A wheel hub burr location and path planning method based on image processing

Abstract

The present invention belongs to the technical field of wheel hub grinding. The present invention uses image processing and ant colony algorithm to obtain the optimal path planning of burr grinding points, and then programs the industrial robot according to the programming rules to realize automatic grinding, which solves the problems of low grinding efficiency and poor working environment existing in the traditional manual grinding method, and ensures the stability of grinding efficiency and wheel hub quality. The present invention accurately locates the burr parts on the wheel hub based on data fitting technology, relies on simple equipment, stable and reliable positioning method, and high accuracy, thereby improving the wheel hub grinding quality. The present invention can perform path planning on the obtained grinding points based on the ant colony algorithm, so that the path and time taken by the robot terminal after grinding all the grinding points are the shortest, thereby ensuring grinding efficiency and improving wheel hub production efficiency.

Description

Technical Field

[0001] The present invention belongs to the technical field of wheel hub grinding, and in particular relates to a wheel hub burr positioning and path planning method based on image processing. Background Art

[0002] Automotive wheels are prone to burrs after low-pressure casting and machining. The polishing method currently employed by manufacturers is primarily the traditional manual polishing method from the demographic dividend era, requiring workers to use grinding tools to polish the wheels. Traditional manual polishing produces sparks, aluminum dust, and noise, seriously impacting workers' physical and mental health. Prolonged handholding of grinding tools can easily lead to finger stiffness and muscle cramps, making it difficult to guarantee polishing efficiency and consistent wheel quality. Manual wheel polishing presents issues such as instability and a high rate of defective products, seriously hindering the advancement of wheel manufacturing technology.

[0003] At present, the deburring methods include electrolytic deburring, vibration deburring, manual deburring and CNC machining deburring. The electrolytic deburring method has high removal efficiency and good quality, but the equipment cost is high and it is difficult for general enterprises to afford it; the vibration deburring method is relatively low in cost, but the removal effect is poor, and it is easy to form a more obvious R angle on the stamping surface, which affects the fitting accuracy and wear resistance of the stamping parts and cannot meet higher use requirements; the manual deburring method is highly arbitrary and easy to control, but the process is complicated, the labor intensity of workers is high, the time utilization rate is low, the cost expenditure is high, and the consistency is poor, which is not suitable for mass production; CNC machining deburring is a commonly used deburring method with high work efficiency and good consistency, but CNC machining has the problems of manual manual operation of the teaching pendant for positioning grinding, which results in complicated steps and waste of time, and the time-consuming and labor-intensive defects of indiscriminate and traversal grinding of the wheel hub. Summary of the Invention

[0004] In order to solve the above problems, the present invention provides the following technical solutions:

[0005] The present invention provides a wheel hub burr location and path planning method based on image processing, the method comprising:

[0006] (1) Extracting and recording the pixel coordinates of the standard wheel hub profile and the wheel hub with burrs from the original image of the workpiece wheel hub captured by the industrial camera;

[0007] (2) Based on the least squares method, the pixel coordinates of the burr grinding points are obtained;

[0008] (3) Based on the ant colony algorithm, path planning is performed for the burr grinding points;

[0009] (4) Programming the controller of the industrial robot according to the grinding sequence and pixel coordinate positions of the burr grinding points;

[0010] (5) The industrial robot starts grinding the wheel hub.

[0011] The burr grinding points are the pixel points on the wheel hub profile where burrs exist.

[0012] Furthermore, the process of extracting and recording the pixel coordinates of the outline of the standard wheel hub and the outline of the wheel hub with burrs in step (1) is as follows:

[0013] (1.1) Customize the reference coordinate system on the wheel hub grinding workbench, establish the image coordinate system on the original image, establish the workpiece coordinate system in the teach pendant, and make the image coordinate system and the workpiece coordinate system coincide with each other through the customized reference coordinate system;

[0014] (1.2) Perform image denoising and image enhancement on the original image of the workpiece hub to highlight the feature points of the hub edge;

[0015] (1.3) The contour edge detection algorithm based on the Canny operator is used to extract the contours of the standard wheel hub and the wheel hub with burrs;

[0016] (1.4) Use MATLAB to extract and record the pixel coordinates of the standard wheel hub and the wheel hub with burrs.

[0017] Furthermore, the method for coinciding the image coordinate system with the workpiece coordinate system is:

[0018] (1.21) Select three points in the custom reference coordinate system and capture the original image of the workpiece hub with the industrial camera;

[0019] (1.22) Extract the coordinates of three points in the custom reference coordinate system in MATLAB, establish an image coordinate system based on the three points, and make the image coordinate system coincide with the custom reference coordinate system;

[0020] (1.23) Record three points in the custom reference coordinate system in the teach pendant so that the workpiece coordinate system and the custom reference coordinate system coincide.

[0021] Furthermore, in step (2), the fitting function of the standard hub profile is obtained by least squares data fitting:

[0022]

[0023] Where D is the distance value;

[0024] V is x p Substitute the value of the hub fitting function y = f(x);

[0025] is (δx, δy), fitting binary function f(x p ,y p ) where x and y are derived to obtain δx and δy respectively;

[0026] Furthermore, the process of obtaining the pixel coordinates of the burr grinding point in step (2) is as follows:

[0027] (2.1) Substitute the pixel coordinates of the contour of the wheel with burrs into the fitting function middle;

[0028] (2.2) Set the threshold k1 and the distance value D to determine the value;

[0029] If the distance value D is greater than the threshold k1, that is, greater than the set threshold k1, then the pixel point is not on the standard wheel hub profile; the burr grinding point is the wheel hub profile pixel point with burrs;

[0030] If the distance value D is less than or equal to the threshold value k1, that is, it is not greater than the set threshold value k1, then the pixel point is on the standard wheel hub profile. A distance check is performed between the standard wheel hub profile pixel point and the pixel point. If the distance between the pixel point and any point in the standard wheel hub profile pixel point is not less than the threshold value k2, then the pixel point is a burr grinding point.

[0031] (2.3) Repeat the above steps to obtain the pixel coordinates of the burr grinding point.

[0032] Furthermore, the rules for programming the controller of the industrial robot in step (4) are:

[0033] A threshold K is set. If the distance between two adjacent burr grinding points is greater than the threshold K, the two adjacent burr grinding points are not in a continuous trajectory segment. The programming program controlled by the joint motion command is used to keep the robot end moving between the two adjacent points.

[0034] If the distance between two adjacent burr grinding points is less than the threshold value K, the two adjacent burr grinding points are in a continuous trajectory segment, and the programming adopts a programming program controlled by a linear motion instruction to keep the robot end moving between the two adjacent points.

[0035] The present invention has the following beneficial effects:

[0036] (1) The present invention uses image processing and ant colony algorithm to obtain the optimal path planning of the burr grinding point, and then programs the industrial robot according to the programming rules to realize automatic grinding, which solves the problems of low grinding efficiency and harsh working environment existing in the traditional manual grinding method, and ensures the stability of grinding efficiency and wheel quality.

[0037] (2) The present invention accurately locates the burr part on the wheel hub based on data fitting technology, relies on simple equipment, and the positioning method is stable and reliable with high accuracy, thereby improving the grinding quality of the wheel hub.

[0038] (3) The present invention can perform path planning for the obtained grinding points based on the ant colony algorithm, so that the path and time taken by the robot end after grinding all the grinding points are the shortest, thereby ensuring the grinding efficiency and improving the hub production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a flow chart of the wheel hub burr positioning and path planning method of the present invention.

[0040] Figure 2 It is a flow chart of extracting pixel coordinates of the present invention.

[0041] Figure 3 It is a schematic diagram of the workbench custom reference coordinate system and image coordinate system of the present invention.

[0042] Figure 4 It is a schematic diagram of the point method recording of the present invention.

[0043] Figure 5 It is the Canny diagram of the standard wheel hub of the present invention.

[0044] Figure 6 It is a Canny diagram of a burred wheel hub of the present invention.

[0045] Figure 7 It is the fitting function decomposition diagram a of the present invention.

[0046] Figure 8 It is the fitting function decomposition diagram b of the present invention.

[0047] Figure 9 This is the fitting function decomposition diagram c of the present invention.

[0048] Figure 10 It is the fitting function decomposition diagram d of the present invention.

[0049] Figure 11 This is a flow chart of the present invention for obtaining pixel coordinates of burr grinding points. Figure 12 This is a diagram analyzing the distance s principle of the present invention.

[0050] Figure 13 This is a collection of pixels that does not require polishing.

[0051] Figure 14 A collection of pixels for burr removal.

[0052] Figure 15 It is the flow chart of the ant colony algorithm.

[0053] Figure 16 This is a diagram of the simulation results of the ant colony algorithm of the present invention.

[0054] Figure 17 This is a graph of the X-coordinate trajectory tracking of the robot end grinding tool of the present invention.

[0055] Figure 18 It is a Y-coordinate trajectory tracking curve diagram of the robot end grinding tool of the present invention. DETAILED DESCRIPTION

[0056] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings. It should be pointed out that the embodiments are only specific explanations of the invention and should not be regarded as limitations of the invention. The purpose of the embodiments is to enable those skilled in the art to better understand and reproduce the technical solutions of the present invention. The scope of protection of the present invention shall still be based on the scope defined by the claims.

[0057] like Figure 1 As shown, the present invention provides a wheel hub burr location and path planning method based on image processing, the method comprising:

[0058] S1, extracting and recording the pixel coordinates of the outline of the standard wheel hub and the outline of the wheel hub with burrs from the original image of the workpiece wheel hub captured by the industrial camera;

[0059] like Figure 2 As shown in the figure, the process of extracting and recording the pixel coordinates of the standard wheel hub outline and the wheel hub with burrs is as follows:

[0060] S11, customize the reference coordinate system on the wheel hub grinding workbench, establish the image coordinate system on the original image, establish the workpiece coordinate system in the teach pendant, and make the image coordinate system and the workpiece coordinate system coincide with each other through the customized reference coordinate system;

[0061] The method for coinciding the image coordinate system and the workpiece coordinate system is as follows:

[0062] On the wheel hub grinding workbench, customize the reference coordinate system and select points X1, X2, and Y1 on the customized reference coordinate system. Collect the customized reference coordinate system and points X1, X2, and Y1 into the original image of the workpiece hub. In the original image of the workpiece hub, establish an image coordinate system based on points X1, X2, and Y1 so that the customized reference coordinate system and the image coordinate system coincide with each other. Figure 3 As shown; record the three points X1, X2 and Y1 in the teaching pendant, as shown Figure 4 As shown, the workpiece coordinate system is established, the workpiece coordinate system is made to coincide with the custom coordinate system, and finally the image coordinate system is made to coincide with the workpiece coordinate system.

[0063] S12, performing image denoising and image enhancement processing on the original image of the workpiece hub to highlight the edge feature points of the hub;

[0064] The image is denoised by bilateral filtering, the denoised image is enhanced by logarithmic transformation, and then the image is binarized to obtain a black and white binary image of the image;

[0065] S13, the contour edge detection algorithm based on the Canny operator extracts the contour of the standard wheel hub and the contour of the wheel hub with burrs, respectively. Figure 5 and Figure 6 As shown;

[0066] S14, extracting and recording the pixel coordinates of the standard wheel profile and the wheel profile with burrs by MATLAB;

[0067] The pixels of the outline of the standard wheel hub and the outline of the wheel hub with burrs are both white pixels.

[0068] After extracting the wheel edge using the Canny operator, we can obtain the Canny edge images of the standard wheel and the wheel with burrs. The coordinates of the white pixels in the standard wheel are extracted using the find function in Matlab.

[0069] Since white pixels can be identified using the built-in find function in MATLAB, the key code is [row, col] = find (Picture == 1), where Picture is the extracted picture.

[0070] The three-dimensional pixel points are converted into two-dimensional pixel points through Picture=Picture(:,:,1), thereby obtaining a two-dimensional pixel point set of the wheel hub edge.

[0071] S2, based on the least squares method, the pixel coordinates of the burr grinding points are obtained;

[0072] The commonly used solution for fitting models is the least squares mathematical method. The specific steps of the least squares method are:

[0073] 1. Select the fitting function;

[0074] 2. Determine the coefficients in the fitting function;

[0075] 3. Select the optimal fitting function based on the distance between the pixel and the fitting function. The smaller the distance, the better the fitting function. The decomposition diagram of the fitting function is shown in Figure 7-10.

[0076] The wheel hub can be fitted with the function y = f(x), which is transformed to z = f(x, y). If the point (x1, y2) is on the wheel hub edge fitting function, then z = f(x, y) = 0. If it is not on the contour, then |z = f(x, y)| > 0. For the binary function z = f(x, y), find a point P(x p , y p ) to the distance of the function can be performed as follows: first calculate f(x p ,y p ), and then find the partial derivatives of x and y to get δx and δy respectively, then the gradient of z is = (δx, δy). Since the length of the gradient can be regarded as the rate of change of z in this direction, that is Finally, the distance D can be Calculated.

[0077] By fitting the data using the least squares method, the fitting function of the standard wheel hub profile is obtained as follows:

[0078]

[0079] Where D is the distance value;

[0080] V is x p Substitute the value of the hub fitting function y = f(x);

[0081] is (δx, δy), fitting binary function f(x p ,y p ) where x and y are derived to obtain δx and δy respectively;

[0082] like Figure 11 As shown in the figure, the process of obtaining the pixel coordinates of the burr grinding point is as follows:

[0083] S21, substitute the pixel coordinates of the contour of the wheel with burrs into the fitting function middle;

[0084] Then, the coordinates of the pixel point A(x a ,y a ) are substituted into the distance formula to calculate and get the distance value. Figure 12 As shown, the solid line part is the fitted binary function f(x,y), where A(x a ,y a ) point is the point to be tested, and the distance s in the figure is the distance from the point to the function.

[0085] S22, setting the threshold k1 and the distance value D to determine the value;

[0086] The specific value of the threshold k1 is related to the image pixels and the specific size of the wheel. The reason for setting the threshold k1 is that the fitting function cannot fully and accurately reflect the actual situation of the standard wheel profile, and there may be deviations in the verification of pixel points. Therefore, a threshold k1 is set to perform range judgment.

[0087] If the distance value D is greater than the threshold k1, that is, greater than the set threshold k1, then the pixel point is not on the standard wheel hub profile;

[0088] If the distance value D is less than or equal to the threshold value k1, that is, it is not greater than the set threshold value k1, then the pixel point is on the standard wheel hub profile, and the pixel points obtained that are not greater than the threshold value k1 are presented in the image, which is the set of pixel points that do not need to be polished in the wheel hub profile image with burrs, as shown in Figure 13.

[0089] Perform a distance check between the standard wheel hub outline pixel and the pixel point (the pixel point that does not need to be polished in the wheel hub outline with burrs). If the distance between the pixel point and any point in the standard wheel hub outline pixel point is not less than the threshold k2, then the pixel point is a burr polishing point, such as Figure 14 As shown;

[0090] S23, repeat the above steps to obtain the pixel coordinates of the burr grinding point.

[0091] S3, based on the ant colony algorithm, performs path planning for the burr grinding points;

[0092] Ant colony algorithm is used to perform path planning based on the obtained polishing points. The flow chart of the ant colony algorithm is as follows: Figure 15 As shown, the grinding path sequence and the shortest path distance of the grinding points are obtained, as shown in Figure 16 shown.

[0093] S4, programming the controller of the industrial robot according to the grinding order and pixel coordinate positions of the burr grinding point plan;

[0094] The rules for programming the controller of an industrial robot are:

[0095] A threshold K is set. If the distance between two adjacent burr grinding points is greater than the threshold K, the two adjacent burr grinding points are not in a continuous trajectory segment. The programming program controlled by the joint motion command is used to keep the robot end moving between the two adjacent points.

[0096] If the distance between two adjacent burr grinding points is less than the threshold value K, the two adjacent burr grinding points are in a continuous trajectory segment, and the programming program using linear motion instruction control keeps the robot end moving between the two adjacent points.

[0097] S5, the industrial robot starts grinding the wheel hub.

[0098] Verify that the trajectory of the robot end in the present invention is the trajectory obtained according to the path planning, and track and record the actual grinding point coordinates of the robot end to obtain the following Figure 17 and Figure 18 The trajectory tracking diagram shown in the figure shows that the coordinates of the actual trajectory points are basically the same as those of the ideal trajectory points, with the average absolute deviation of the X coordinate being 1.967 mm and the average absolute deviation of the Y coordinate being 1.863 mm.

[0099] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

Claims

1. A wheel hub burr location and path planning method based on image processing, characterized in that: The method comprises: (1) Extract and record the pixel coordinates of the standard wheel hub profile and the wheel hub with burrs from the original image of the workpiece wheel hub captured by the industrial camera; (2) Based on the least squares method, the pixel coordinates of the burr grinding points are obtained; By fitting the data using the least squares method, the fitting function of the standard wheel hub profile is obtained as follows: ； in, is the distance value; V is x p Substitute the value of the wheel hub fitting function y = f (x), where x p is the horizontal coordinate value of point P; is (δx, δy), fitting the binary function f (x p , y p ) where x and y are derived to obtain δx and δy respectively; The process of obtaining the pixel coordinates of the burr grinding point is as follows: (2.1) Substitute the pixel coordinates of the contour of the wheel with burrs into the fitting function middle; (2.2) Set the threshold k1 and distance value Make numerical value judgment; If the distance value D>threshold k1, that is, greater than the set threshold k1, then the pixel point is not on the standard wheel hub profile. If the distance value D is less than or equal to the threshold value k1, that is, it is not greater than the set threshold value k1, then the pixel point is on the standard wheel hub profile. A distance check is performed between the standard wheel hub profile pixel point and the pixel point. If the distance between the pixel point and any point in the standard wheel hub profile pixel point is not less than the threshold value k2, then the pixel point is a burr grinding point. (2.3) Repeat the above steps to obtain the pixel coordinates of the burr grinding points; (3) Based on the ant colony algorithm, path planning is performed for the burr grinding points; (4) Program the controller of the industrial robot according to the grinding sequence and pixel coordinate positions of the burr grinding points; (5) The industrial robot starts grinding the wheel hub.

2. The wheel hub burr location and path planning method based on image processing according to claim 1, characterized in that: The process of extracting and recording the pixel coordinates of the standard wheel hub profile and the wheel hub with burrs in step (1) is as follows: (1.1) Customize the reference coordinate system on the wheel hub grinding workbench, establish the image coordinate system on the original image, establish the workpiece coordinate system in the teach pendant, and make the image coordinate system and the workpiece coordinate system coincide with each other through the customized reference coordinate system; (1.2) Perform image denoising and image enhancement on the original image of the workpiece hub to highlight the feature points of the hub edge; (1.3) The contour edge detection algorithm based on the Canny operator extracts the contours of the standard wheel hub and the contours of the wheel hub with burrs; (1.4) Use MATLAB to extract and record the pixel coordinates of the standard wheel hub profile and the wheel hub with burrs.

3. The wheel hub burr location and path planning method based on image processing according to claim 2, characterized in that: The method for coinciding the image coordinate system and the workpiece coordinate system is as follows: (1.21) Select three points in the custom reference coordinate system and capture the original image of the workpiece hub with the industrial camera; (1.22) In MATLAB, extract the coordinates of three points in the custom reference coordinate system and establish an image coordinate system based on these three points so that the image coordinate system and the custom reference coordinate system coincide. (1.23) Record three points in the custom reference coordinate system in the teach pendant so that the workpiece coordinate system and the custom reference coordinate system coincide.

4. The wheel hub burr location and path planning method based on image processing according to claim 1, characterized in that: The rules for programming the controller of the industrial robot in step (4) are: A threshold K is set. If the distance between two adjacent burr grinding points is greater than the threshold K, the two adjacent burr grinding points are not in a continuous trajectory segment. A programming program controlled by joint motion instructions is used to keep the robot end moving between the two adjacent points. If the distance between two adjacent burr grinding points is less than the threshold value K, the two adjacent burr grinding points are in a continuous trajectory segment, and the programming adopts a programming program controlled by a linear motion instruction to keep the robot end moving between the two adjacent points.

Images