Robot scraping method and system applied to machine tool working table

Abstract

The invention belongs to the field of automatic machining of machine tools, and relates to a robot scraping method and system applied to a working table of a machine tool, and the method comprises the following steps: acquiring a workpiece image, and calibrating the acquired workpiece image; performing working area identification on the workpiece image, and extracting contour information in the workpiece image to obtain a region of interest (ROI); in a region of interest (ROI), generating a pattern-shoveling contour with given parameters and a uniformly-distributed dot matrix; traversing points in the generated dot matrix, judging whether each point in the dot matrix is in the shoveling pattern contour or not, determining that the points in the shoveling pattern contour are heavy shoveling points and the points outside the shoveling pattern contour are light shoveling points, and obtaining light and heavy shoveling point information; the light and heavy shoveling point information is sent to the robot through an instruction, and the robot conducts scraping and flower shoveling work according to the light and heavy shoveling point information. Corresponding scraping contours can be selected according to actual requirements, the surface of the workbench is subjected to pattern shoveling according to the scraping contours, the automation degree is high, and the problem that the product quality is reduced due to the fact that the machine tool workbench is smooth is effectively solved.

Description

Technical Field

[0001] The invention belongs to the field of machine tool automated processing, and in particular relates to a robot scraping method and system applied to a machine tool work surface. Background Art

[0002] The scraping process is one of the key processes to ensure the accuracy of machine tools during the manufacturing process of machine tools, and is one of the key factors that determine the accuracy and mechanical properties of machine tools. The scraping process is currently operated manually throughout the industry, which is inefficient, labor-intensive, highly dependent on the work experience of the scraping workers, and has a long-term impact on physical and mental health. With the development of industrial automation and artificial intelligence technology, the conditions for industrial application of machines replacing manual scraping are ready.

[0003] At present, the surface of the machine tool worktable uses grinding technology to ensure flatness, but at the same time, the worktable surface is relatively smooth and has a low friction coefficient. When the workpiece fixture is clamped on the worktable for processing, vibration will occur, and the worktable surface is too smooth and prone to displacement. When the force between the workpiece fixture and the worktable is too large, the workpiece fixture will be deformed, making it difficult to ensure product quality. In addition, the wear of the worktable surface is difficult to detect; if the wear of the worktable surface is not detected in time, it will have a direct impact on product quality. Summary of the invention

[0004] In order to solve the defects existing in the prior art, the purpose of the present invention is to provide a robot scraping method and system for application to a machine tool work table, which can select the corresponding scraping profile according to the scraping needs in the actual production process, and then scrape the work table surface after grinding according to the scraping profile. The manual labor intensity is low and the degree of automation is high. The work table surface after scraping can effectively solve the problem of reduced product quality due to the smooth machine tool work table.

[0005] The method of the present invention is implemented by the following technical solution: A robot scraping method applied to a machine tool work surface comprises the following steps:

[0006] Collecting workpiece images and calibrating the collected workpiece images;

[0007] Identify the working area of ​​the workpiece image, extract the contour information in the workpiece image, and obtain the region of interest ROI;

[0008] In the region of interest (ROI), a scraping contour with given parameters and a uniformly distributed dot matrix are generated;

[0009] Traverse the points in the generated dot matrix to determine whether each point in the dot matrix is ​​within the scraping contour. The points within the scraping contour are heavy scraping points, and the points outside the scraping contour are light scraping points, and obtain the information of heavy and light scraping points;

[0010] The information of light and heavy shoveling points is sent to the robot through instructions, and the robot performs scraping and scraping operations based on the information of light and heavy shoveling points.

[0011] The system of the present invention is implemented by the following technical solution: A robot scraping system applied to a machine tool work surface, comprising:

[0012] An image acquisition module is used to acquire workpiece images and calibrate the acquired workpiece images;

[0013] The region recognition module is used to identify the working area of ​​the workpiece image, extract the contour information in the workpiece image, and obtain the region of interest ROI;

[0014] The contour and dot matrix generation module generates the scraping contour and evenly distributed dot matrix with given parameters in the region of interest (ROI);

[0015] The module for distinguishing light and heavy shoveling points traverses the points in the generated dot matrix to determine whether each point in the dot matrix is ​​within the scraping contour. The points within the scraping contour are heavy scraping points, and the points outside the scraping contour are light scraping points, thereby obtaining information on light and heavy scraping points.

[0016] The scraping and grinding module sends the information of light and heavy scraping points to the robot through instructions, and the robot performs scraping and grinding operations based on the information of light and heavy scraping points.

[0017] Compared with the prior art, the present invention has the following advantages:

[0018] 1. The present invention can select a corresponding scraping profile according to the scraping requirements in the actual production process, and then scrape the surface of the workbench after grinding according to the scraping profile; the workbench surface after scraping can increase the contact rigidity between the machined part and the workbench surface, effectively solve the problem that the workpiece is easy to shift due to the smooth table surface, and improve the quality of part processing, such as improving the finish of the machined parts and effectively reducing the deformation of the parts caused by over-tightening.

[0019] 2. The work surface after scraping of the present invention is easy to observe and find out whether it is worn; therefore, the wear of the work surface can be found in time, ensuring that the product quality is not affected by the wear of the surface.

[0020] 3. The present invention uses a scraped surface scraped by a robot to replace the traditional machine-processed grinding surface, which can significantly reduce the labor intensity of workers, replace heavy manual labor operations, effectively reduce the labor intensity of workers, and realize automation. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a flow chart of a robot scraping method applied to a machine tool work surface in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The present invention will be further described below in conjunction with the accompanying drawings and embodiments, but the embodiments of the present invention are not limited thereto.

[0023] Example

[0024] This embodiment provides a robot scraping method applied to a machine tool work surface, comprising the following steps:

[0025] Step S1: Use an industrial camera to capture a workpiece image, and calibrate the captured workpiece image.

[0026] The workpiece images captured by the industrial camera have high imaging quality and the resolution meets the detection requirements. The camera calibration can accurately convert the actual coordinate values ​​and pixel coordinate values ​​of the workpiece; thus, in the subsequent steps, the robot can accurately scrape the workpiece according to the scraping points obtained from the image.

[0027] Step S2: Use edge detection algorithm to identify the working area of ​​the workpiece image, extract contour information in the workpiece image, and obtain the region of interest ROI.

[0028] The identification of the working area is realized by an edge detection algorithm, and the contour information in the image is extracted to obtain the ROI. In this embodiment, the edge detection algorithm identifies the working area by calculating the gradient value of the workpiece image; the calculation formula of the gradient value of the workpiece image is as follows:

[0029] g x (i,j)=f(i+1,j+1)-f(i-1,j-1)+f(i-1,j+1)-f(i+1,j-1)+f(i,j+1)-f(i,j-1);

[0030] g y (i,j)=f(i+1,j+1)-f(i-1,j-1)-f(i-1,j+1)+f(i+1,j-1)+f(i,j+1)-f(i,j-1);

[0031]

[0032] θ=arctan(g y (i,j) / g x (i,j));

[0033] Where g x (i, j) is the X-direction gradient of the workpiece image, g y(i, j) is the Y-direction gradient of the workpiece image, i is the X-coordinate of the workpiece image, j is the Y-coordinate of the workpiece image, f(i, j) is the grayscale value of the workpiece image, g(i, j) is the gradient value of the workpiece image, and θ is the gradient direction of the workpiece image. When the workpiece image gradient value g(i, j) is greater than the set threshold, the point information is saved.

[0034] Step S3: Generate a scraping contour with given parameters and a uniformly distributed dot matrix in the ROI obtained in step S2.

[0035] In this embodiment, in order to meet various requirements in the industrial production process, the scraping contour includes shapes such as circle, rectangle, diamond, etc. The scraping contour calculates contour points according to the selected shape and step length, and sequentially generates a dot matrix with equal spacing in the X direction and the Y direction in the contour.

[0036] Step S4, traverse the points in the dot matrix generated in step S3, determine whether each point in the dot matrix is ​​within the scraping contour, take the points within the scraping contour as heavy scraping points, and the points outside the scraping contour as light scraping points, and obtain light and heavy scraping point information.

[0037] More specifically, the points with darker colors and larger areas within the scraping contour are heavy scraping points, and the points with lighter colors and smaller areas outside the scraping contour are light scraping points. The heavy scraping points have deep scraping marks, large areas, and more scraping times; the light scraping points have shallow scraping marks, small areas, and fewer scraping times.

[0038] In this embodiment, it is determined whether the point in the dot matrix of step S3 is within the scraping contour by traversing each point in the dot matrix and determining the relationship between each point and the scraping contour in turn. The algorithm is as follows:

[0039]

[0040] Where φ is the sum of the angles between the test point and all the points of the spade contour, n is the number of points in the contour, pi is the i-th point in the contour, P(i+1)%n is the i+1-th point in the contour, Pt is the test point, Pi.x is the X coordinate of the i-th point in the contour, Pi.y is the Y coordinate of the i-th point in the contour, Pt.x is the X coordinate of the test point, and Pt.y is the Y coordinate of the test point. If φ=2π, it means that the test point is inside the spade contour, and φ=0 means that it is outside the spade contour.

[0041] Step S5, after the scraping point identification is completed, the light and heavy scraping point information is sent to the robot through instructions, and the robot performs scraping and polishing operations according to the light and heavy scraping point information, that is, scraping.

[0042] The robot performs scraping and grinding tasks according to the information of light and heavy scraping points, which can realize automated scraping and grinding, effectively reduce the intensity of manual work, and solve the problem of easy displacement of workpieces with smooth table surfaces.

[0043] In addition, the robot parameters can be modified according to actual needs to adjust knife patterns of different shapes.

[0044] This embodiment also provides a robot scraping system applied to a machine tool work surface, which specifically includes the following modules:

[0045] An image acquisition module is used to acquire workpiece images and calibrate the acquired workpiece images;

[0046] The region recognition module is used to identify the working area of ​​the workpiece image, extract the contour information in the workpiece image, and obtain the region of interest ROI;

[0047] The contour and dot matrix generation module generates the scraping contour and evenly distributed dot matrix with given parameters in the region of interest (ROI);

[0048] The module for distinguishing light and heavy shoveling points traverses the points in the generated dot matrix to determine whether each point in the dot matrix is ​​within the scraping contour. The points within the scraping contour are heavy scraping points, and the points outside the scraping contour are light scraping points, thereby obtaining information on light and heavy scraping points.

[0049] The scraping and grinding module sends the information of light and heavy scraping points to the robot through instructions, and the robot performs scraping and grinding operations based on the information of light and heavy scraping points.

[0050] The above modules are respectively used to execute the various steps of the scraping method in this embodiment. The detailed execution process can be found in steps S1-S5 and will not be described in detail.

[0051] The above is only the better / preferred implementation method of the invention of the present invention, but the protection scope of the invention patent is not limited to this. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principle of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A robot scraping method applied to a machine tool work surface, characterized in that: The following steps are involved: Collecting workpiece images and calibrating the collected workpiece images; Identify the working area of ​​the workpiece image, extract the contour information in the workpiece image, and obtain the region of interest ROI; In the region of interest (ROI), a scraping contour with given parameters and a uniformly distributed dot matrix are generated; Traverse the points in the generated dot matrix to determine whether each point in the dot matrix is ​​within the scraping contour. The points within the scraping contour are heavy scraping points, and the points outside the scraping contour are light scraping points, and obtain the information of heavy and light scraping points; The information of light and heavy shoveling points is sent to the robot through instructions, and the robot performs scraping and scraping operations based on the information of light and heavy shoveling points.

2. The robot scraping method according to claim 1, characterized in that: The edge detection algorithm is used to identify the working area of ​​the workpiece image.

3. The robot scraping method according to claim 2, characterized in that: The edge detection algorithm identifies the working area by calculating the gradient value of the workpiece image; the calculation formula of the workpiece image gradient value is: g x (i,j)=f(i+1,j+1)-f(i-1,j-1)+f(i-1,j+1)-f(i+1,j-1)+f(i,j+1)-f(i,j-1); g y (i,j)=f(i+1,j+1)-f(i-1,j-1)-f(i-1,j+1)+f(i+1,j-1)+f(i,j+1)-f(i,j-1); θ=arctan(g y (i,j) / g x (i,j)); Where g x (i, j) is the X-direction gradient of the workpiece image, g y (i, j) is the Y-direction gradient of the workpiece image, i is the X-coordinate of the workpiece image, j is the Y-coordinate of the workpiece image, f(i, j) is the grayscale value of the workpiece image, g(i, j) is the gradient value of the workpiece image, and θ is the gradient direction of the workpiece image.

4. The robot scraping method according to claim 1, characterized in that: Determine whether a point in the dot matrix is ​​within the scraping contour, traverse each point in the dot matrix, and determine the relationship between each point and the scraping contour in turn: Where φ is the sum of the angles between the test point and all the points of the contour, n is the number of points in the contour, pi is the i-th point in the contour, P(i+1)%n is the i+1-th point in the contour, Pt is the test point, Pi.x is the X coordinate of the i-th point in the contour, Pi.y is the Y coordinate of the i-th point in the contour, Pt.x is the X coordinate of the test point, and Pt.y is the Y coordinate of the test point; if φ=2π, the test point is judged to be within the scraping contour, and φ=0 means that the test point is outside the scraping contour.

5. The robot scraping method according to claim 1, characterized in that: The scraped outline is circular, rectangular or diamond-shaped.

6. A robot scraping system for use on a machine tool work surface, characterized in that: include: An image acquisition module is used to acquire workpiece images and calibrate the acquired workpiece images; The region recognition module is used to identify the working area of ​​the workpiece image, extract the contour information in the workpiece image, and obtain the region of interest ROI; The contour and dot matrix generation module generates the scraping contour and evenly distributed dot matrix with given parameters in the region of interest (ROI); The module for distinguishing light and heavy shoveling points traverses the points in the generated dot matrix to determine whether each point in the dot matrix is ​​within the scraping contour. The points within the scraping contour are heavy scraping points, and the points outside the scraping contour are light scraping points, thereby obtaining information on light and heavy scraping points. The scraping and grinding module sends the information of light and heavy scraping points to the robot through instructions, and the robot performs scraping and grinding operations based on the information of light and heavy scraping points.

7. The robot scraping system according to claim 6, characterized in that: The area recognition module uses edge detection algorithm to identify the working area of ​​the workpiece image; The edge detection algorithm identifies the working area by calculating the gradient value of the workpiece image; the calculation formula of the workpiece image gradient value is: g x (i,j)=f(i+1,j+1)-f(i-1,j-1)+f(i-1,j+1)-f(i+1,j-1)+f(i,j+1)-f(i,j-1); g y (i,j)=f(i+1,j+1)-f(i-1,j-1)-f(i-1,j+1)+f(i+1,j-1)+f(i,j+1)-f(i,j-1); θ=arctan(g y (i,j) / g x (i,j)); Where g x (i, j) is the X-direction gradient of the workpiece image, g y (i, j) is the Y-direction gradient of the workpiece image, i is the X-coordinate of the workpiece image, j is the Y-coordinate of the workpiece image, f(i, j) is the grayscale value of the workpiece image, g(i, j) is the gradient value of the workpiece image, and θ is the gradient direction of the workpiece image.

8. The robot scraping system according to claim 6, characterized in that: The light and heavy shovel point discrimination module determines whether a point in the dot matrix is ​​within the shovel contour, traverses each point in the dot matrix, and determines the relationship between each point and the shovel contour in turn: Where φ is the sum of the angles between the test point and all the points of the contour, n is the number of points in the contour, pi is the i-th point in the contour, P(i+1)%n is the i+1-th point in the contour, Pt is the test point, Pi.x is the X coordinate of the i-th point in the contour, Pi.y is the Y coordinate of the i-th point in the contour, Pt.x is the X coordinate of the test point, and Pt.y is the Y coordinate of the test point; if φ=2π, the test point is judged to be within the scraping contour, and φ=0 means that the test point is outside the scraping contour.

9. The robot scraping system according to claim 6, characterized in that: The scraped outline is circular, rectangular or diamond-shaped.

Images