Automatic polishing method and device for free edge of ship part and related equipment

Abstract

The invention relates to the technical field of part processing, and discloses an automatic polishing method and device for free edges of ship parts and related equipment.The automatic polishing method comprises the steps that overall point cloud data are obtained through photographing, point cloud data of the ship parts to be polished are extracted, and the point cloud data of the ship parts to be polished are converted into pixel images; according to a pre-created classification library, classification and recognition are carried out, categories are determined, image edges of to-be-polished ship parts are extracted based on pixel images, the to-be-polished ship parts are corrected based on the categories and the image edges, and coordinates of to-be-polished points of the to-be-polished ship parts are determined based on the categories and the corrected to-be-polished ship parts. And a polishing track is generated based on the to-be-polished points, and the mechanical arm is controlled for polishing. According to the method, the point cloud data of the part is obtained and converted into the pixel image, the category and the image edge of the part are determined based on the pixel image, the part is corrected according to the category and the image edge, the coordinates of the point to be polished are determined, and finally the polishing track is determined and polished.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of part processing, and in particular to a ship part free edge automatic polishing method and device and related equipment. BACKGROUND

[0002] In the ship industry, due to the fact that the ship sails in the marine environment for a long time, the internal space is always in a special working atmosphere with high salt and high humidity. Under this environment, salt mist formed by evaporation of seawater will adhere to the surface of the ship part, and the moisture in the air will also continuously penetrate, which is extremely easy to cause corrosion of the part. The free edge of the ship part, i.e. the outermost part in the ship design, which is not connected with other structures of the ship body and has not been specially treated, is a "disaster area" of corrosion. Once the free edge is corroded, not only will it cause the structural strength of the part to decrease, affecting the overall safety and stability of the ship, but also it may cause a chain failure, increasing the maintenance cost and sailing risk of the ship. Therefore, professional polishing treatment of the free edge of the ship part is a key process to block the corrosion path and ensure the service life of the ship part.

[0003] However, for a long time, the ship industry has mainly relied on traditional manual polishing methods for the free edge of the part. The limitations of this method are obvious: in terms of polishing quality, the effect of manual polishing completely depends on the experience, technical level and working state of the operator. There are differences in polishing force, angle and speed among different operators, and even the same operator may cause deviation in polishing standards due to fatigue after a long time of work, resulting in uneven smoothness and difficulty in unifying the size accuracy of the free edge of the processed ship part, poor uniformity, hidden dangers for subsequent corrosion prevention treatment, and serious impact on the assembly accuracy and overall performance of the ship part. More importantly, the working environment of manual polishing poses a great threat to human health. A large amount of metal dust and grinding wheel debris is generated during the polishing process, and these fine particles diffuse in the air. After long-term inhalation by the operator, respiratory system diseases such as pneumoconiosis and bronchitis are easily caused, seriously endangering the health and safety of front-line workers. With the development of the ship industry towards large-scale, high-end and intelligentization, higher requirements are put forward for the processing accuracy, quality stability and production efficiency of ship parts, and the traditional manual polishing method has been difficult to meet the development needs of the industry. Therefore, how to realize automatic polishing of the free edge of the ship part is a direction that people have been researching. SUMMARY

[0004] Therefore, the present application provides a ship part free edge automatic polishing method and device and related equipment to facilitate automatic polishing of the free edge of the ship part.

[0005] In order to achieve the above-mentioned purpose, the present scheme is as follows: A free edge automatic polishing method of a ship part, comprising: Triggering a mechanical arm to grab and take a picture, to obtain a grabbing area image; Based on the grabbing area image, determining a grabbing point, and controlling the mechanical arm to grab the ship part to be polished and place it in the shooting area; Taking a picture of the ship part to be polished to obtain overall point cloud data; Extracting the point cloud data of the ship part to be polished from the overall point cloud data; Judging whether the point cloud data of the ship part to be polished is available; If yes, converting the point cloud data of the ship part to be polished into a pixel image; According to a pre-created classification library, classifying and identifying the pixel image to determine the category of the ship part to be polished; Based on the pixel image, extracting the image edge of the ship part to be polished; Based on the category and the image edge, aligning the ship part to be polished to obtain an aligned ship part to be polished; Based on the category and the aligned ship part to be polished, determining the coordinates of the points to be polished on the free edge of the ship part to be polished; Judging whether the coordinates of the points to be polished are incorrect; If no, based on the points to be polished, generating a free edge polishing track and a knife position covering all the polishing areas of the ship part to be polished; Based on the free edge polishing track and the knife position, controlling the mechanical arm to polish.

[0006] Optionally, the extracting of the point cloud data of the ship part to be polished from the overall point cloud data comprises: Extracting the point cloud data of the ship part to be polished and the magnet device from the overall point cloud data; Filtering the point cloud data of the ship part to be polished and the magnet device to remove the surrounding outliers; Segmenting and obtaining the point cloud data of the magnet plane from the filtered point cloud data of the ship part to be polished and the magnet device; Obtaining the minimum circumscribed cube corner coordinates of the point cloud data of the magnet plane; Removing the point cloud data inside the space cube composed of the minimum circumscribed cube corner coordinates; Using the curvature difference between the point cloud data of the ship part to be polished and other remaining point cloud data to extract the point cloud data of the ship part to be polished.

[0007] Optionally, converting the point cloud data of the ship parts to be polished into pixel images includes: Create a white background of fixed size; The point cloud data is projected onto a white background in black to obtain a pixel image.

[0008] Optionally, the step of aligning the ship part to be polished based on the category and image edges to obtain the aligned ship part to be polished includes: Based on the image edges, determine the pixel coordinates of each edge; Based on the category and the pixel coordinates of each side, the straightening direction is determined; The ship parts to be polished are aligned according to the alignment direction to obtain the aligned ship parts to be polished.

[0009] Optionally, determining the coordinates of the grinding point on the free edge of the ship part to be ground, based on the category and the aligned part, includes: Based on the category, retrieve the algorithm for determining the coordinates of the grinding points on the free edge of the ship parts to be ground from the pre-created classification library; Based on the aligned ship parts to be polished, determine the coordinates of each side of the aligned ship parts to be polished; Based on the determination algorithm and the coordinates of each side of the ship part to be polished after alignment, the coordinates of the polishing point on the free side of the ship part to be polished are determined.

[0010] Optionally, generating a free-edge grinding trajectory and tool entry position covering the entire grinding area of ​​the ship part to be ground, based on the coordinates of the point to be ground, includes: Based on the coordinates of the points to be polished, a free edge polishing trajectory covering the entire polishing area of ​​the ship parts to be polished is generated using B-spline curves. The cutting position is determined based on the free edge grinding trajectory.

[0011] Optionally, taking photos of the ship parts to be polished to obtain overall point cloud data includes: The ship parts to be polished are photographed to obtain initial overall point cloud data; Remove outliers around the initial overall point cloud data to obtain the overall point cloud data.

[0012] An automated grinding device for free edges of ship parts, comprising: The grasping and photographing module is used to trigger the robotic arm to grasp and photograph the vision, thereby obtaining an image of the grasped area; The parts gripping module is used to determine the gripping point based on the image of the gripping area, and control the robotic arm to grip the ship parts to be polished and place them in the photo area; The overall point cloud data acquisition module is used to photograph the ship parts to be polished and acquire overall point cloud data. The part point cloud data acquisition module is used to extract the point cloud data of the ship part to be polished from the overall point cloud data. The part point cloud data detection module is used to determine whether the point cloud data of the ship part to be polished is available; A pixel image conversion module is used to convert the point cloud data of the ship parts to be polished into pixel images when the point cloud data of the ship parts to be polished is available. The parts category determination module is used to classify and identify the pixel images according to a pre-created classification library to determine the category of the ship parts to be polished; An image edge extraction module is used to extract the image edges of the ship parts to be polished based on the pixel image; The parts alignment module is used to align the ship parts to be polished based on the category and image edges, so as to obtain the aligned ship parts to be polished. The grinding point determination module is used to determine the coordinates of the grinding point on the free edge of the ship part to be ground based on the category and the aligned ship part to be ground. The grinding point detection module is used to determine whether the coordinates of the grinding point are incorrect; The grinding data generation module is used to generate a free edge grinding trajectory and cutter position covering the entire grinding area of ​​the ship part to be ground, based on the coordinates of the grinding point when the coordinates of the grinding point are correct. The free edge grinding module is used to control the robotic arm to perform grinding based on the free edge grinding trajectory and the cutting position.

[0013] An automated grinding device for free edges of ship parts includes: a memory and a processor; The memory is used to store programs; The processor is used to execute the program to implement each step of the automated grinding method for the free edge of ship parts as described above.

[0014] A readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the various steps of any of the aforementioned automated grinding methods for the free edges of ship parts.

[0015] As can be seen from the above technical solution, the automated grinding method for free edges of ship parts provided in this application includes: triggering a robotic arm to grasp and take a visual photograph to obtain an image of the grasping area; determining the grasping point based on the grasping area image; controlling the robotic arm to grasp the ship part to be ground and place it in the photographing area; taking a photograph of the ship part to be ground to obtain overall point cloud data; extracting the point cloud data of the ship part to be ground from the overall point cloud data; determining whether the point cloud data of the ship part to be ground is usable; if so, converting the point cloud data of the ship part to be ground into a pixel image and processing it according to a pre-created classification. This application uses a library to classify and identify pixel images, determining the category of the ship parts to be polished. Based on the pixel images, it extracts the image edges of the ship parts to be polished. Based on the category and image edges, it aligns the ship parts to be polished, obtaining the aligned ship parts. Based on the category and aligned ship parts, it determines the coordinates of the points to be polished on the free edges of the ship parts. It checks whether the coordinates of the points to be polished are incorrect. If not, it generates a free edge polishing trajectory and a cutting position covering the entire area to be polished on the ship parts. Based on the free edge polishing trajectory and cutting position, it controls a robotic arm to perform polishing. This application acquires point cloud data of the ship parts to be polished by taking pictures and converting it into pixel images. Based on the pixel images, it determines the category and image edges of the ship parts to be polished. Based on the category and image edges, it aligns the ship parts to be polished and determines the coordinates of the points to be polished. Finally, it determines the polishing trajectory and polishes the free edges, realizing automated polishing of the free edges of ship parts.

[0016] Furthermore, for each ship part to be polished, this application determines the final polishing point coordinates and polishing trajectory based on the actual image edge. Compared with pre-setting the polishing path, this can reduce the impact of manufacturing deviations to a certain extent and avoid over-polishing or under-polishing. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0018] Figure 1 A flowchart of an automated grinding method for free edges of ship parts provided in this application embodiment; Figure 2 A schematic diagram of an automated grinding device for free edges of ship parts provided in this application embodiment; Figure 3This is a hardware structure block diagram of an automated grinding equipment for the free edge of ship parts, provided in an embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] Figure 1 A flowchart of an automated grinding method for free edges of ship parts is provided as an embodiment of this application, with reference to... Figure 1 As shown, the method may include the following steps: Step S100: Trigger the robotic arm to capture a visual image and obtain an image of the grasping area.

[0021] Specifically, a binocular structured light camera can be used to photograph the surface of the ship parts to be polished. Compared to 2D vision, which requires stable lighting and a clean bottom plate, 3D vision obtained using a binocular structured light camera does not have particularly high environmental requirements. Moreover, the accuracy of 2D vision is not as high as that of 3D vision, and 3D vision can reduce the recognition error of the three-dimensional contour of free edges to a certain extent.

[0022] Step S101: Based on the image of the gripping area, determine the gripping point, and control the robotic arm to grip the ship parts to be polished and place them in the photo area.

[0023] Specifically, the gripping area image obtained in the above steps can be used to determine the gripping point of the ship parts to be polished, thereby controlling the robotic arm to grip the ship parts to be polished and place them in the photo area.

[0024] Step S102: Take photos of the ship parts to be polished to obtain overall point cloud data.

[0025] Specifically, by triggering a binocular structured light 3D camera to take pictures, the RGB image captured by the camera provides the X and Y coordinates in the pixel coordinate system, while the depth map directly provides the Z coordinate in the camera coordinate system, which is the distance between the camera and the point. Based on the information of the RGB-D image and the camera's intrinsic parameters, the coordinates of any pixel in the camera coordinate system can be calculated. Each coordinate represents a point, and the set of points constitutes the point cloud, thus forming point cloud data.

[0026] Specifically, after taking pictures of the ship parts to be polished, a pass-through filter can be used to extract the point cloud of the ship parts to be polished, the magnet, and the magnet mounting platform as the initial overall point cloud data. At this time, there will still be a large number of outliers around the initial overall point cloud data. Therefore, a statistical outlier filter can be used to remove the outliers around the initial overall point cloud data, and finally obtain the overall point cloud data.

[0027] Step S103: Extract the point cloud data of the ship parts to be polished from the overall point cloud data.

[0028] Specifically, the overall point cloud data can include: point cloud data of the ship parts to be polished, point cloud data of the magnet plane, point cloud data of the magnet side, and point cloud data of the magnet mounting platform. Due to the variety of ship workpieces and the edge bevel slope, the ship parts to be polished and the magnet plane are sometimes very closely attached, making it difficult to segment the point cloud data.

[0029] Step S104: Determine whether the point cloud data of the ship parts to be polished is available.

[0030] Specifically, if the point cloud data of the ship parts to be polished is available, then step S105 is executed; if the point cloud data of the ship parts to be polished is not available, then the process returns to step S100.

[0031] Step S105: Convert the point cloud data of the ship parts to be polished into pixel images.

[0032] Specifically, there are many methods for converting point cloud data into pixel images. The following describes one such method. This application provides a method for converting point cloud data of a ship part to be polished into a pixel image. First, a white background of a fixed size is created. Then, the point cloud data of the ship part to be polished is projected onto the white background in black to obtain a pixel image of the ship part to be polished.

[0033] Step S106: Based on the pre-created classification library, classify and identify the pixel images to determine the category of the ship parts to be polished.

[0034] Specifically, the classification library is formed by putting the images of each type of workpiece into a folder, with each folder representing a type of workpiece, and then putting these folders into another folder.

[0035] Step S107: Extract the image edges of the ship parts to be polished based on the pixel image; specifically, this is achieved by combining the boundary function with the pixel image, which generates the boundary of the data through the "α-shape" algorithm.

[0036] Step S108: Based on the category and image edge, straighten the ship parts to be polished to obtain the straightened ship parts to be polished.

[0037] Specifically, during the pre-creation of the classification library, the orientation of each category of ship parts can be determined. By simply adjusting the ship parts to be polished to the orientation, the corresponding category's algorithm can be directly called to process their trajectory, thereby ensuring the accuracy of the free edge polishing of ship parts to a certain extent.

[0038] Step S109: Based on the category and the aligned ship parts to be polished, determine the coordinates of the polishing points on the free edge of the ship parts to be polished.

[0039] Step S110: Determine if the coordinates of the point to be polished are incorrect.

[0040] Specifically, by determining whether the coordinates of the point to be polished obtained in step S109 are empty, if the coordinates of the point to be polished are not empty, that is, the coordinates of the point to be polished are correct, then step S111 is executed; if the coordinates of the point to be polished are empty, that is, the coordinates of the point to be polished are incorrect, then the robotic arm is controlled to grab the incorrect workpiece.

[0041] Step S111: Based on the coordinates of the point to be polished, generate a free edge polishing trajectory and cutter position covering the entire polishing area of ​​the ship part to be polished.

[0042] Specifically, when generating the free-edge grinding trajectory and tool entry position covering the entire grinding area of ​​the ship part to be ground based on the coordinates of the grinding point, it is necessary to ensure that the free-edge grinding trajectory covers all grinding points and that the grinding trajectory transitions smoothly to a certain extent, while preventing tool collisions or reaching the limits of the robotic arm. Based on this, B-spline curves can be used to generate the free-edge grinding trajectory covering the entire grinding area of ​​the ship part to be ground, and the tool entry position can be determined based on the free-edge grinding trajectory.

[0043] Step S112: Based on the free edge grinding trajectory and the cutting position, control the robotic arm to perform grinding.

[0044] Specifically, based on the free edge grinding trajectory and cutting position determined in the above steps, a command is sent to control the robotic arm to grind the free edge of the ship part to be ground.

[0045] As can be seen from the above technical solution, the automated grinding method for free edges of ship parts provided in this application includes: triggering a robotic arm to grasp and take a visual photograph to obtain an image of the grasping area; determining the grasping point based on the grasping area image; controlling the robotic arm to grasp the ship part to be ground and place it in the photographing area; taking a photograph of the ship part to be ground to obtain overall point cloud data; extracting the point cloud data of the ship part to be ground from the overall point cloud data; determining whether the point cloud data of the ship part to be ground is usable; if so, converting the point cloud data of the ship part to be ground into a pixel image and processing it according to a pre-created classification. This application uses a library to classify and identify pixel images, determining the category of the ship parts to be polished. Based on the pixel images, it extracts the image edges of the ship parts to be polished. Based on the category and image edges, it aligns the ship parts to be polished, obtaining the aligned ship parts. Based on the category and aligned ship parts, it determines the coordinates of the points to be polished on the free edges of the ship parts. It checks whether the coordinates of the points to be polished are incorrect. If not, it generates a free edge polishing trajectory and a cutting position covering the entire area to be polished on the ship parts. Based on the free edge polishing trajectory and cutting position, it controls a robotic arm to perform polishing. This application acquires point cloud data of the ship parts to be polished by taking pictures and converting it into pixel images. Based on the pixel images, it determines the category and image edges of the ship parts to be polished. Based on the category and image edges, it aligns the ship parts to be polished and determines the coordinates of the points to be polished. Finally, it determines the polishing trajectory and polishes the free edges, realizing automated polishing of the free edges of ship parts.

[0046] Furthermore, for each ship part to be polished, this application determines the final polishing point coordinates and polishing trajectory based on the actual image edge. Compared with pre-setting the polishing path, this can reduce the impact of manufacturing deviations to a certain extent and avoid over-polishing or under-polishing.

[0047] In some embodiments of this application, step S103, the process of extracting the point cloud data of the ship parts to be polished from the overall point cloud data, may include: S11. Extract the point cloud data of the ship parts to be polished and the magnet device from the overall point cloud data.

[0048] Specifically, the magnet device may include a magnet and a magnet mounting platform, and the point cloud data of the magnet device may include point cloud data of the magnet plane, point cloud data of the magnet side, and point cloud data of the magnet mounting platform.

[0049] S12. Filter the point cloud data of the ship parts and magnet devices to be polished to remove outliers.

[0050] S13. From the filtered point cloud data of the ship parts to be polished and the magnet device, segment and obtain the point cloud data of the magnet plane.

[0051] Specifically, Euclidean clustering can be used to obtain point cloud data of the magnet plane.

[0052] S14. Obtain the coordinates of the smallest outer cube corner of the point cloud data of the magnet plane.

[0053] Specifically, based on the point cloud data of the acquired magnet plane, the corner coordinates of its smallest circumscribed cube are determined.

[0054] S15. Remove point cloud data inside the spatial cube formed by the coordinates of the corner points of the smallest circumscribed cube.

[0055] Specifically, after removing the data, the remaining data are the point cloud data of the ship parts to be polished, the point cloud data of the side of the magnet, and the point cloud data of the magnet mounting platform.

[0056] S16. Using the curvature difference between the point cloud data of the ship part to be polished and other remaining point cloud data, the point cloud data of the ship part to be polished is extracted.

[0057] Specifically, since the ship parts to be polished are in very close contact with the magnet plane, directly extracting the point cloud data of the ship parts to be polished from the point cloud data of the ship parts to be polished and the magnet device by using the curvature difference method is very likely to include the point cloud data of the magnet plane. Therefore, it is necessary to first remove the point cloud data of the magnet plane, and then use the curvature difference between the point cloud data of the ship parts to be polished and other remaining point cloud data to extract the point cloud data of the ship parts to be polished.

[0058] In some embodiments of this application, step S108, the process of aligning the ship parts to be polished based on category and image edges to obtain the aligned ship parts to be polished, may include: S21. Based on the image edges, determine the pixel coordinates of each edge.

[0059] Specifically, a coordinate system can be established with the top left corner as the origin, the x-axis pointing to the right, and the y-axis pointing downwards. Based on the created coordinate system, the position of each edge is determined based on the image edge, and the pixel coordinates of each edge are determined.

[0060] S22. Determine the orientation based on the category and the pixel coordinates of each side.

[0061] Specifically, the above steps determine the category of the ship parts to be ground, thus obtaining the pre-established alignment state corresponding to that category. Based on the pixel coordinates of each side, it can be determined which side of the ship parts to be ground belongs to which category. According to the determined correspondence, the alignment direction can be determined. The alignment direction can include the rotation angle and rotation direction around each coordinate axis.

[0062] S23. According to the alignment direction, align the ship parts to be polished to obtain the aligned ship parts to be polished.

[0063] Specifically, based on the alignment direction determined in the above steps, the ship parts to be polished are aligned so that the placement of the ship parts to be polished is consistent with the alignment status of the corresponding category in the classification library, which facilitates subsequent processing.

[0064] In some embodiments of this application, step S109, the process of determining the coordinates of the grinding point on the free edge of the ship part to be ground based on the category and the aligned part to be ground, may include: S31. From the pre-created classification library, based on the category, retrieve the algorithm for determining the coordinates of the points to be polished on the free edge corresponding to the ship parts to be polished. This algorithm can correct the image of the ship parts, then calculate the corner points (i.e., corner points) in the edge trajectory points through Harris corner detection, obtain the coordinates of these corner points, and then remove the non-polishing points.

[0065] Specifically, during the creation of the classification library, an algorithm can be generated for each category to determine the coordinates of the grinding points on the free edges of the ship parts to be ground. This allows the coordinates of the grinding points on the ship parts to be ground to be determined using the algorithm after the coordinates of each edge of the aligned ship parts have been determined.

[0066] S32. Based on the aligned ship parts to be ground, determine the coordinates of each side of the aligned ship parts to be ground.

[0067] Specifically, based on the aligned ship parts to be polished obtained in the above steps, the coordinate values ​​of each side of the aligned ship parts to be polished can be determined. Using the coordinate values ​​of each side, the coordinate values ​​of the points to be polished can be determined.

[0068] S33. Based on the determination algorithm and the coordinates of each side of the ship part to be polished after alignment, determine the coordinates of the polishing point on the free side of the ship part to be polished.

[0069] Specifically, since the free edges of each type of ship part are different, different algorithms exist for determining the grinding points on the free edges of different ship parts. The algorithm combines the coordinates of each edge of the aligned ship part to determine the coordinates of the grinding points on the free edges. Each trajectory is generated specifically for the actual ship part to be ground using a defined algorithm and actual coordinates. Compared to pre-programmed fixed trajectories, this approach can address manufacturing deviations in ship parts and, to some extent, avoid under-grinding or even more serious accidents.

[0070] Based on the automated grinding method for free edges of ship parts provided in the embodiments of this application, this application can also provide an automated grinding device for free edges of ship parts. The automated grinding device for free edges of ship parts provided in the embodiments of this application is described below. The automated grinding device for free edges of ship parts described below can be referred to in correspondence with the automated grinding method for free edges of ship parts described above.

[0071] Figure 2 This is a schematic diagram of an automated grinding device for the free edges of ship parts, provided in an embodiment of this application. (Refer to...) Figure 2 As shown, the automated grinding device for the free edges of ship parts may include: The grasping and photographing module 101 is used to trigger the robotic arm to grasp and photograph the vision, thereby obtaining an image of the grasping area. The parts gripping module 102 is used to determine the gripping point based on the gripping area image, and control the robotic arm to grip the ship parts to be polished and place them in the photo area. The overall point cloud data acquisition module 103 is used to take pictures of the ship parts to be polished and acquire overall point cloud data. The part point cloud data acquisition module 104 is used to extract the point cloud data of the ship parts to be polished from the overall point cloud data. The part point cloud data detection module 105 is used to determine whether the point cloud data of the ship parts to be polished is available; The pixel image conversion module 106 is used to convert the point cloud data of the ship parts to be polished into pixel images when the point cloud data of the ship parts to be polished is available. The part category determination module 107 is used to classify and recognize pixel images and determine the category of the ship parts to be polished based on a pre-created classification library. Image edge extraction module 108 is used to extract the image edges of the ship parts to be polished based on pixel images; The part alignment module 109 is used to align the ship parts to be polished based on the category and image edge to obtain the aligned ship parts to be polished. The grinding point determination module 110 is used to determine the coordinates of the grinding point on the free edge of the ship part to be ground based on the category and the aligned ship part to be ground. The grinding point detection module 111 is used to determine whether the coordinates of the grinding point are incorrect; The grinding data generation module 112 is used to generate a free edge grinding trajectory and cutter position covering the entire grinding area of ​​the ship parts to be ground, based on the coordinates of the grinding point when the coordinates of the grinding point are correct. The free edge grinding module 113 is used to control the robotic arm to perform grinding based on the free edge grinding trajectory and the cutting position.

[0072] As can be seen from the above technical solutions, the automated grinding device for free edges of ship parts provided in this application includes: a gripping and photographing module 101, used to trigger a robotic arm to grip and photograph the image of the gripping area; a part gripping module 102, used to determine the gripping point based on the image of the gripping area, and control the robotic arm to grip the ship part to be ground and place it in the photographing area; an overall point cloud data acquisition module 103, used to photograph the ship part to be ground and acquire overall point cloud data; a part point cloud data acquisition module 104, used to extract the point cloud data of the ship part to be ground from the overall point cloud data; a part point cloud data detection module 105, used to determine whether the point cloud data of the ship part to be ground is available; a pixel image conversion module 106, used to convert the point cloud data of the ship part to be ground into a pixel image when the point cloud data of the ship part to be ground is available; and a part category determination module 107, used to... The system performs classification and recognition on pixel images, determining the category of the ship parts to be polished based on a pre-created classification library; an image edge extraction module 108 extracts the image edges of the ship parts to be polished based on pixel images; a part alignment module 109 aligns the ship parts to be polished based on the category and image edges to obtain the aligned ship parts to be polished; a point to be polished determination module 110 determines the coordinates of the points to be polished on the free edges of the ship parts to be polished based on the category and the aligned ship parts to be polished; a point to be polished detection module 111 determines whether the coordinates of the points to be polished are incorrect; a polishing data generation module 112 generates a free edge polishing trajectory and cutting position covering the entire polishing area of ​​the ship parts to be polished based on the coordinates of the points to be polished when the coordinates of the points to be polished are correct; and a free edge polishing module 113 controls the robotic arm to perform polishing based on the free edge polishing trajectory and cutting position. This application acquires point cloud data of ship parts to be polished by taking pictures and converting it into pixel images. Based on the pixel images, the category and image edges of the ship parts to be polished are determined. Based on the category and image edges, the ship parts to be polished are aligned and the coordinates of the points to be polished are determined. Finally, the polishing trajectory is determined and the free edges are polished, thus realizing automated polishing of the free edges of ship parts.

[0073] Furthermore, for each ship part to be polished, this application determines the final polishing point coordinates and polishing trajectory based on the actual image edge. Compared with pre-setting the polishing path, this can reduce the impact of manufacturing deviations to a certain extent and avoid over-polishing or under-polishing.

[0074] Optionally, the process of extracting the point cloud data of the ship part to be polished from the overall point cloud data by the part point cloud data acquisition module 104 may include: Extract point cloud data of the ship parts to be polished and the magnet device from the overall point cloud data; The point cloud data of the ship parts and magnet devices to be polished is filtered to remove surrounding outliers; From the filtered point cloud data of the ship parts to be polished and the magnet device, the point cloud data of the magnet plane is segmented and obtained; Obtain the coordinates of the smallest outer cube corner of the point cloud data of the magnet plane; Remove point cloud data from the interior of the spatial cube, which is composed of the coordinates of the corner points of the smallest circumscribed cube; The point cloud data of the ship parts to be polished is extracted by using the curvature difference between the point cloud data of the parts to be polished and the remaining point cloud data.

[0075] Optionally, the pixel image conversion module 106 may perform the process of converting the point cloud data of the ship part to be polished into a pixel image, which may include: Create a white background of fixed size; The point cloud data is projected onto a white background in black to obtain a pixel image.

[0076] Optionally, the process by which the part alignment module 109 aligns the ship part to be polished based on the category and image edges to obtain the aligned ship part to be polished may include: Based on the image edges, determine the pixel coordinates of each edge; Based on the category and the pixel coordinates of each side, the straightening direction is determined; The ship parts to be polished are aligned according to the alignment direction to obtain the aligned ship parts to be polished.

[0077] Optionally, the process by which the grinding point determination module 110 determines the coordinates of the grinding point on the free edge of the ship part to be ground based on the category and the aligned part to be ground may include: An algorithm for determining the coordinates of the grinding points on the free edges of the ship parts to be ground, based on the categories from a pre-created classification library; Based on the aligned ship parts to be polished, determine the coordinates of each side of the aligned ship parts to be polished; Based on the algorithm for determining the coordinates of the points to be polished and the coordinates of each side of the ship part to be polished after alignment, the coordinates of the points to be polished on the free side of the ship part to be polished are determined.

[0078] Optionally, the grinding data generation module 112 may perform a process of generating a free-edge grinding trajectory and cutter position covering the entire grinding area of ​​the ship part to be ground based on the coordinates of the point to be ground, which may include: Based on the coordinates of the points to be polished, a free edge polishing trajectory covering the entire polishing area of ​​the ship parts to be polished is generated using B-spline curves. The cutting position is determined based on the free edge grinding trajectory.

[0079] Optionally, the process of the overall point cloud data acquisition module 103 taking pictures of the ship parts to be polished and acquiring overall point cloud data may include: The ship parts to be polished are photographed to obtain initial overall point cloud data; Remove outliers around the initial overall point cloud data to obtain the overall point cloud data.

[0080] This application also provides an automated grinding device for the free edges of ship parts. Figure 3 The hardware structure block diagram of the automated grinding equipment for free edges of ship parts is shown. (Refer to...) Figure 3 The hardware structure of the automated grinding equipment for free edges of ship parts may include: at least one processor 1, at least one communication interface 2, at least one memory 3, and at least one communication bus 4. In this embodiment of the application, the number of processor 1, communication interface 2, memory 3, and communication bus 4 is at least one, and processor 1, communication interface 2, and memory 3 communicate with each other through communication bus 4; Processor 1 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. Memory 3 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk storage device; The memory stores a program, and the processor can call the program stored in the memory. The program is used to implement each processing step in the aforementioned automated grinding method for free edges of ship parts.

[0081] This application embodiment also provides a storage medium that can store a program suitable for processor execution, the program being used to implement each processing step in the aforementioned automated grinding method for free edges of ship parts.

[0082] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0083] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined with each other, and the same or similar parts can be referred to each other.

[0084] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An automated grinding method for the free edges of ship parts, characterized in that, include: Trigger the robotic arm to capture a visual image of the grasped area; Based on the image of the grasping area, the grasping point is determined, and the robotic arm is controlled to grasp the ship parts to be polished and place them in the photo area. Take photos of the ship parts to be polished to obtain overall point cloud data; Extract the point cloud data of the ship parts to be polished from the overall point cloud data; Determine whether the point cloud data of the ship parts to be polished is available; If so, the point cloud data of the ship parts to be polished is converted into pixel images; The pixel images are classified and identified according to a pre-created classification library to determine the category of the ship parts to be polished; Based on the pixel image, extract the image edges of the ship part to be polished; Based on the category and image edges, the ship parts to be polished are aligned to obtain the aligned ship parts to be polished; Based on the aforementioned category and the aligned ship parts to be polished, determine the coordinates of the polishing points on the free edges of the ship parts to be polished; Determine if the coordinates of the point to be polished are incorrect; If not, then based on the point to be polished, generate a free edge polishing trajectory and cutter position covering the entire polishing area of ​​the ship part to be polished; Based on the free edge grinding trajectory and the cutting position, the robotic arm is controlled to perform grinding.

2. The method according to claim 1, characterized in that, The step of extracting the point cloud data of the ship part to be polished from the overall point cloud data includes: Extract the point cloud data of the ship parts to be polished and the magnet device from the overall point cloud data; The point cloud data of the ship parts to be polished and the magnet device are filtered to remove surrounding outliers; From the filtered point cloud data of the ship parts to be polished and the magnet device, the point cloud data of the magnet plane is segmented and obtained. Obtain the coordinates of the smallest outer cube corner of the point cloud data of the magnet plane; Remove the point cloud data inside the spatial cube formed by the coordinates of the corner points of the minimum circumscribed cube; The point cloud data of the ship part to be polished is extracted by using the curvature difference between the point cloud data of the ship part to be polished and other remaining point cloud data.

3. The method according to claim 1, characterized in that, The step of converting the point cloud data of the ship parts to be polished into pixel images includes: Create a white background of fixed size; The point cloud data is projected onto a white background in black to obtain a pixel image.

4. The method according to claim 1, characterized in that, The process of aligning the ship parts to be polished based on the category and image edges to obtain the aligned ship parts to be polished includes: Based on the image edges, determine the pixel coordinates of each edge; Based on the category and the pixel coordinates of each side, the straightening direction is determined; The ship parts to be polished are aligned according to the alignment direction to obtain the aligned ship parts to be polished.

5. The method according to claim 1, characterized in that, The determination of the coordinates of the grinding point on the free edge of the ship part to be ground, based on the category and the aligned part, includes: Based on the category, retrieve the algorithm for determining the coordinates of the grinding points on the free edge of the ship parts to be ground from the pre-created classification library; Based on the aligned ship parts to be polished, determine the coordinates of each side of the aligned ship parts to be polished; Based on the determination algorithm and the coordinates of each side of the ship part to be polished after alignment, the coordinates of the polishing point on the free side of the ship part to be polished are determined.

6. The method according to claim 1, characterized in that, The step of generating a free-edge grinding trajectory and cutter position covering the entire grinding area of ​​the ship part to be ground, based on the coordinates of the point to be ground, includes: Based on the coordinates of the points to be polished, a free edge polishing trajectory covering the entire polishing area of ​​the ship parts to be polished is generated using B-spline curves. The cutting position is determined based on the free edge grinding trajectory.

7. The method according to claim 1, characterized in that, The step of photographing the ship parts to be polished to obtain overall point cloud data includes: The ship parts to be polished are photographed to obtain initial overall point cloud data; Remove outliers around the initial overall point cloud data to obtain the overall point cloud data.

8. An automated grinding device for the free edges of ship parts, characterized in that, include: The grasping and photographing module is used to trigger the robotic arm to grasp and photograph the vision, thereby obtaining an image of the grasped area; The parts gripping module is used to determine the gripping point based on the image of the gripping area, and control the robotic arm to grip the ship parts to be polished and place them in the photo area; The overall point cloud data acquisition module is used to photograph the ship parts to be polished and acquire overall point cloud data. The part point cloud data acquisition module is used to extract the point cloud data of the ship part to be polished from the overall point cloud data. The part point cloud data detection module is used to determine whether the point cloud data of the ship part to be polished is available; A pixel image conversion module is used to convert the point cloud data of the ship parts to be polished into pixel images when the point cloud data of the ship parts to be polished is available. The parts category determination module is used to classify and identify the pixel images according to a pre-created classification library to determine the category of the ship parts to be polished; An image edge extraction module is used to extract the image edges of the ship parts to be polished based on the pixel image; The parts alignment module is used to align the ship parts to be polished based on the category and image edges, so as to obtain the aligned ship parts to be polished. The grinding point determination module is used to determine the coordinates of the grinding point on the free edge of the ship part to be ground based on the category and the aligned ship part to be ground. The grinding point detection module is used to determine whether the coordinates of the grinding point are incorrect; The grinding data generation module is used to generate a free edge grinding trajectory and cutter position covering the entire grinding area of ​​the ship part to be ground, based on the coordinates of the grinding point when the coordinates of the grinding point are correct. The free edge grinding module is used to control the robotic arm to perform grinding based on the free edge grinding trajectory and the cutting position.

9. An automated grinding equipment for the free edges of ship parts, characterized in that, include: Memory and processor; The memory is used to store programs; The processor is used to execute the program to implement each step of the automated grinding method for free edges of ship parts as described in any one of claims 1-7.

10. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements each step of the automated grinding method for free edges of ship parts as described in any one of claims 1-7.

Images