A method and device for trajectory planning of steel plate bevel grinding

Abstract

The application discloses a kind of trajectory planning methods and devices of steel plate bevel polishing, first identify steel plate point cloud data and convert into steel plate CAD number, then steel plate point cloud data is segmented into steel plate plane point cloud data, according to the edge information in steel plate CAD number, edge estimation is carried out to steel plate plane point cloud data, and the point cloud set of intersection of steel plate plane and bevel surface is obtained, according to the initial polishing trajectory point generated, finally according to initial polishing pose generated by initial polishing trajectory point and correction transformation matrix, final polishing pose is obtained. By calculating the point cloud of the intersection of the steel plate plane and the bevel surface, the accuracy of the bevel polishing can be improved. At the same time, the polishing trajectory is planned in combination with the steel plate CAD and the bevel information to improve the adaptability to complex bevel shapes. Automatic acquisition of steel plate point cloud can save steel plate tool positioning time, adapt to different steel plate polishing path planning, improve the automation degree of polishing workstation, and improve the bevel polishing efficiency.

Description

Technical Field

[0001] This application relates to the field of machining technology, and more specifically, to a trajectory planning method and apparatus for beveling steel plates. Background Technology

[0002] Beveling of steel plates involves using various beveling tools to process the surface of steel plates, achieving specific shapes and dimensions to meet diverse application requirements. One common type of beveling is grinding. Beveling improves the surface quality and lifespan of steel plates. Beveling results in a smoother surface with less roughness, enhanced wear and corrosion resistance, and increased durability during application. Furthermore, beveling can alter the shape of the steel plate to meet different needs. For example, for steel plates requiring cutting or bending, beveling can change their shape and size to better suit specific application scenarios.

[0003] Steel plate beveling and grinding trajectory planning, as one of the key technologies in intelligent grinding workstations, plays an indispensable role in modern welding engineering. Its importance is growing, primarily because it serves as a crucial step in improving welding quality and the overall performance of the workpiece.

[0004] However, in existing technologies, steel plate beveling is typically performed manually or semi-automatically. Manual operation requires experienced technicians, while semi-automatic equipment lacks intelligence in trajectory planning and control, leading to slow processing speeds and low efficiency. Traditional grinding methods may also result in insufficient processing accuracy due to operator inexperience, operational errors, and inaccurate tooling positioning, affecting the overall performance and reliability of the workpiece. Furthermore, traditional methods struggle to handle bevels with complex geometries, easily leading to processing dead angles and uneven surfaces, which impact welding quality and the overall appearance of the workpiece. Therefore, addressing these issues has become one of the research directions for those skilled in the art. Summary of the Invention

[0005] The purpose of this application is to overcome the shortcomings of existing technologies and provide a trajectory planning method and apparatus for steel plate beveling. By calculating the point cloud at the intersection of the steel plate plane and the bevel surface, the accuracy of beveling can be improved. At the same time, by combining steel plate CAD and bevel information to plan the grinding trajectory, the adaptability to complex bevel shapes can be improved. The automatic acquisition of steel plate point cloud can save the steel plate tooling positioning time, adapt to different steel plates to plan grinding paths, improve the automation level of the grinding workstation, and improve the beveling grinding efficiency.

[0006] The objective of this application is achieved through the following technical solution:

[0007] Firstly, this application proposes a trajectory planning method for beveling steel plates, the method comprising:

[0008] Identify the point cloud data of the steel plate and convert the point cloud data of the steel plate into a CAD number of the steel plate;

[0009] The steel plate point cloud data is segmented into steel plate planar point cloud data;

[0010] Based on the bevel edge information in the steel plate CAD number, edge estimation is performed on the steel plate plane point cloud data to obtain the point cloud set at the intersection of the steel plate plane and the bevel surface.

[0011] The initial grinding trajectory points are generated by aggregating the points at the intersection of the steel plate plane and the bevel surface.

[0012] The final grinding pose is obtained from the initial grinding pose generated from the initial grinding trajectory points and the correction transformation matrix.

[0013] In one possible implementation, the step of identifying steel plate point cloud data and converting the steel plate point cloud data into steel plate CAD numbers includes:

[0014] Point cloud data of the steel plate is acquired using a 3D camera;

[0015] The point cloud data is converted to the robot's base coordinate system and then subjected to noise reduction processing.

[0016] Steel plate point cloud data is obtained by extracting steel plate data from point cloud data.

[0017] The point cloud data of the steel plate is converted into a two-dimensional steel plate image and template matching is performed to obtain the CAD number of the steel plate.

[0018] In one possible implementation, the step of segmenting the steel plate point cloud data into steel plate planar point cloud data includes:

[0019] Obtain the angle α between the normal to each point in the steel plate point cloud data and the z-axis of the base coordinate system. i ;

[0020] Angle α i The point cloud data of the steel plate is obtained by comparing it with the separation threshold.

[0021] In one possible implementation, the step of performing edge estimation on the point cloud data of the steel plate plane based on the bevel edge information in the steel plate CAD number to obtain the point cloud set at the intersection of the steel plate plane and the bevel surface includes:

[0022] The bevel surface projection width is calculated based on the bevel edge information in the steel plate CAD number.

[0023] Edge estimation is performed on the planar point cloud data of the steel plate to obtain the edge point cloud of the steel plate;

[0024] Calculate the shortest distance from the point cloud at the edge of the steel plate to the edge information of the bevel, and obtain the point cloud set at the intersection of the steel plate plane and the bevel surface based on the shortest distance and the projection width of the bevel surface.

[0025] In one possible implementation, the step of generating initial grinding trajectory points based on the point cloud at the intersection of the steel plate plane and the bevel surface includes:

[0026] The point cloud at the intersection of the steel plate plane and the bevel surface is obtained by fitting the point cloud at the intersection according to the edge type of the steel plate.

[0027] The point cloud at the junction is divided into initial grinding trajectory points, which include straight grinding trajectory points, circular grinding trajectory points and arc grinding trajectory points.

[0028] Secondly, this application proposes a trajectory planning device for beveling steel plates, the device comprising:

[0029] The identification module is used to identify the point cloud data of the steel plate and convert the point cloud data of the steel plate into the CAD number of the steel plate;

[0030] The segmentation module is used to segment the steel plate point cloud data into steel plate planar point cloud data;

[0031] The edge estimation module is used to perform edge estimation on the point cloud data of the steel plate plane based on the bevel edge information in the steel plate CAD number to obtain the point cloud set at the intersection of the steel plate plane and the bevel surface.

[0032] The trajectory point generation module is used to generate initial grinding trajectory points based on the point cloud at the intersection of the steel plate plane and the bevel surface;

[0033] The grinding pose generation module is used to obtain the final grinding pose based on the initial grinding pose generated from the initial grinding trajectory points and the correction transformation matrix.

[0034] In one possible implementation, the identification module is further configured to:

[0035] Point cloud data of the steel plate is acquired using a 3D camera;

[0036] The point cloud data is converted to the robot's base coordinate system and then subjected to noise reduction processing.

[0037] Steel plate point cloud data is obtained by extracting steel plate data from point cloud data.

[0038] The point cloud data of the steel plate is converted into a two-dimensional steel plate image and template matching is performed to obtain the CAD number of the steel plate.

[0039] In one possible implementation, the segmentation module is further configured to:

[0040] Obtain the angle α between the normal to each point in the steel plate point cloud data and the z-axis of the base coordinate system. i ;

[0041] Angle α i The point cloud data of the steel plate is obtained by comparing it with the separation threshold.

[0042] In one possible implementation, the edge estimation module is further configured to:

[0043] The bevel surface projection width is calculated based on the bevel edge information in the steel plate CAD number.

[0044] Edge estimation is performed on the planar point cloud data of the steel plate to obtain the edge point cloud of the steel plate;

[0045] Calculate the shortest distance from the point cloud at the edge of the steel plate to the edge information of the bevel, and obtain the point cloud set at the intersection of the steel plate plane and the bevel surface based on the shortest distance and the projection width of the bevel surface.

[0046] In one possible implementation, the trajectory point generation module is further configured to:

[0047] The point cloud at the intersection of the steel plate plane and the bevel surface is obtained by fitting the point cloud at the intersection according to the edge type of the steel plate.

[0048] The point cloud at the junction is divided into initial grinding trajectory points, which include straight grinding trajectory points, circular grinding trajectory points and arc grinding trajectory points.

[0049] The main solution and its various further alternatives described above can be freely combined to form multiple solutions, all of which are solutions that can be adopted and are claimed in this application; furthermore, the (non-conflicting alternatives) can also be freely combined with each other and with other alternatives. Those skilled in the art, after understanding the solution of this application, will realize from the prior art and common general knowledge that there are many combinations, all of which are technical solutions to be protected by this application, and will not be exhaustively listed here.

[0050] This application discloses a trajectory planning method and apparatus for beveling steel plates. First, the steel plate point cloud data is identified and converted into a steel plate CAD number. Then, the steel plate point cloud data is segmented into planar point cloud data. Based on the bevel edge information in the steel plate CAD number, edge estimation is performed on the planar point cloud data to obtain the point cloud set at the intersection of the steel plate plane and the bevel surface. Initial grinding trajectory points are generated based on these points. Finally, the final grinding pose is obtained from the initial grinding pose generated from the initial grinding trajectory points and the correction transformation matrix. By calculating the point cloud at the intersection of the steel plate plane and the bevel surface, the accuracy of beveling grinding can be improved. Furthermore, combining the steel plate CAD and bevel information to plan the grinding trajectory enhances adaptability to complex bevel shapes. Automatic acquisition of the steel plate point cloud saves steel plate tooling positioning time, adapts to different steel plates, plans grinding paths, improves the automation level of the grinding workstation, and increases beveling grinding efficiency. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 A schematic flowchart of a trajectory planning method for steel plate beveling and grinding proposed in an embodiment of this application is shown.

[0053] Figure 2 An image corresponding to a steel plate CAD number proposed in an embodiment of this application is shown.

[0054] Figure 3 A schematic diagram of the grinding trajectory shown in the embodiment of this application is illustrated. Detailed Implementation

[0055] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0056] Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this application.

[0057] Beveling of steel plates involves using various beveling tools to process the surface of steel plates, achieving specific shapes and dimensions to meet diverse application requirements. One common type of beveling is grinding. Beveling improves the surface quality and lifespan of steel plates. Beveling results in a smoother surface with less roughness, enhanced wear and corrosion resistance, and increased durability during application. Furthermore, beveling can alter the shape of the steel plate to meet different needs. For example, for steel plates requiring cutting or bending, beveling can change their shape and size to better suit specific application scenarios.

[0058] Steel plate beveling and grinding trajectory planning, as one of the key technologies in intelligent grinding workstations, plays an indispensable role in modern welding engineering. Its importance is growing, primarily due to its crucial role as a core step in improving welding quality and the overall performance of the workpiece.

[0059] In existing technologies, steel plate beveling is typically performed manually or semi-automatically. Manual operation requires experienced technicians, while semi-automatic equipment lacks intelligence in trajectory planning and control, resulting in slow processing speed and low efficiency. Traditional grinding methods may also lead to insufficient processing accuracy due to operator inexperience, operational errors, and inaccurate tooling positioning, affecting the overall performance and reliability of the workpiece. Furthermore, traditional methods struggle to handle bevels with complex geometries, easily leading to processing dead corners and uneven surfaces, which negatively impact welding quality and the overall appearance of the workpiece.

[0060] Therefore, in order to solve the problems of low efficiency, insufficient precision and poor adaptability of existing steel plate beveling grinding methods, this application proposes a trajectory planning method and device for steel plate beveling grinding. It can overcome the shortcomings of traditional methods, improve production efficiency and processing precision, and adapt to the processing needs of different geometries and materials, so as to meet the needs of modern manufacturing industry for high quality and high efficiency. The following is a detailed description of it.

[0061] Please refer to Figure 1 , Figure 1 The diagram illustrates a trajectory planning method for steel plate beveling and grinding according to an embodiment of this application, including the following steps:

[0062] S1. Identify the point cloud data of the steel plate and convert it into a CAD number for the steel plate.

[0063] S2. Divide the steel plate point cloud data into steel plate planar point cloud data;

[0064] S3. Based on the bevel edge information in the steel plate CAD number, perform edge estimation on the steel plate plane point cloud data to obtain the point cloud set at the intersection of the steel plate plane and the bevel surface.

[0065] S4. Generate the initial grinding trajectory points based on the point cluster at the intersection of the steel plate plane and the bevel surface;

[0066] S5. Obtain the final grinding pose based on the initial grinding trajectory points generated and the correction transformation matrix.

[0067] First, the CAD number of the steel plate is obtained by identifying the point cloud of the steel plate. Second, the point cloud of the steel plate is divided into the steel plate plane and the bevel surface, and the point cloud set at the intersection of the steel plate plane and the bevel surface is calculated. Then, the initial grinding trajectory points and grinding direction are generated according to the bevel edge information. Finally, the final grinding pose is generated according to the actual grinding process.

[0068] Step S1, the step of identifying the steel plate point cloud data and converting the steel plate point cloud data into steel plate CAD numbers, includes:

[0069] Point cloud data of the steel plate is acquired using a 3D camera;

[0070] The point cloud data is converted to the robot's base coordinate system and then subjected to noise reduction processing.

[0071] Steel plate point cloud data is obtained by extracting steel plate data from point cloud data.

[0072] The point cloud data of the steel plate is converted into a two-dimensional steel plate image and template matching is performed to obtain the CAD number of the steel plate.

[0073] A 3D camera is used to photograph the steel plate to be polished to obtain the corresponding point cloud data. This point cloud data is then converted to the robot's base coordinate system. After noise reduction processing, the steel plate data is extracted from the point cloud to obtain the steel plate data point cloud, {P}={p i |p i =(x i ,y i ,z i ),i∈[1,n]}, where n is the number of points in the steel plate point cloud, which can be 262839.

[0074] Because the point cloud quality of the bevel portion in the steel plate point cloud is often poor or incomplete, it is not easy to directly plan the trajectory of the bevel surface. Therefore, it is necessary to identify the steel plate CAD number and obtain the bevel edge information. The above steel plate point cloud data is converted into a two-dimensional image, and template matching is performed to identify the steel plate CAD number m=4. Figure 2 An image corresponding to a steel plate CAD number proposed in an embodiment of this application is shown.

[0075] Step S2, the step of segmenting the steel plate point cloud data into steel plate planar point cloud data, includes:

[0076] Obtain the angle α between the normal to each point in the steel plate point cloud data and the z-axis of the base coordinate system. i ;

[0077] Angle α i The point cloud data of the steel plate is obtained by comparing it with the separation threshold.

[0078] Each point in the steel plate point cloud data is p i The set of normals corresponding to it is {N} = {n} i |n i =(x′) i ,y′ i ,z′ i ),i∈[1,n]}, obtain the angle α between the normal line corresponding to each point in the steel plate point cloud data and the z-axis of the base coordinate system. i A preset segmentation threshold θ th , make the included angle α i With the separation threshold θ th Comparing (25°), if the included angle α i Less than the separation threshold θ th Then the planar point cloud data of the steel plate is determined, if the included angle α i Greater than the separation threshold θ th Then the bevel surface data is determined. Therefore, point p i The point cloud representation is as follows:

[0079]

[0080] Among them, {P plane} represents the planar data of the steel plate, {P groove} represents the bevel surface data.

[0081] Step S3, the step of segmenting the steel plate point cloud data into steel plate planar point cloud data, includes:

[0082] Calculate the projected width of the bevel surface based on the bevel edge information in the steel plate CAD number;

[0083] Edge estimation is performed on the planar point cloud data of the steel plate to obtain the edge point cloud of the steel plate;

[0084] Calculate the shortest distance from the point cloud at the edge of the steel plate to the edge information of the bevel, and obtain the point cloud set at the intersection of the steel plate plane and the bevel surface based on the shortest distance and the projection width of the bevel surface.

[0085] For steel plates to be ground, their edges are divided into two categories: independent edges and edges that intersect with the bevel surface. The formula for calculating the projected width w of the bevel surface based on the bevel edge information in the steel plate's CAD number is: w = (hb) * tan(θ) groove ), where h is the thickness of the steel plate, b is the width of the blunt edge, and θgroove The bevel angle.

[0086] In one possible embodiment, h = 16 mm is taken as the steel plate thickness, b = 6 mm as the blunt edge width, and θ groove =45° is the bevel angle. The formula for calculating the projected width w of the bevel surface is: w = (hb) * tan(θ) groove =10mm.

[0087] Edge estimation algorithms are used to perform edge estimation on planar point cloud data of steel plates to obtain the edge point cloud {P} of the steel plate. edge}={p edgei |p edgei =(x edgei ,y edgei ,z edgei ),i∈[1,num]}, where num is the number of edge points of the steel plate, which can be 262839.

[0088] Calculate the point cloud p of the steel plate edge based on each bevel edge in the steel plate to be ground. edgei The shortest distance d of the bevel edge information i If the shortest distance d i If the width of the bevel surface is less than the projected width w, then the edge point p edgei For the intersection point corresponding to the bevel edge, all points that meet the condition are combined to form a point cloud corresponding to the bevel edge, resulting in a point cloud set divided according to the bevel edge of the steel plate. This point cloud set is the point cloud set {S} at the intersection of the steel plate plane and the bevel surface.

[0089] Step S4, which involves generating the initial grinding trajectory points based on the point cloud at the intersection of the steel plate plane and the bevel surface, includes:

[0090] The point cloud at the intersection of the steel plate plane and the bevel surface is obtained by fitting the point cloud at the intersection according to the edge type of the steel plate.

[0091] The point cloud at the junction is divided into initial polishing trajectory points, which include straight-line polishing trajectory points, circular polishing trajectory points, and arc polishing trajectory points.

[0092] Steel plate edge types include straight lines, circles, and arcs. Different strategies are needed to plan paths for different steel plate edge types. Therefore, it is necessary to fit the point cloud data at the intersection of the steel plate plane and the bevel surface according to the corresponding steel plate edge type to obtain the point cloud at the intersection.

[0093] The equation of the spatial line corresponding to the line is: Where (x0, y0, z0) is a point on the line, which can be set as (7.09, -189.42, 0.22), (m l ,n l ,pl The direction vector of the line can be set to (0.999, 0.022, -0.001). Based on the line equation and its corresponding point cloud, the two endpoints of the current line edge (the grinding trajectory points) can be obtained.

[0094] The spatial equation corresponding to the intersection of a circle and a plane is: Where (x) o ,y o ,z o Let (a, b, c) be the center of the circle, R be the radius, and (a, b, c) be the plane normal vector. A point set is obtained by taking points at certain intervals according to the spatial equation corresponding to the circle; this point set forms the grinding trajectory points of the circle.

[0095] The process for circular arcs involves adding a starting angle and a deflection angle to the basic circular arc. The processing is similar to that for circular arcs and will not be explained further here. For the grinding trajectory points on the circular arc, the selection of points is based on the circular arc, but with added angular range restrictions to obtain the corresponding point set.

[0096] The specified grinding direction is either counterclockwise or clockwise around the steel plate, and each trajectory point p(x,y,z) corresponds to a grinding direction. The trajectory point p(x,y,z) can be set to p(800.04,-193.88,200.46), and the grinding direction can be set. for

[0097] In step S5, the final grinding pose is obtained based on the initial grinding trajectory points generated by the initial grinding pose and the correction transformation matrix. Since the grinding head shape and size parameters may be different for different grinding processes, the grinding methods will be different, such as side grinding and end face grinding. Different grinding methods will also correspond to different grinding postures.

[0098] First, set the grinding direction to counterclockwise around the steel plate to be ground, with the initial grinding posture R as follows: The initial grinding pose T is formed based on the initial grinding posture and trajectory point p: When grinding the side, use the side of the grinding head to contact the bevel surface for grinding. and for: in Let z be the z-axis direction vector of the base coordinate system.

[0099] In the actual grinding process, the grinding trajectory is located at the center line of the bevel surface and adapts to the bevel angle θ. groove According to the bevel angle θ groove To correct the transformation matrix T m :

[0100]

[0101] When grinding the end face, the end face of the grinding head should be in contact with the bevel surface for grinding. and for: The corresponding actual grinding trajectory should be located at the centerline of the bevel surface and adapted to the bevel angle θ. groove According to the bevel angle θ groove The corrected transformation matrix T m :

[0102]

[0103] The final polishing pose T′ is: T′=T*T m The grinding trajectory is continuously generated for each bevel edge, thus completing the grinding trajectory planning and obtaining the grinding trajectory. Figure 3 A schematic diagram of the grinding trajectory shown in the embodiment of this application is illustrated.

[0104] In one possible embodiment, for end face grinding, the end face of the grinding head is used to contact the bevel surface for grinding. and for: Its transformation matrix T m for:

[0105]

[0106] Finally, the corresponding polishing pose T′ is obtained:

[0107]

[0108] Therefore, this application reduces the requirements for steel plate tooling positioning by acquiring steel plate point clouds through a 3D camera and converting them into robot space; by processing and analyzing the point clouds, the point clouds at the junction of the steel plate plane and the bevel surface are calculated, improving the accuracy of bevel grinding; by combining steel plate CAD and bevel information to plan the grinding trajectory, the adaptability to complex bevel shapes is improved; the automatic acquisition of steel plate point clouds saves steel plate tooling positioning time, can adapt to different steel plates to plan grinding paths, improves the automation level of the grinding workstation, and improves bevel grinding efficiency.

[0109] Compared with the prior art, the embodiments of this application have the following beneficial effects:

[0110] First, using a 3D camera to acquire point clouds of steel plates and converting them into robot space can reduce the requirements for positioning steel plate tooling.

[0111] Secondly, by processing and analyzing the point cloud to calculate the point cloud at the junction of the steel plate plane and the bevel surface, the accuracy of bevel grinding can be improved.

[0112] Third, combining steel plate CAD and bevel information to plan the grinding trajectory can improve adaptability to complex bevel shapes. Automatic acquisition of steel plate point clouds saves steel plate tooling positioning time, can adapt to different steel plates to plan grinding paths, improves the automation level of the grinding workstation, and improves bevel grinding efficiency.

[0113] The following is a possible implementation of a trajectory planning device for steel plate beveling, which is used to execute the various execution steps and corresponding technical effects of the trajectory planning method for steel plate beveling shown in the above embodiments and possible implementations.

[0114] The device includes:

[0115] The identification module is used to identify steel plate point cloud data and convert the steel plate point cloud data into steel plate CAD numbers;

[0116] The segmentation module is used to segment the steel plate point cloud data into steel plate planar point cloud data;

[0117] The edge estimation module is used to perform edge estimation on the point cloud data of the steel plate plane based on the bevel edge information in the steel plate CAD number to obtain the point cloud set at the intersection of the steel plate plane and the bevel surface.

[0118] The trajectory point generation module is used to generate initial grinding trajectory points based on the point cloud at the intersection of the steel plate plane and the bevel surface;

[0119] The grinding pose generation module is used to obtain the final grinding pose based on the initial grinding pose generated from the initial grinding trajectory points and the correction transformation matrix.

[0120] In one possible implementation, the identification module is further configured to:

[0121] Point cloud data of the steel plate is acquired using a 3D camera;

[0122] The point cloud data is converted to the robot's base coordinate system and then subjected to noise reduction processing.

[0123] Steel plate point cloud data is obtained by extracting steel plate data from point cloud data.

[0124] The point cloud data of the steel plate is converted into a two-dimensional steel plate image and template matching is performed to obtain the CAD number of the steel plate.

[0125] In one possible implementation, the segmentation module is further configured to:

[0126] Obtain the angle α between the normal to each point in the steel plate point cloud data and the z-axis of the base coordinate system. i ;

[0127] Angle α i The point cloud data of the steel plate is obtained by comparing it with the separation threshold.

[0128] In one possible implementation, the edge estimation module is further configured to:

[0129] Calculate the projected width of the bevel surface based on the bevel edge information in the steel plate CAD number;

[0130] Edge estimation is performed on the planar point cloud data of the steel plate to obtain the edge point cloud of the steel plate;

[0131] Calculate the shortest distance from the point cloud at the edge of the steel plate to the edge information of the bevel, and obtain the point cloud set at the intersection of the steel plate plane and the bevel surface based on the shortest distance and the projection width of the bevel surface.

[0132] In one possible implementation, the trajectory point generation module is further configured to:

[0133] The point cloud at the intersection of the steel plate plane and the bevel surface is obtained by fitting the point cloud at the intersection according to the edge type of the steel plate.

[0134] The point cloud at the junction is divided into initial polishing trajectory points, which include straight-line polishing trajectory points, circular polishing trajectory points, and arc polishing trajectory points.

[0135] In summary, the trajectory planning method and apparatus for steel plate beveling grinding disclosed in this application first identifies the steel plate point cloud data and converts it into a steel plate CAD number. Then, the steel plate point cloud data is segmented into planar point cloud data. Based on the bevel edge information in the steel plate CAD number, edge estimation is performed on the planar point cloud data to obtain the point cloud set at the intersection of the steel plate plane and the bevel surface. Initial grinding trajectory points are generated based on these points. Finally, the final grinding pose is obtained from the initial grinding pose generated from the initial grinding trajectory points and the correction transformation matrix. By calculating the point cloud at the intersection of the steel plate plane and the bevel surface, the accuracy of beveling grinding can be improved. Furthermore, combining the steel plate CAD and bevel information to plan the grinding trajectory enhances adaptability to complex bevel shapes. Automatic acquisition of the steel plate point cloud saves steel plate tooling positioning time, adapts to different steel plates, plans grinding paths, improves the automation level of the grinding workstation, and increases beveling grinding efficiency.

[0136] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A trajectory planning method for beveling steel plates, characterized in that, The method includes: Identify the point cloud data of the steel plate and convert the point cloud data of the steel plate into a CAD number of the steel plate; Obtain the angle between the normal to each point in the steel plate point cloud data and the z-axis of the base coordinate system. ; angle The steel plate planar point cloud data and bevel surface data are obtained by comparing with the separation threshold; Based on the bevel edge information in the steel plate CAD number, edge estimation is performed on the steel plate plane point cloud data to obtain the point cloud set at the intersection of the steel plate plane and the bevel surface. The initial grinding trajectory points are generated by aggregating the points at the intersection of the steel plate plane and the bevel surface. The final grinding pose is obtained by generating the initial grinding pose and the correction transformation matrix based on the initial grinding trajectory points. The grinding trajectory is then continuously generated by traversing each bevel edge to complete the grinding trajectory planning and obtain the grinding trajectory.

2. The trajectory planning method for steel plate beveling grinding as described in claim 1, characterized in that, The steps of identifying steel plate point cloud data and converting the steel plate point cloud data into steel plate CAD numbers include: Point cloud data of the steel plate is acquired using a 3D camera; The point cloud data is converted to the robot's base coordinate system and then subjected to noise reduction processing. Steel plate point cloud data is obtained by extracting steel plate data from point cloud data. The point cloud data of the steel plate is converted into a two-dimensional steel plate image and template matching is performed to obtain the CAD number of the steel plate.

3. The trajectory planning method for steel plate beveling grinding as described in claim 1, characterized in that, The steps for obtaining the point cloud set at the intersection of the steel plate plane and the bevel surface by edge estimation of the steel plate plane point cloud data based on the bevel edge information in the steel plate CAD number include: The bevel surface projection width is calculated based on the bevel edge information in the steel plate CAD number. Edge estimation is performed on the planar point cloud data of the steel plate to obtain the edge point cloud of the steel plate; Calculate the shortest distance from the point cloud at the edge of the steel plate to the edge information of the bevel, and obtain the point cloud set at the intersection of the steel plate plane and the bevel surface based on the shortest distance and the projection width of the bevel surface.

4. The trajectory planning method for steel plate beveling grinding as described in claim 1, characterized in that, The steps for generating initial grinding trajectory points based on the point cloud at the intersection of the steel plate plane and the bevel surface include: The point cloud at the intersection of the steel plate plane and the bevel surface is obtained by fitting the point cloud at the intersection according to the edge type of the steel plate. The point cloud at the junction is divided into initial grinding trajectory points, which include straight grinding trajectory points, circular grinding trajectory points and arc grinding trajectory points.

5. A trajectory planning device for beveling steel plates, characterized in that, The device includes: The identification module is used to identify the point cloud data of the steel plate and convert the point cloud data of the steel plate into the CAD number of the steel plate; The segmentation module is used to obtain the angle between the normal to each point in the steel plate point cloud data and the z-axis of the base coordinate system. ; angle The steel plate planar point cloud data and bevel surface data are obtained by comparing with the separation threshold; The edge estimation module is used to perform edge estimation on the point cloud data of the steel plate plane based on the bevel edge information in the steel plate CAD number to obtain the point cloud set at the intersection of the steel plate plane and the bevel surface. The trajectory point generation module is used to generate initial grinding trajectory points based on the point cloud at the intersection of the steel plate plane and the bevel surface; The grinding pose generation module is used to obtain the final grinding pose based on the initial grinding pose generated from the initial grinding trajectory points and the correction transformation matrix. It continuously traverses and generates the grinding trajectory for each bevel edge, completes the grinding trajectory planning, and obtains the grinding trajectory.

6. The trajectory planning device for steel plate beveling grinding as described in claim 5, characterized in that, The recognition module is also used for: Point cloud data of the steel plate is acquired using a 3D camera; The point cloud data is converted to the robot's base coordinate system and then subjected to noise reduction processing. Steel plate point cloud data is obtained by extracting steel plate data from point cloud data. The point cloud data of the steel plate is converted into a two-dimensional steel plate image and template matching is performed to obtain the CAD number of the steel plate.

7. The trajectory planning device for steel plate beveling grinding as described in claim 5, characterized in that, The edge estimation module is further used for: The bevel surface projection width is calculated based on the bevel edge information in the steel plate CAD number. Edge estimation is performed on the planar point cloud data of the steel plate to obtain the edge point cloud of the steel plate; Calculate the shortest distance from the point cloud at the edge of the steel plate to the edge information of the bevel, and obtain the point cloud set at the intersection of the steel plate plane and the bevel surface based on the shortest distance and the projection width of the bevel surface.

8. The trajectory planning device for steel plate beveling grinding as described in claim 5, characterized in that, The trajectory point generation module is also used for: The point cloud at the intersection of the steel plate plane and the bevel surface is obtained by fitting the point cloud at the intersection according to the edge type of the steel plate. The point cloud at the junction is divided into initial grinding trajectory points, which include straight grinding trajectory points, circular grinding trajectory points and arc grinding trajectory points.

Images