the pixel coordinates of the key points; () represents the coordinates of the image center; The actual physical size corresponding to a single pixel; By matching the workpiece coordinates with the robot coordinate system coordinates of key points using the singular value decomposition algorithm, the precise transformation relationship between the workpiece and the robot coordinate system can be solved.

2. The visual positioning method for an automated drilling robot for tire molds as described in claim 1, characterized in that, The target markings consist of nine perfectly circular, uniformly spaced, and uniformly sized circles arranged in a 3×3 pattern, pre-made at the center of the lower edge of the tire mold; The image feature template is used to identify the target surface markings. The template includes nine perfectly circular shapes of the same size and spacing, arranged in a 3×3 configuration.

3. The visual positioning method for an automated drilling robot for tire molds as described in claim 1, characterized in that, The key points include the center point of the vent hole to be drilled and eight boundary points, with the nine key points arranged in a 3×3 pattern.

4. The visual positioning method for an automated drilling robot for tire molds as described in claim 2, characterized in that, The specific physical size corresponding to the single pixel is as follows: Where D and d are the actual distance and pixel distance of any perfect circle in the target surface marking, respectively.

5. The visual positioning method for an automated drilling robot for tire molds as described in claim 1, characterized in that, Based on the camera intrinsic parameters, the workpiece coordinates and pixel coordinates marked on the target surface, the SQPnP algorithm is used to solve the initial transformation relationship of the workpiece relative to the robot coordinate system.

6. The visual positioning method for an automated drilling robot for tire molds as described in claim 1, characterized in that, The calculation of the photographic pose of key points in the robot coordinate system based on the initial transformation relationship and the workpiece coordinates of key points specifically includes: Calculate the initial position of the key points in the robot coordinate system based on the initial transformation relationship and the workpiece coordinates of the key points; Shift the initial position backward along the key point normal vector The distance is used to determine the shooting pose; among which For camera focal length, The x-coordinate of a single pixel. This represents the actual physical size corresponding to a single pixel.

7. A vision positioning system for an automated drilling robot for tire molds, characterized in that, include: The calibration module is used to calibrate the camera's intrinsic and extrinsic parameters, create image feature templates, and train the YOLO model. The workpiece coordinate extraction module is used to extract the workpiece coordinates of the target surface markings and key points on the tire mold; The template matching module is used to identify the target surface of the tire mold based on image feature templates, and obtain the pixel coordinates of the marking and the actual physical size corresponding to a single pixel; The initial solution module is used to solve the initial transformation relationship of the workpiece relative to the robot coordinate system based on the camera intrinsic parameters, the workpiece coordinates and pixel coordinates marked on the target surface; The dynamic positioning module is used to calculate the photographic pose of key points in the robot coordinate system based on the initial transformation relationship and the workpiece coordinates of key points, generate camera movement commands, drive the camera to the photographic pose to take pictures, and obtain the pixel coordinates of key points based on the pre-trained YOLO model; The coordinate calculation module is used to calculate the actual coordinates of keypoints relative to the robot coordinate system based on the pixel coordinates of the keypoints, the image pose, and the camera intrinsic parameters. Specifically, it includes: ， in, The distance the camera pose is shifted backward, and the actual coordinates of the key points relative to the robot's coordinate system are: The pose for taking the photo is ), These are the pixel coordinates of the key points; () represents the coordinates of the image center; The actual physical size corresponding to a single pixel; The precise solution module is used to match the workpiece coordinates and robot coordinates of key points using the singular value decomposition algorithm, and solve the precise transformation relationship between the workpiece and the robot coordinate system.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the visual positioning method for an automated drilling robot for tire molds as described in any one of claims 1-6.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the visual positioning method for an automated drilling robot for tire molds as described in any one of claims 1-6.

Images