An automatic scraping system with a visual recognition robot

Abstract

The invention discloses an automatic glue scraping system with a visual recognition robot, which belongs to the field of glue scraping equipment, and comprises: a frame; a positioning core shaft and a first driving device for driving the positioning core shaft to rotate are arranged on the frame; a glue coating device, which comprises a glue scraping mechanism, a mechanical arm and a second driving device for driving the mechanical arm to move back and forth along the axial direction of the positioning core shaft; a visual recognition device, which is arranged on the mechanical arm; the invention obtains the diameter of the pipe to be processed by the visual recognition device, calculates the circumference of the pipe, and the mechanical arm starts to perform glue scraping on the pipe to be processed according to the recognition information of the visual recognition device, and completes a half circumference corresponding to a glue scraping roller brush as a cycle, and the second driving device starts to drive the mechanical arm to move the width of a glue scraping roller brush, and the visual recognition device re-recognizes the diameter of the pipe to be processed, starts roller brush positioning, and the next step of glue scraping starts, and the operation is cyclically performed in sequence.

Description

Technical Field

[0001] The invention relates to a glue scraping system, in particular to an automatic glue scraping system with a visual recognition robot. Background Art

[0002] In the process of manufacturing resin pipes, when winding fiberglass fiber on the core column, chemical adhesive needs to be continuously added to solidify it into a one-piece pipe. In this manufacturing process, each time a layer of fiberglass fiber is wound, excess adhesive will flow down the cylindrical surface. Chemical adhesive will pollute the environment, and the toxic gas emitted will harm workers on site, and it will also cause a waste of raw materials for the production enterprise.

[0003] At present, enterprises manufacturing pipes use manual labor to complete this process, which has several problems: first, the harsh environment brings occupational hazards such as respiratory diseases to workers; second, the manual operation is not precise enough, causing the glue to flow out and pollute the environment. Third, the manual operation is not standard, resulting in waste of glue and uneven thickness of glue on the surface of the pipe, affecting the quality of the pipe. Summary of the invention

[0004] The purpose of the present invention is to solve at least one of the technical problems existing in the prior art and to provide an automatic glue scraping system with a visual recognition robot.

[0005] The technical solution of the present invention is as follows:

[0006] An automatic scraping system with a visual recognition robot, comprising:

[0007] A frame; the frame is provided with a positioning mandrel and a first driving device for driving the positioning mandrel to rotate;

[0008] A glue coating device, comprising a glue scraping mechanism, a mechanical arm, and a second driving device for driving the mechanical arm to move back and forth along the axial direction of the positioning mandrel;

[0009] A visual recognition device is arranged on the mechanical arm.

[0010] Preferably, the first driving device includes a motor and a roller group arranged at both ends of the frame, the roller group includes at least two driving rollers, the distance between the two driving rollers is smaller than the diameter of the pipe to be processed, and the motor can drive the driving rollers to rotate.

[0011] Preferably, the second driving device includes a servo motor and a slide rail arranged parallel to the positioning core shaft, a rack is arranged at the side end of the slide rail, a gear meshing with the rack is arranged at the output end of the servo motor, a movable plate is slidably arranged on the slide rail, and the robotic arm is arranged on the movable plate.

[0012] Preferably, the visual recognition device is a visual camera.

[0013] Preferably, the robotic arm is a six-axis robotic arm, and a first rotating mechanism is provided at the front end thereof.

[0014] Preferably, the visual recognition device is arranged on the first rotating mechanism.

[0015] Preferably, the scraping mechanism is arranged on the first rotating mechanism.

[0016] Preferably, the glue scraping mechanism comprises a glue scraping roller brush and a glue storage cylinder, and the glue scraping roller brush is located above the glue storage cylinder.

[0017] Preferably, it also includes a control device, which can control the actions of the robot arm, the first drive device and the second drive device according to the recognition signal of the visual recognition device.

[0018] The present invention has at least one of the following beneficial effects:

[0019] (1) The present invention is an automatic glue scraping system with a visual recognition robot. The visual recognition device obtains the diameter of the pipe to be processed and calculates the circumference of the pipe. The robot arm starts to scrape glue on the pipe to be processed based on the recognition information of the visual recognition device. The cycle ends when the half circumference corresponding to a scraping roller brush is completed. The second drive device starts to drive the robot arm to move the width of a scraping roller brush. The visual recognition device re-identifies the diameter of the pipe to be processed and starts to position the roller brush. The next step of scraping glue begins and the operation is cyclically performed. At the same time, the problem of uneven thickness of residual glue on the pipe wall and difficulty in scraping glue is solved. At the same time, it can also be used to scrape glue on pipes of different diameters, realizing a more intelligent scraping action.

[0020] (2) The automatic glue scraping system with a visual recognition robot of the present invention is provided with a glue scraping mechanism on which a glue storage cylinder is provided, so that the scraped glue can be collected while performing the glue scraping operation, which is more conducive to centralized processing and recycling and avoids waste.

[0021] (3) The automatic scraping system with a visual recognition robot of the present invention realizes flexible scraping action through a six-axis robotic arm, and is combined with a second driving device and a control device to achieve unattended, automated and intelligent operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a flow chart of the image detection algorithm of the present invention;

[0023] Figure 2 It is the first search graph in the eight-neighborhood precise search of the present invention;

[0024] Figure 3 It is the second search graph in the eight-neighborhood precise search of the present invention;

[0025] Figure 4 This is a flowchart of the pixel edge point search program of the present invention.

[0026] Figure 5 The structure of the preferred embodiment of the present invention is shown in FIG. Figure 1 ;

[0027] Figure 6 The structure of the preferred embodiment of the present invention is shown in FIG. Figure 2 ;

[0028] Figure 7 yes Figure 6 A local enlarged schematic diagram;

[0029] Figure 8 yes Figure 6 Side view of

[0030] Fig. 9 is a schematic structural diagram of the second driving device in a preferred embodiment of the present invention;

[0031] Fig.10 yes Fig. 9 B is a partial enlarged schematic diagram;

[0032] In the figure, 100-frame, 200-positioning mandrel, 300-driving roller, 400-visual recognition device, 500-glue scraping mechanism, 501-glue scraping roller brush, 502-glue storage cylinder, 600-mechanical arm, 601-first rotating mechanism, 602-first rotating arm, 700-second driving device, 701-servo motor, 702-gear, 703-rack, 704-slide rail, 705-movable plate. DETAILED DESCRIPTION

[0033] This section will describe in detail the specific embodiments of the present invention. The preferred embodiments of the present invention are shown in the accompanying drawings. The purpose of the accompanying drawings is to supplement the description of the text part of the specification with graphics, so that people can intuitively and vividly understand each technical feature and the overall technical solution of the present invention, but it cannot be understood as a limitation on the scope of protection of the present invention.

[0034] In the description of the present invention, it should be understood that descriptions involving orientations, such as up, down, front, back, left, right, etc., and orientations or positional relationships indicated are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0035] In the description of the present invention, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed" etc. are understood as not including the number itself, and "above", "below", "within" etc. are understood as including the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing the technical features, and cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0036] In the description of the present invention, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in the present invention based on the specific content of the technical solution.

[0037] Reference Figures 1 to 10 , the preferred embodiment of the present invention:

[0038] An automatic scraping system with a visual recognition robot, comprising:

[0039] The frame 100 is provided with a positioning mandrel 200 and a first driving device for driving the positioning mandrel 200 to rotate;

[0040] A glue coating device, comprising a glue scraping mechanism 500, a mechanical arm 600, and a second driving device 700 for driving the mechanical arm 600 to move back and forth along the axial direction of the positioning mandrel 200;

[0041] The visual recognition device 400 is disposed on the robot arm 600 .

[0042] In this preferred embodiment, the diameter of the pipe to be processed is obtained through the visual recognition device 400, and the circumference of the pipe is calculated. The mechanical arm 600 starts to scrape the pipe to be processed according to the recognition information of the visual recognition device 400. The half circumference corresponding to a scraping roller brush is completed as a cycle. The second drive device 700 starts to drive the mechanical arm 600 to move, and the visual recognition device re-identifies the diameter of the pipe to be processed, starts to position the roller brush, and the next step of scraping begins. The operation is cyclically performed in sequence. When the end of the pipe corresponding to the first radius is completed, the first drive device drives the positioning mandrel 200 to rotate the pipe to be processed to the second radius surface, and the scraping operation is continued according to the above method. The pipe to be processed in this embodiment is a glass fiber reinforced plastic pipe, and the problem of uneven thickness of residual glue on the pipe wall and difficult scraping operation is solved. At the same time, it can also be applied to scraping glue on pipes of different diameters; a more intelligent scraping action is achieved.

[0043] As a preferred embodiment of the present invention, it may also have the following additional technical features:

[0044] The first driving device includes a motor and a roller group arranged at both ends of the frame 100, the roller group includes at least two driving rollers 300, the distance between the two driving rollers 300 is smaller than the diameter of the pipe to be processed, and the motor can drive the driving rollers 300 to rotate.

[0045] The second driving device 700 includes a servo motor 701 and a slide rail 704 arranged parallel to the positioning core shaft 200. A rack 703 is arranged at the side end of the slide rail 704. The output end of the servo motor 701 is provided with a gear 702 meshing with the rack 703. A movable plate 705 is slidably arranged on the slide rail 704. The robot arm 600 is arranged on the movable plate 705. This arrangement enables the robot arm 600 to be automated and move parallel to the positioning core shaft 200, thereby reducing the omission of the scraping action, and making the scraping action more precise.

[0046] The visual recognition device 400 is a visual camera, which can make the scraping of glue more accurate. The visual camera obtains the diameter of the pipe to be processed and calculates the circumference of the pipe, so that the robot arm can scrape glue accurately according to the circumference.

[0047] The robot arm 600 adopts a six-axis robot arm, which has more sensitive movement accuracy, that is, more flexibility. A first rotating mechanism 601 is provided at its front end. The six-axis robot arm 600 realizes flexible scraping action and is combined with the second driving device to realize automated operation.

[0048] The visual recognition device 400 is disposed on the first rotating mechanism 601 so that the visual recognition device 400 moves along with the first rotating mechanism 601 .

[0049] The scraping mechanism 500 is disposed on the first rotating mechanism 601 , so that the scraping mechanism 400 can move along with the first rotating mechanism 601 .

[0050] Specifically, the first rotating mechanism may be hinged to achieve up and down movement on the first rotating arm 602, or may be other forms of rotating connection, and is not limited thereto.

[0051] In some other applications, the scraper mechanism 500 includes a scraper roller brush 501 and a glue storage cylinder 502, and the scraper roller brush 501 is located above the glue storage cylinder 502. This arrangement can collect the scraped glue while performing the scraping operation, which is more conducive to centralized processing and recycling and avoids waste; specifically, the second drive device 700 drives the robotic arm 600 to move the width of a scraper roller brush 501.

[0052] It also includes a control device, which can control the actions of the robot arm 600, the first drive device and the second drive device 700 according to the recognition signal of the visual recognition device 400, specifically, it controls the actions of the robot arm 600, the motor, and the servo motor 701 to achieve automation.

[0053] Specifically, the specific scraping method of the above-mentioned scraping system based on visual recognition is:

[0054] Step 1: The visual recognition device obtains an image of the pipe to be processed, processes the image through a visual image processing algorithm to obtain the diameter D of the pipe to be processed, and calculates the circumference L of the pipe to be processed and the curvature of the circumference of the pipe to be processed; L = πD = 2πr, where: D is the diameter and r is the radius.

[0055] The curvature is the rotation rate of the tangent direction angle α at a certain point on the curve to the arc length s. It is defined by differentiation and indicates the degree to which the curve deviates from a straight line.

[0056] K=lim|Δα / Δs|, when Δs tends to 0, k is defined as the curvature.

[0057] Step 2: According to the curvature information in step 1, the scraping mechanism on the robot arm in the scraping system is adjusted to position the scraping brush at the top of the pipe to be processed. From here, it moves down along the curvature route of the pipe to be processed to the lowest point, completing the first half-circumference scraping action;

[0058] Step 3: The second driving device starts to drive the robot arm to move the width of a roller brush, and returns to step 1 for a cycle process until the first half of the surface of the pipe to be processed is scraped with glue;

[0059] Step 4: Rotate the pipe to be processed and return to step 1 to cycle until the second half of the surface of the pipe to be processed is scraped with glue.

[0060] The visual image processing algorithm specifically includes:

[0061] S1: A visual recognition device collects image information of the pipe to be processed;

[0062] S2: performing error compensation processing on the image information described in S1 to obtain a corrected image;

[0063] S3: extracting and analyzing the corrected image in S2 to obtain circumferential edge information of the pipe to be processed;

[0064] S4: Fitting and calculating the circumferential edge information in S3 to obtain the diameter of the pipe to be processed.

[0065] In S2, the specific method of error compensation is as follows:

[0066] In the field of view of the visual recognition device, a point (x, y) on the pipe to be processed is located at a point (u, v) on the image information. According to the cause of the imaging distortion error, the following relationship exists:

[0067]

[0068]

[0069] Where a ij , b ij are the coefficients of the polynomial, n is the degree of the polynomial; (i, j represent the number of iterations in the pixel domain).

[0070] In S3, the specific method of extraction and analysis is: grayscale processing is performed on the corrected image, a grayscale histogram is obtained, a grayscale threshold is set, and image segmentation is performed on the grayscale histogram; the segmented image is scanned, the pixel currently being scanned is marked, and the connectivity of the pixel and several neighboring pixels previously scanned is checked, wherein connectivity refers to whether the pixels are connected and connected into a piece, and if they are disconnected, it means that the connectivity is not good; if they are connected, they are marked with the same area symbol, and if they are not connected, they are distinguished; after completing the marking of each area, the marked area to which the image center belongs is determined, and after obtaining the marking number, the boundary image point corresponding to the area is obtained, and the preliminary edge contour of the pipe to be processed is extracted; finally, the preliminary edge contour is accurately searched to determine the precise edge contour, thereby obtaining the circumferential edge information of the pipe to be processed.

[0071] The precise search adopts eight-field precise search. The specific method is as follows:

[0072] The first search is to traverse the image to find the first non-zero pixel point, then this point must be a boundary point. Let this point be the starting point m, and search clockwise for the first non-zero pixel point m' (point m' is also a boundary point) encountered in the eight-neighborhood of this point. Let m = m', and then search clockwise in the neighborhood, and the starting point of the search is the zero point before m' in the process from m to m'.

[0073] Figure 2 In the example, at point m, the search is performed in the order of 0-1-2-3-4... 6 is the first non-zero pixel found, so point 6 in the figure is the boundary point, and the starting position of the m' search is Figure 2 The 5 in the table is 4, which corresponds to m'. The search order for m' is 4-5-6-7-0-1-2-3.

[0074] By the same token, when m' is 2 or 3, the search starting point is 0; when m' is 4 or 5, the search starting point is 2; when m' is 6 or 7, the search starting point is 4; and when m' is 0 or 1, the search starting point is 6.

[0075] It can be stored using map mapping.

[0076] Map[0]=6, map[1]=6; map[2]=0, map[3]=0; map[4]=2, map[5]=2, Map[6]=4, map[7]=4.

[0077] The value found in the previous step is d( Figure 2 In the example, d is 6), so the initial j is set to map[d] (map[6]=4, i.e. j=4;), but we need to continue searching for 0, 1, 2, and 3, so here we take the remainder after dividing by 8. d is set to (d+i) / 8 (i=0,1,2...8).

[0078] In S4, the fitting calculation is specifically as follows: according to the circumferential edge information of the pipe to be processed, the coordinates of the edge point set are determined, and the corrected coordinate point is obtained after correction and conversion as (x i ,y i ), the coordinates of the center of the circle are (A, B), and the radius of the circle is r. If there is no error, the radius can be calculated based on the equation of the circle. The specific equation is: (x i -A) 2 +(y i -B) 2 =r 2 ;

[0079] Since the actual size of the pipe to be processed cannot be an absolute circle, there must be some points that do not fall on the circumference, so optimization is required when fitting the circumference so that the fitting result is closest to the true value. The commonly used fitting method is the least squares method, and the main idea is to find the coefficients so that the residual square sum of the equation is minimized. It is also necessary to perform the least squares fitting on the corrected circumference edge information; the following formula is obtained, specifically:

[0080] When the visual acquisition device collects at least 3 feature points, the values ​​are substituted into the above formula to obtain the following matrix formula:

[0081]

[0082] After finding the optimal solution of the coefficient matrix, a, b, and c are solved, and the coordinates of the center of the circle are: Its diameter D is

[0083] Under the premise that no conflict occurs, those skilled in the art may freely combine and superimpose the above-mentioned additional technical features.

[0084] The above description is only a preferred implementation mode of the present invention. Any technical solution that achieves the purpose of the present invention by substantially the same means belongs to the protection scope of the present invention.

Claims

1. An automatic scraping system with a visual recognition robot, characterized in that: include: frame; The frame is provided with a positioning mandrel and a first driving device for driving the positioning mandrel to rotate; A glue coating device, comprising a glue scraping mechanism, a mechanical arm, and a second driving device for driving the mechanical arm to move back and forth along the axial direction of the positioning mandrel; A visual recognition device, which is arranged on the mechanical arm; The scraping method also includes the following steps: Step 1: The visual recognition device obtains an image of the pipe to be processed, processes the image through a visual image processing algorithm to obtain the diameter D of the pipe to be processed, and calculates the circumference L of the pipe to be processed and the curvature of the circumference of the pipe to be processed; Step 2: According to the curvature information in step 1, the scraping mechanism on the robot arm in the scraping system is adjusted to position the scraping brush at the top of the pipe to be processed. From here, it moves down along the curvature route of the pipe to be processed to the lowest point, completing the first half-circumference scraping action; Step 3: The second driving device starts to drive the robot arm to move the width of a roller brush, and returns to step 1 for a cycle process until the first half of the surface of the pipe to be processed is scraped with glue; Step 4: Rotate the pipe to be processed, and return to step 1 to cycle until the second half of the surface of the pipe to be processed is scraped; The visual image processing algorithm specifically includes: S1: A visual recognition device collects image information of the pipe to be processed; S2: performing error compensation processing on the image information described in S1 to obtain a corrected image; S3: extracting and analyzing the corrected image in S2 to obtain circumferential edge information of the pipe to be processed; S4: fitting and calculating the circumferential edge information in S3 to obtain the diameter of the pipe to be processed; In S2, the specific method of error compensation is as follows: In the field of view of the visual recognition device, a point (x, y) on the pipe to be processed is located at a point (u, v) on the image information. According to the cause of the imaging distortion error, the following relationship exists: Where a ij , b ij are the coefficients of the polynomial, and n is the degree of the polynomial; i, j represents the number of iterations in the pixel domain.

2. The automatic scraping system with a visual recognition robot according to claim 1, characterized in that: In S3, the specific method of extraction and analysis is: grayscale processing is performed on the corrected image, a grayscale histogram is obtained, a grayscale threshold is set, and image segmentation is performed on the grayscale histogram; the segmented image is scanned, the pixel currently being scanned is marked, and the connectivity of the pixel and several neighboring pixels previously scanned is checked, wherein connectivity refers to whether the pixels are connected and connected into a piece, and if they are disconnected, it means that the connectivity is not good; if they are connected, they are marked with the same area symbol, and if they are not connected, they are distinguished; after completing the marking of each area, the marked area to which the image center belongs is determined, and after obtaining the marking number, the boundary image point corresponding to the area is obtained, and the preliminary edge contour of the pipe to be processed is extracted; finally, the preliminary edge contour is accurately searched to determine the precise edge contour, thereby obtaining the circumferential edge information of the pipe to be processed; The precise search adopts eight-field precise search; the specific method is as follows: The first search is to traverse the image to find the first non-zero pixel point, then this point must be a boundary point. Let this point be the starting point m, and search clockwise for the first non-zero pixel point m' encountered in the eight-neighborhood of this point. Point m' is also a boundary point; let m = m', and then search clockwise in the neighborhood. The starting point of the search is the zero point just before m' in the process from m to m'; In S4, the fitting calculation is specifically as follows: according to the circumferential edge information of the pipe to be processed, the coordinates of the edge point set are determined, and the corrected coordinate point is obtained after correction and conversion as (x i ,y i ), the coordinates of the center of the circle are (A, B), and the radius of the circle is r. If there is no error, the radius can be calculated based on the equation of the circle. The specific equation is: (x i -A) 2 +(y i -B) 2 =r 2 ; Since the actual size of the pipe to be processed cannot be an absolute circle, there must be some points that do not fall on the circumference, so optimization is required when fitting the circumference to make the fitting result closest to the true value; the commonly used fitting method is the least squares method, and the main idea is to find the coefficients so that the residual square sum of the equation is minimized; it is also necessary to perform the least squares fitting on the corrected circumference edge information; the following formula is obtained, specifically: When the visual acquisition device collects at least 3 feature points, the values ​​are substituted into the above formula to obtain the following matrix formula: After finding the optimal solution of the coefficient matrix, a, b, and c are solved, and the coordinates of the center of the circle are: Its diameter D is 3. The automatic scraping system with a visual recognition robot according to claim 1, characterized in that: The first driving device includes a motor and a roller group arranged at both ends of the frame, the roller group includes at least two driving rollers, the distance between the two driving rollers is smaller than the diameter of the pipe to be processed, and the motor can drive the driving rollers to rotate.

4. The automatic scraping system with a visual recognition robot according to claim 1, characterized in that: The second driving device includes a servo motor and a slide rail arranged parallel to the positioning core shaft, a rack is arranged at the side end of the slide rail, a gear meshing with the rack is arranged at the output end of the servo motor, a movable plate is slidably arranged on the slide rail, and the mechanical arm is arranged on the movable plate.

5. The automatic scraping system with a visual recognition robot according to claim 1, characterized in that: The visual recognition device is a visual camera.

6. The automatic scraping system with a visual recognition robot according to claim 1, characterized in that: The mechanical arm is a six-axis mechanical arm, and a first rotating mechanism is arranged at the front end thereof.

7. The automatic scraping system with a visual recognition robot according to claim 6, characterized in that: The visual recognition device is arranged on the first rotating mechanism.

8. The automatic scraping system with a visual recognition robot according to claim 6, characterized in that: The scraping mechanism is arranged on the first rotating mechanism.

9. The automatic scraping system with a visual recognition robot according to claim 1, characterized in that: The scraping mechanism comprises a scraping roller brush and a glue storage cylinder, and the scraping roller brush is located above the glue storage cylinder.

10. The automatic scraping system with a visual recognition robot according to claim 1, characterized in that: It also includes a control device, which can control the actions of the robot arm, the first drive device and the second drive device according to the recognition signal of the visual recognition device.

Images