A wind power blade internal glue removal mechanical arm control method based on visual guidance

Abstract

The application provides a wind power blade internal glue removal mechanical arm control method based on visual guidance, relates to an industrial robot, and comprises the following steps: S1, obtaining a first color image by collecting an image in front of the glue removal mechanism through an RGB camera, and obtaining a second color image and a depth image by collecting an image of a wind power blade leading edge through a depth camera; S2, converting the first and second color images to an HSV color space, and segmenting a target area through an OTSU method, if no residual glue is included in the contour in the first target area, executing S3, otherwise executing S4; S3, calculating a residual glue position according to the second target area and the depth image, and controlling the mechanical arm to drive the glue removal mechanism to approach the residual glue; and S4, determining a pixel coordinate of the residual glue center according to the first target area, and controlling the mechanical arm to drive the glue removal mechanism to remove the residual glue. The application has the beneficial effect that the residual glue is accurately recognized based on color features and spatial features, and the glue removal mechanical arm can accurately control the removal of the residual glue in real time.

Description

Technical Field

[0001] This invention relates to the field of industrial robot technology, and in particular to a vision-guided control method for a robotic arm for removing adhesive from inside wind turbine blades. Background Technology

[0002] Wind turbine blades are made by bonding shells together with adhesive. Adhesive is manually applied to the web, main beam cap, and leading and trailing edges. After the blade is molded using tooling, excess adhesive overflows. A significant amount of adhesive is squeezed out and hangs at the bonding corner edge at the leading edge. The narrow space at the blade tip makes it impossible for workers to clean up the excess adhesive. This residual adhesive falls during blade operation, causing noise, damaging baffles and the blade tip shell, harming the lightning protection system, and clogging drainage holes, resulting in significant quality hazards. The residual adhesive on the inner leading edge of the blade is distributed in a spatial curve. While a glue removal robot can reach the narrow areas at the blade tip that are inaccessible to manual work, accurately controlling the robotic arm to reach the glue removal mechanism and precisely removing the adhesive remains a challenge. Summary of the Invention

[0003] In view of this, in order to solve the problems of controlling the robotic arm to drive the adhesive removal mechanism to the residual adhesive position and to remove adhesive accurately during the adhesive removal operation of the adhesive removal robot, the embodiments of the present invention provide a vision-guided control method for the robotic arm for adhesive removal inside wind turbine blades.

[0004] Embodiments of the present invention provide a vision-guided control method for a robotic arm for removing adhesive from inside wind turbine blades. The robotic arm is mounted on a mobile platform and has an adhesive removal mechanism at its end. The control method includes the following steps:

[0005] S1. Obtain a first color image by acquiring an image of the front of the adhesive removal mechanism using an RGB camera, and obtain a second color image and a depth image by acquiring an image of the leading edge of the wind turbine blade using a depth camera.

[0006] S2. Convert the first color image and the second color image to the HSV color space, and segment the target region using the maximum inter-class variance method to obtain the first target region and the second target region respectively. If the contour within the first target region does not contain residual adhesive, proceed to step S3; otherwise, proceed to step S4.

[0007] S3. Calculate the location of residual adhesive based on the second target area and the depth image, and control the robotic arm to drive the adhesive removal mechanism to approach the residual adhesive.

[0008] S4. Determine the pixel coordinates of the residual adhesive center based on the first target area, and control the robotic arm to drive the adhesive removal mechanism to remove the adhesive.

[0009] Further, in step S2, the method for converting the first color image and the second color image to the HSV color space is as follows: each pixel in the color image is converted from the RGB color space to the HSV color space.

[0010]

[0011]

[0012] v = max

[0013] Where r, g, and b are the red, green, and blue pixel values ​​of the three channels in the RGB color space, respectively; max is the maximum value among the three channels r, g, and b; min is the minimum value among the three channels r, g, and b; and h, s, and v are the hue, saturation, and brightness of the three channels in the HSV color space.

[0014] Furthermore, before converting the first color image to the HSV color space, a section of wind turbine blade base plate image and a section of residual adhesive image are spliced ​​onto both sides of the first color image.

[0015] Furthermore, the target region is segmented using the Otsu's method, including:

[0016] Extract the H channel of the HSV image and perform median filtering on the H channel image;

[0017] The Otsu's method is used to perform adaptive threshold segmentation on the filtered image to obtain a binary image mask.

[0018] Extract the contours of the mask image, calculate the area enclosed by each contour in the image, and select the contour with the largest area as the contour of the target region.

[0019] Furthermore, the method for determining whether the outline within the first target area includes residual adhesive is as follows: if a portion of the outline within the first target area is within the first color image before splicing, then it includes residual adhesive; otherwise, it does not include residual adhesive.

[0020] Furthermore, the method for calculating the location of residual adhesive based on the second target region and the depth image includes:

[0021] S3.1 Calculate the minimum bounding rectangle of the second target area. Take p pixels above the lower boundary of the minimum bounding rectangle as the center line of the residual adhesive. Extract n points as candidate points on the center line near the front of the adhesive removal mechanism.

[0022] S3.2, For the i-th candidate point, its pixel coordinates are (u i ,v i The depth value corresponding to the same pixel coordinates in the depth image is Z. iWhen the depth value Z i When the value is 0, the candidate point is discarded, and the intrinsic parameters (f) of the depth camera are considered. x ,f y The three-dimensional spatial coordinates (X, u0, v0) of the valid candidate points are obtained. i ,Y i Z i ):

[0023]

[0024]

[0025] S3.3 Calculate the center of the valid candidate points:

[0026]

[0027] Where m≤n, m is the number of effective points, and the center point is the position center of the residual glue in front of the glue removal mechanism.

[0028] Furthermore, the robotic arm has translational and rotational degrees of freedom, and controlling the robotic arm to drive the adhesive removal mechanism closer to the residual adhesive specifically includes:

[0029] S3.4. Based on the three-dimensional spatial position of the residual adhesive center and the relative position of the depth camera and the base coordinate system of the robotic arm, determine the position of the residual adhesive relative to the base coordinate system of the robotic arm as (x, y).

[0030] S3.5. Perform inverse kinematics solution based on the structure of the robotic arm:

[0031]

[0032] d = x - Lcosθ

[0033] Where L is the length of the robotic arm, θ is the angle between the robotic arm and the direction of movement of the degree of freedom, and d is the translation distance of the degree of freedom relative to the base coordinate system of the robotic arm.

[0034] Furthermore, the step S4 of determining the pixel coordinates of the residual adhesive center based on the first target region specifically includes:

[0035] S4.1 Calculate the minimum bounding rectangle of the first target region contour, and use the left side of the minimum bounding rectangle as the left boundary of the residual adhesive, with the pixel x-coordinate as u = u L The x-coordinate u = u is taken as q pixels to the right of the left boundary as the center of the residual adhesive. c =u L +q.

[0036] Furthermore, step S4, controlling the robotic arm to drive the adhesive removal mechanism to remove adhesive, specifically includes:

[0037] S4.2, Make the robotic arm press down with a controllable rotational degree of freedom and a constant torque;

[0038] S4.3, The preset target desired pixel horizontal coordinate is u D Constructing image errors:

[0039] e = u D -u C

[0040] If u C Within the first color image before stitching, the moving speed of the robotic arm's degree of freedom is:

[0041] v=-λe

[0042] Otherwise, v = 0.

[0043] Furthermore, the RGB camera is mounted on the upper part of the adhesive removal mechanism and faces forward, while the depth camera is mounted on the side of the mobile platform near the adhesive removal mechanism and facing the adhesive removal mechanism and its front.

[0044] The beneficial effects of the technical solution provided by the embodiments of the present invention are as follows: The present invention provides a vision-guided control method for a degumming robotic arm in wind turbine blades, which converts color images acquired by RGB cameras and depth cameras to the HSV color space and segments the target area using the maximum inter-class variance method; calculates the location of residual adhesive by combining the target area and depth value in the color image acquired by the depth camera, and guides the degumming mechanism at the end of the robotic arm to the vicinity of the residual adhesive; obtains the pixel coordinates of the center of the residual adhesive from the target area in the color image acquired by the RGB camera, and guides the degumming mechanism at the end of the robotic arm to accurately remove the adhesive; improves the control accuracy of the robotic arm by accurately identifying the residual adhesive based on color features and spatial features; adopts a coarse positioning method based on the depth camera to ensure that the residual adhesive can enter the field of view of the RGB camera with a small field of view; after the residual adhesive enters the field of view of the RGB camera, switching to a vision-guided method based on image error can control the degumming robotic arm to remove the residual adhesive in real time and accurately. Attached Figure Description

[0045] Figure 1 This is a flowchart of a vision-guided control method for a robotic arm that removes adhesive from inside wind turbine blades according to the present invention.

[0046] Figure 2 This is a schematic diagram of a glue removal robot;

[0047] Figure 3 This is a schematic diagram of the adhesive removal robot in operation;

[0048] Figure 4 This is a simplified diagram of the motion principle of a robotic arm in two-dimensional planar motion;

[0049] Figure 5 This is a diagram illustrating the processing steps of the first color image.

[0050] In the diagram: 1. Robotic arm; 2. Mobile platform; 3. Adhesive removal mechanism; 4. RGB camera; 5. Depth camera; 100. Wind turbine blade; 100a. Leading edge; 200. Residual adhesive. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be further described below with reference to the accompanying drawings. The following description presents a preferred embodiment of the various possible embodiments of the present invention, intended to provide a basic understanding of the invention, but not intended to identify key or decisive elements of the invention or to limit the scope of protection sought.

[0052] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0053] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0054] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures. Also, it should be understood that, for ease of description, the dimensions of the various parts shown in the figures are not drawn to actual scale.

[0055] In the description of this invention, it should be noted that the circuits, electronic components and modules involved in this invention are all prior art, which can be fully implemented by those skilled in the art, and need not be elaborated upon.

[0056] It should be further noted that, unless otherwise explicitly specified and limited, the terms "installation" and "connection" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0057] Please refer to Figure 1The embodiments of the present invention provide a vision-guided control method for a degumming robotic arm inside wind turbine blades, which is mainly applied to degumming robots.

[0058] like Figure 2 and 3 As shown, the adhesive removal robot mainly includes a mobile platform 2, a robotic arm 1, and an adhesive removal mechanism 3. The robotic arm 1 is mounted on the mobile platform 2, which mainly consists of a traction trolley and an adhesive removal platform mounted on the traction trolley. The mobile platform 2 can move inside the wind turbine blade 100, thereby driving the robotic arm 1 to move inside the wind turbine blade 100.

[0059] The adhesive removal mechanism 3 is installed at the end of the robotic arm 1 and is located on one side of the mobile platform 2. When the mobile platform 2 moves inside the wind turbine blade 100, the front port of the adhesive removal mechanism 3 is brought into contact with the adhesive removal surface, and the residual adhesive 200 can enter through the front port of the adhesive removal mechanism 3 and be collected by the adhesive removal mechanism 3.

[0060] The robotic arm 1 generally has translational and rotational degrees of freedom, meaning it can drive the adhesive removal mechanism 3 to rotate and translate. The translational degree of freedom is typically achieved through a lead screw drive mechanism, and the rotational degree of freedom through a gear drive mechanism. The gear drive mechanism can be driven by a ROBOTIS servo motor, thus allowing precise control over the rotation angle and torque of the robotic arm 1. In this embodiment, the robotic arm 1 can translate laterally along the moving platform 2 and rotate around the direction of the translational degree of freedom.

[0061] Furthermore, in order to ensure that the adhesive removal mechanism 3 has a passive compliant degree of freedom without drive, that is, when the adhesive removal mechanism 3 contacts the adhesive removal surface at the bottom of the leading edge 100a of the wind turbine blade 100 and the curvature of the adhesive removal surface changes, the adhesive removal mechanism 3 can adaptively adjust to ensure that the front port of the adhesive removal mechanism 3 is in contact with the adhesive removal surface.

[0062] When the adhesive removal robot moves within the wind turbine blade 100 via the mobile platform 2, and the adhesive removal mechanism 3 moves close to the inner leading edge 100a of the wind turbine blade 100 to remove residual adhesive 200 from the inner leading edge, the movement of the robotic arm is controlled by a vision-guided method for controlling the adhesive removal robotic arm within the wind turbine blade. This vision-guided method specifically includes the following steps:

[0063] S1. Obtain a first color image by acquiring an image of the front of the adhesive removal mechanism 3 through an RGB camera 4, and obtain a second color image and a depth image by acquiring an image of the leading edge 100a of the wind turbine blade 100 through a depth camera 5.

[0064] The RGB camera 4 is mounted on top of the adhesive removal mechanism 3 and faces forward, tilted downwards. When the moving platform 2 moves inside the wind turbine blade 100, if there is residual adhesive 200 in front of the adhesive removal mechanism 3, the content of the first color image includes the residual adhesive. The depth camera 5 is mounted on the moving platform 2 near the adhesive removal mechanism 3 and facing the adhesive removal mechanism 3 and its front. Because the depth camera 5 is located on one side of the adhesive removal mechanism 3, it has a wide field of view, and generally the content of the second color image includes the residual adhesive 200 and the adhesive removal mechanism 3.

[0065] The RGB camera 4 and the depth camera 5 can be flexibly selected according to the actual application scenario. For example, in this embodiment, the RGB camera 4 is an IMX415 industrial camera and the depth camera 5 is a RealSense D405 camera.

[0066] S2. Convert the first color image and the second color image to the HSV color space, and segment the target region using the maximum inter-class variance method to obtain the first target region and the second target region respectively. If the contour within the first target region does not contain residual adhesive, proceed to step S3; otherwise, proceed to step S4. Specifically:

[0067] S2.1 Before converting the first color image to the HSV color space, a section of wind turbine blade base plate image and a section of residual adhesive image are spliced ​​onto both sides of the first color image. For example, in this embodiment, a wind turbine blade base plate image with a pixel width of 200 is spliced ​​onto the left side of the first color image, and a residual adhesive image with a pixel width of 200 is spliced ​​onto the right side.

[0068] S2.2 The method for converting the stitched first color image and the second color image to the HSV color space is as follows: Each pixel in the color image is converted from the RGB color space to the HSV color space:

[0069]

[0070]

[0071] v = max

[0072] Where r, g, and b are the red, green, and blue pixel values ​​of the three channels in the RGB color space, respectively; max is the maximum value among the three channels r, g, and b; min is the minimum value among the three channels r, g, and b; and h, s, and v are the hue, saturation, and brightness of the three channels in the HSV color space.

[0073] Then, the target region is segmented using the Otsu's method in the HSV color space obtained from the first and second color images, respectively:

[0074] S2.3 Extract the H channel of the HSV image and perform median filtering on the H channel image. In this embodiment, a median filter with a kernel size of 35*35 is performed on the H channel image, which can suppress noise while maintaining the boundary contour of the residual adhesive.

[0075] S2.4. Adaptive threshold segmentation is performed on the filtered image using the maximum inter-class variance method to obtain a binary image mask;

[0076] S2.5 Extract the contours of the mask image, calculate the area enclosed by each contour in the image, and select the contour with the largest area as the contour of the target region. A threshold for the largest area contour can be set here; only contours with a larger area than the threshold are selected as the contour of the target region. The purpose of setting the area threshold is that, because the residual adhesive 200 is continuous and has a relatively large area within the wind turbine blade 100, only contours with a larger area than the threshold are considered valid target regions, rather than interfering noise. In this embodiment, if the largest area contour is greater than 1000, then the largest contour is selected as the contour of the target region. Thus, the first target region can be obtained from the first color image, and the second target region can be obtained from the second color image.

[0077] S2.6 Determine whether the first target area includes residual adhesive. If the outline of the first target area does not include residual adhesive 200, it means that there is no residual adhesive 200 in front of the adhesive removal mechanism 3, and the adhesive removal mechanism 3 needs to be moved closer to the residual adhesive 200. Then, execute step S3, and drive the adhesive removal mechanism 3 to move closer to the residual adhesive through the robotic arm 1. Otherwise, it means that there is residual adhesive 200 in front of the adhesive removal mechanism 3, and the adhesive can be removed directly. Then, execute step S4, and drive the adhesive removal mechanism 3 to remove the adhesive through the robotic arm 1.

[0078] The method for determining whether the outline within the first target area includes residual adhesive is as follows: if a portion of the outline within the first target area is within the first color image before splicing, then residual adhesive is included; otherwise, residual adhesive is not included.

[0079] S3. Calculate the location of residual adhesive based on the second target area and the depth image, and control the robotic arm to drive the adhesive removal mechanism to approach the residual adhesive.

[0080] The method for calculating the location of residual adhesive based on the second target region and the depth image includes:

[0081] S3.1 Calculate the minimum bounding rectangle of the second target area. Take p pixels above the lower boundary of the minimum bounding rectangle as the center line of the residual adhesive. Extract n points as candidate points on the center line near the front of the adhesive removal mechanism. In this embodiment, p is 70 and n is 10.

[0082] S3.2, For the i-th candidate point, its pixel coordinates are (u i ,v i The depth value corresponding to the same pixel coordinates in the depth image is Z. i When the depth value Z i When the value is 0, the candidate point is discarded, and the intrinsic parameters (f) of the depth camera are considered. x ,f y The three-dimensional spatial coordinates (X, u0, v0) of the valid candidate points are obtained. i ,Y i Z i ):

[0083]

[0084]

[0085] S3.3 Calculate the center of the valid candidate points:

[0086]

[0087] Where m≤n, m is the number of effective points, and the center point is the position center of the residual glue in front of the glue removal mechanism.

[0088] Controlling the robotic arm to drive the adhesive removal mechanism to approach the residual adhesive specifically includes:

[0089] S3.4. Based on the three-dimensional spatial position of the residual adhesive center and the relative position of the depth camera 5 and the base coordinates of the robotic arm 1, determine the position of the residual adhesive relative to the base coordinate system of the robotic arm 1 as (x, y).

[0090] S3.5, such as Figure 4 As shown, since the robotic arm 1 only moves in a two-dimensional plane, its motion principle can be simplified to two-dimensional planar motion. The inverse kinematics solution can be performed based on the structure of the robotic arm 1:

[0091]

[0092] d = x - Lcosθ

[0093] Where L is the length of the robotic arm 1, θ is the angle between the robotic arm 1 and the direction of movement of the degree of freedom, and d is the translation distance of the degree of freedom relative to the base coordinate system of the robotic arm 1.

[0094] S4. Determine the pixel coordinates of the residual adhesive center based on the first target area, and control the robotic arm 1 to drive the adhesive removal mechanism 3 to remove the adhesive.

[0095] Specifically, determining the pixel coordinates of the residual adhesive center based on the first target region includes:

[0096] S4.1 Calculate the minimum bounding rectangle of the first target region contour, and use the left side of the minimum bounding rectangle as the left boundary of the residual adhesive, with the pixel x-coordinate as u = u L The x-coordinate u = u is taken as q pixels to the right of the left boundary as the center of the residual adhesive. c =u L +q. In this embodiment, q is 200, and the corresponding x-coordinate of the residual glue center is u = u c =u L +200.

[0097] Controlling the robotic arm 1 to drive the adhesive removal mechanism 200 to remove adhesive specifically includes:

[0098] S4.2, Make the robotic arm 1 press down with a controllable rotational degree of freedom and a constant torque;

[0099] S4.3, The preset target desired pixel horizontal coordinate is u D Constructing image errors:

[0100] e = u D -u C

[0101] If u C Within the first color image before stitching, the moving speed of the robotic arm's degree of freedom is:

[0102] v=-λe

[0103] Otherwise, v = 0.

[0104] During the cleaning of the wind turbine blade 100, the adhesive removal robot moves from the root to the tip of the blade via the mobile platform 2. The bottom surface of one side of the leading edge 100a of the wind turbine blade 100 is a curved surface with constantly changing curvature, and the height and horizontal position of the residual adhesive 200 will change. The adhesive removal robot continuously controls the movement of the robotic arm 1 through the vision-guided wind turbine blade internal adhesive removal robotic arm control method. Based on the depth camera 5, the residual adhesive is coarsely located, and the robotic arm 1 is controlled to drive the adhesive removal mechanism 3 to approach the residual adhesive 200, so that the residual adhesive 200 enters the field of view of the RGB camera 4 with a small field of view. Then, the pixel coordinates of the center of the residual adhesive are obtained from the target area in the color image acquired by the RGB camera 4, which guides the adhesive removal mechanism 3 at the end of the robotic arm 1 to remove the adhesive accurately, effectively improving the control accuracy of the robotic arm 1.

[0105] In this document, the directional terms such as front, back, top, and bottom are defined based on the position of the components in the accompanying drawings and their relative positions to each other, solely for the purpose of clarity and convenience in expressing the technical solution. It should be understood that these are relative concepts and can vary depending on different methods of use and placement; the use of these directional terms should not limit the scope of protection claimed in this application.

[0106] Where there is no conflict, the embodiments and features described above can be combined with each other. The above descriptions are merely preferred embodiments of the present invention and are not intended to limit the invention. Any modifications, equivalent substitutions, or improvements 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 vision-guided control method for a robotic arm for removing adhesive from inside wind turbine blades, wherein the robotic arm is mounted on a mobile platform and has an adhesive removal mechanism at its end, characterized in that... The control method includes the following steps: S1. Obtain a first color image by acquiring an image of the front of the adhesive removal mechanism using an RGB camera, and obtain a second color image and a depth image by acquiring an image of the leading edge of the wind turbine blade using a depth camera. S2. Convert the first color image and the second color image to the HSV color space, and segment the target region using the maximum inter-class variance method to obtain the first target region and the second target region respectively. If the contour within the first target region does not contain residual adhesive, proceed to step S3; otherwise, proceed to step S4. S3. Calculate the location of residual adhesive based on the second target area and the depth image, and control the robotic arm to drive the adhesive removal mechanism to approach the residual adhesive. S4. Determine the pixel coordinates of the residual adhesive center based on the first target area, and control the robotic arm to drive the adhesive removal mechanism to remove the adhesive. The step S4, which involves determining the pixel coordinates of the residual adhesive center based on the first target region, specifically includes: S4.1 Calculate the minimum bounding rectangle of the first target region contour, and use the left side of the minimum bounding rectangle as the left boundary of the residual adhesive, wherein the x-coordinate of the left boundary pixel is... The x-coordinate of the center of the residual adhesive is q pixels to the right of the left boundary. ; The step S4, which involves controlling the robotic arm to drive the adhesive removal mechanism to remove adhesive, specifically includes: S4.2, Make the robotic arm press down with a controllable rotational degree of freedom and a constant torque; S4.3, The preset target desired pixel x-coordinate is Constructing image errors: ； like Within the first color image before stitching, the moving speed of the robotic arm's degree of freedom is: ;in; This is the proportional gain coefficient; otherwise .

2. The vision-guided robotic arm control method for degumming inside wind turbine blades as described in claim 1, characterized in that, In step S2, the method for converting the first color image and the second color image to the HSV color space is as follows: Each pixel in the color image is converted from the RGB color space to the HSV color space. ； ； ； Where r, g, and b are the red, green, and blue pixel values ​​of the three channels in the RGB color space, respectively; max is the maximum value among the three channels r, g, and b; min is the minimum value among the three channels r, g, and b; and h, s, and v are the hue, saturation, and brightness of the three channels in the HSV color space.

3. The vision-guided robotic arm control method for degumming inside wind turbine blades as described in claim 2, characterized in that: Before converting the first color image to the HSV color space, the process also includes splicing a section of wind turbine blade base plate image and a section of residual adhesive image onto both sides of the first color image.

4. The vision-guided robotic arm control method for degumming inside wind turbine blades as described in claim 2, characterized in that, The target region segmented using the Otsu's inter-class variance method includes: Extract the H channel of the HSV image and perform median filtering on the H channel image; The Otsu's method is used to perform adaptive threshold segmentation on the filtered image to obtain a binary image mask. Extract the contours of the mask image, calculate the area enclosed by each contour in the image, and select the contour with the largest area as the contour of the target region.

5. The vision-guided robotic arm control method for degumming inside wind turbine blades as described in claim 4, characterized in that, The method for determining whether the outline within the first target area includes residual adhesive is as follows: if a portion of the outline within the first target area is within the first color image before splicing, then residual adhesive is included; otherwise, residual adhesive is not included.

6. The vision-guided robotic arm control method for degumming inside wind turbine blades as described in claim 1, characterized in that, The method for calculating the location of residual adhesive based on the second target region and the depth image includes: S3.1 Calculate the minimum bounding rectangle of the second target area. Take p pixels above the lower boundary of the minimum bounding rectangle as the center line of the residual adhesive. Extract n points as candidate points on the center line near the front of the adhesive removal mechanism. S3.2, For the i-th candidate point, its pixel coordinates are: The depth value corresponding to the same pixel coordinates in the depth image is When the depth value If the candidate point is not found, then the intrinsic parameters of the depth camera are considered. Obtain the three-dimensional spatial coordinates of the valid candidate points : ； ； S3.3 Calculate the center of the valid candidate points: ； in m represents the number of effective points, and the center point is used as the center of the residual glue position in front of the glue removal mechanism.

7. The vision-guided robotic arm control method for degumming inside wind turbine blades as described in claim 6, characterized in that: The robotic arm has translational and rotational degrees of freedom. Controlling the robotic arm to drive the adhesive removal mechanism to approach the residual adhesive specifically includes: S3.

4. Based on the three-dimensional spatial position of the residual adhesive center and the relative position of the depth camera and the base coordinates of the robotic arm, determine the position of the residual adhesive relative to the base coordinate system of the robotic arm as (x, y). S3.

5. Perform inverse kinematics solution based on the structure of the robotic arm: ； ； Where L is the length of the robotic arm, θ is the angle between the robotic arm and the direction of movement of the degree of freedom, and d is the translation distance of the degree of freedom relative to the base coordinate system of the robotic arm.

8. The vision-guided robotic arm control method for degumming inside wind turbine blades as described in claim 1, characterized in that: The RGB camera is mounted on the upper part of the adhesive removal mechanism and faces forward, while the depth camera is mounted on the side of the mobile platform near the adhesive removal mechanism and facing the adhesive removal mechanism and its front.

Images