Fillet weld identification method and device

Abstract

The invention provides a fillet weld recognition method and device. The fillet weld recognition method comprises the steps that point cloud information of a to-be-welded structure and two positioning assemblies is collected; converting to obtain the direction of the fillet weld, the posture direction of the welding gun and the starting point initial coordinate and the ending point initial coordinate of the fillet weld; on the basis of the initial coordinates of the starting point and the initial coordinates of the end point, point cloud information of two local areas is intercepted from the point cloud information of the structure to be welded, point cloud data analysis is conducted on the point cloud information of the two local areas, and actual coordinates of the starting point and actual coordinates of the end point are obtained; based on the starting point actual coordinate, the end point actual coordinate, the direction of the fillet weld and the posture direction of the welding gun, recognition of the fillet weld is completed; according to the method and device, the initial coordinates of the starting point and the ending point of the fillet weld are obtained through the point cloud information of the positioning assembly, local point cloud information is intercepted according to the initial coordinates for fine analysis of the point cloud data, the data processing operand in the point cloud data analysis process is greatly reduced, and therefore the fillet weld recognition speed is increased, and the recognition precision is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of industrial robot vision control, and in particular relates to a fillet weld recognition method and device. Background Art

[0002] High-rise steel structures, transmission towers, and other steel structures often have numerous complex steel joints. These joints are welded together from numerous gusset plates, resulting in highly complex geometry. Because the welds at these joints are short and numerous, welders typically need to use handheld torches to complete the work.

[0003] With the development of information technology, industrial robotic arms are increasingly being used for welding in equipment such as automobiles and ships. These devices require accurate weld path information to function properly. For fillet welds in complex steel structure joints, this weld path information includes the 3D coordinates of the weld ends and the spatial posture of the welding gun during welding.

[0004] Currently, the main methods for determining the weld path for industrial robot arms are:

[0005] 1) Manual teaching. This method requires a worker to manually guide the industrial robot to the critical weld point. Compared to manual welding, this method offers no cost advantage. Even if the worker's welding skills are qualified, this method can be more expensive than manual welding.

[0006] 2) Utilize key points of the 3D geometric model of the welded workpiece to provide guidance information. This method requires retrieving key information from the 3D model to obtain key information about the weld. However, the main drawbacks of this method are: ① Many rough or complex workpieces often do not have corresponding 3D models; ② The geometric dimensions of the 3D model and the actual workpiece differ significantly, often deviating by around 5mm, resulting in the weld path generated by this method being too inaccurate to be used; ③ This method requires accurate spatial transformation to obtain the actual weld path based on the relationship between the node installation position and the virtual position used in the welding path simulation.

[0007] 3) Image information method. This method first collects point cloud or image information from the surface of the welded workpiece, then uses machine vision algorithms to analyze the geometric features of the point cloud to determine the weld path. Currently, machine vision positioning methods can be divided into two categories: 1) Using visible light to take photos and form a plan view to determine the weld location. This method has the advantage of fast calculation speed, but the main disadvantage is large Z-axis ranging error, and the produced welding path often cannot meet the requirements; 2) Collecting three-dimensional laser point cloud information from the surface of the welded workpiece, and then using artificial intelligence and other methods to identify the weld. Because the accuracy of the welding path cannot be less than 1mm, and the length of the weld is generally greater than 200mm, the point cloud resolution is required to be at least within 0.1mm. This results in too many data points in the point cloud and too long calculation time, often exceeding 30 minutes, making it unsuitable for actual engineering needs. Therefore, there is an urgent need for a method and device that is simple to operate and has high weld identification efficiency and accuracy. Summary of the Invention

[0008] In order to overcome the above-mentioned deficiencies of the prior art, the present invention proposes a fillet weld identification method, comprising:

[0009] Based on the point cloud acquisition device, point cloud information of the structure to be welded and point cloud information of two positioning components respectively arranged on two fillet weld surfaces of the structure to be welded are acquired;

[0010] Based on the point cloud information of the two positioning components, the direction of the fillet weld, the posture direction of the welding gun, the initial coordinates of the starting point of the fillet weld, and the initial coordinates of the end point are converted;

[0011] Taking the initial coordinates of the starting point and the initial coordinates of the end point as centers, respectively, determining two local areas based on a preset range, intercepting point cloud information of the two local areas from the point cloud information of the structure to be welded, performing point cloud data analysis on the point cloud information of the two local areas, and obtaining actual coordinates of the starting point and the end point;

[0012] The fillet weld is identified based on the actual coordinates of the starting point, the actual coordinates of the end point, the direction of the fillet weld, and the posture direction of the welding gun.

[0013] Preferably, each of the positioning components includes a fillet weld starting point marker, a fillet weld end point marker and an auxiliary point marker that are not arranged in a colinear manner.

[0014] Preferably, a line connecting two of the fillet weld starting point marking members and the starting point of the fillet weld is perpendicular to the fillet weld, and a line connecting two of the fillet weld ending point marking members and the ending point of the fillet weld is perpendicular to the fillet weld.

[0015] Preferably, the direction of the fillet weld, the posture direction of the welding gun, the initial coordinates of the starting point and the initial coordinates of the end point of the fillet weld are converted based on the point cloud information of the two positioning components, including:

[0016] Based on the point cloud information of the two positioning components, the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning components are analyzed and obtained;

[0017] Based on the spatial coordinates of the fillet weld starting point marker and the fillet weld end point marker in any one of the positioning assemblies, a unit vector from the fillet weld starting point marker to the fillet weld end point marker is calculated as a fillet weld direction;

[0018] Based on the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in each positioning assembly, a spatial analytical geometry algorithm is used to calculate the plane equation of the fillet weld welding surface where each positioning assembly is located;

[0019] Based on the plane equations of the two fillet weld surfaces, a spatial analytical geometry algorithm is used to calculate the negative vector of the normal angle bisector of the two fillet weld surfaces as the posture direction of the welding gun;

[0020] Based on the spatial coordinates of the two fillet weld starting point markers and the plane equations of the two fillet weld welding surfaces, a spatial analytic geometry algorithm is used to calculate the spatial coordinates of the projection point of the midpoint of the line connecting the two fillet weld starting point markers to the intersection line of the two fillet weld welding surfaces, and use them as the initial coordinates of the starting point of the fillet weld;

[0021] Based on the spatial coordinates of the two fillet weld end point markers and the plane equations of the two fillet weld welding surfaces, a spatial analytical geometry algorithm is used to calculate the spatial coordinates of the projection point from the midpoint of the line connecting the two fillet weld end point markers to the intersection of the two fillet weld welding surfaces as the initial coordinates of the end point of the fillet weld.

[0022] Preferably, the analyzing and obtaining the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning components based on the point cloud information of the two positioning components includes:

[0023] Based on the attribute characteristics of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components, the point cloud information of the two positioning components is distinguished to obtain the point cloud information of each of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components; wherein the attribute characteristics of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components are different;

[0024] Based on the point cloud information of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning components, the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning components are analyzed and obtained.

[0025] Preferably, the attribute characteristics include one or more of the following:

[0026] Size characteristics, color characteristics.

[0027] Based on the same inventive concept, the present invention also provides a fillet weld identification device, comprising:

[0028] Two positioning components, point cloud acquisition equipment and data processing modules;

[0029] The positioning assembly adopts the positioning assembly described above;

[0030] The point cloud acquisition device is used to acquire point cloud information of the structure to be welded and point cloud information of two positioning components respectively arranged on two welding surfaces of the fillet weld of the structure to be welded;

[0031] The data processing module is electrically connected to the point cloud acquisition device, and is used to convert the direction of the fillet weld, the posture direction of the welding gun, the initial coordinates of the starting point and the initial coordinates of the end point of the fillet weld based on the point cloud information of the two positioning components; based on the initial coordinates of the starting point and the initial coordinates of the end point as the center, based on a preset range, intercept the information of two analysis areas in the point cloud information of the structure to be welded, perform point cloud data analysis on the point cloud information of the two local areas, and obtain the actual coordinates of the starting point and the actual coordinates of the end point; based on the actual coordinates of the starting point, the actual coordinates of the end point, the direction of the fillet weld and the posture direction of the welding gun, complete the identification of the fillet weld.

[0032] Preferably, the data processing module includes:

[0033] A first processing submodule is configured to analyze and obtain the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning assemblies based on the point cloud information of the two positioning assemblies;

[0034] a second processing submodule, configured to calculate, based on the spatial coordinates of the fillet weld starting point marker and the fillet weld ending point marker in any one of the positioning assemblies, a unit vector from the fillet weld starting point marker to the fillet weld ending point marker as a fillet weld direction;

[0035] a third processing submodule, configured to calculate, based on the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in each positioning assembly, a plane equation of the fillet weld welding surface where each positioning assembly is located using a spatial analytical geometry algorithm;

[0036] a fourth processing submodule, configured to calculate, based on the plane equations of the two fillet weld surfaces, a negative vector of the normal angle bisector of the two fillet weld surfaces using a spatial analytical geometry algorithm, as a posture direction of the welding gun;

[0037] a fifth processing submodule, configured to calculate, based on the spatial coordinates of the two fillet weld starting point markers and the plane equations of the two fillet weld welding surfaces, using a spatial analytic geometry algorithm to obtain the spatial coordinates of a projection point from a midpoint of a line connecting the two fillet weld starting point markers to an intersection line of the two fillet weld welding surfaces, as the initial coordinates of the starting point of the fillet weld;

[0038] The sixth processing submodule is used to calculate the spatial coordinates of the projection point from the midpoint of the line connecting the two fillet weld end point markers to the intersection line of the two fillet weld welding surfaces based on the spatial coordinates of the two fillet weld end point markers and the plane equations of the two fillet weld welding surfaces, using a spatial analytical geometry algorithm, as the initial coordinates of the end point of the fillet weld.

[0039] Preferably, the first processing submodule is specifically configured to:

[0040] Based on the attribute characteristics of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components, the point cloud information of the two positioning components is distinguished to obtain the point cloud information of each of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components; wherein the attribute characteristics of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components are different;

[0041] Based on the point cloud information of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning components, the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning components are analyzed and obtained.

[0042] Preferably, the attribute characteristics include one or more of the following:

[0043] Size characteristics, color characteristics.

[0044] Preferably, the fillet weld starting point marker, fillet weld end point marker and auxiliary point marker of the positioning assembly all include:

[0045] Hemispherical magnet.

[0046] Preferably, the hemispherical magnet is a hemispherical electromagnet.

[0047] Preferably, the diameter of the hemispherical magnet ranges from 10 mm to 20 mm.

[0048] Based on the same inventive concept, the present invention also provides a computer device, comprising: one or more processors;

[0049] a memory for storing one or more programs;

[0050] When the one or more programs are executed by the one or more processors, a fillet weld identification method as described above is implemented.

[0051] Based on the same inventive concept, the present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed, the fillet weld identification method as described above is implemented.

[0052] Compared with the closest prior art, the present invention has the following beneficial effects:

[0053] The present invention provides a fillet weld identification method and device, comprising: collecting point cloud information of a structure to be welded and point cloud information of two positioning components respectively arranged on two welding surfaces of the fillet welds of the structure to be welded based on a point cloud acquisition device; converting the direction of the fillet weld, the posture direction of the welding gun, the initial coordinates of the starting point and the initial coordinates of the end point of the fillet weld based on the point cloud information of the two positioning components; determining two local areas based on a preset range with the initial coordinates of the starting point and the initial coordinates of the end point as the center respectively, intercepting the point cloud information of the two local areas from the point cloud information of the structure to be welded, and performing a relative calculation on the two local areas. The point cloud information is used to perform point cloud data analysis to obtain the actual coordinates of the starting point and the actual coordinates of the end point; the fillet weld is identified based on the actual coordinates of the starting point, the actual coordinates of the end point, the direction of the fillet weld, and the posture direction of the welding gun; the method and device obtain the initial coordinates of the starting point and the end point of the fillet weld through the point cloud information of the positioning component, and intercept local point cloud information from the complete point cloud information of the structure to be welded according to the initial coordinates to perform a detailed analysis of the point cloud data, which greatly reduces the amount of data processing calculations in the point cloud data analysis process, can speed up the recognition speed of the fillet weld, and improve the recognition accuracy, and is suitable for actual engineering needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A schematic flow chart of a fillet weld identification method provided by the present invention;

[0055] Figure 2 A schematic diagram of the positional relationship between the structure to be welded and the positioning assembly provided by the present invention;

[0056] Figure 3 A schematic diagram of the spatial posture of the industrial robot welding gun provided by the present invention;

[0057] Figure 4 A schematic structural diagram of a fillet weld identification device provided by the present invention;

[0058] Figure 5 A schematic diagram of the structure of a hemispherical electromagnet provided by the present invention;

[0059] Figure 6 This is a structural schematic diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0060] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0061] Example 1:

[0062] The present invention provides a fillet weld identification method, such as Figure 1 and Figure 2 Shown, including:

[0063] S1. Using a point cloud acquisition device, collect point cloud information of a structure to be welded and point cloud information of two positioning components respectively provided on two fillet weld surfaces of the structure to be welded;

[0064] S2. Based on the point cloud information of the two positioning components, calculate the direction of the fillet weld, the posture direction of the welding gun, the initial coordinates of the starting point of the fillet weld, and the initial coordinates of the end point of the fillet weld;

[0065] S3, respectively taking the initial coordinates of the starting point and the initial coordinates of the end point as centers, determining two local areas based on a preset range, intercepting point cloud information of the two local areas from the point cloud information of the structure to be welded, and performing point cloud data analysis on the point cloud information of the two local areas to obtain actual coordinates of the starting point and the end point;

[0066] S4. Complete identification of the fillet weld based on the actual coordinates of the starting point, the actual coordinates of the end point, the direction of the fillet weld, and the posture direction of the welding gun.

[0067] For complex, crude steel structures such as bridges and high-rise steel structures, these workpieces have complex geometric shapes, low machining precision, and often lack a refined three-dimensional model suitable for robotic welding. This makes it difficult to cost-effectively determine the precise welding path, which poses significant challenges for robotic automated welding. The present invention obtains the initial coordinates of the fillet weld's starting and ending points by locating the component's point cloud information. Based on these initial coordinates, local point cloud information is extracted from the complete point cloud information of the structure to be welded for detailed analysis of the point cloud data. This significantly reduces the amount of data processing required during point cloud data analysis, speeding up fillet weld recognition while also improving accuracy, making it suitable for practical engineering needs.

[0068] In this embodiment, the point cloud acquisition device includes a surface laser sensor.

[0069] In this embodiment, each of the positioning components includes a fillet weld starting point marker, a fillet weld end point marker, and an auxiliary point marker that are not arranged in a collinear manner.

[0070] It should be noted that the purpose of non-collinear layout is to enable the three markers, namely, the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker, to determine a fillet weld welding surface, that is, the plane equation of the fillet weld welding surface where they are located can be determined based on the coordinate positions of the three markers, namely, the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker.

[0071] It should be noted that, considering the convenience of disassembly of the positioning component and the recognition accuracy of the positioning component by the surface laser sensor, in this embodiment, the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker are all hemispherical magnets. The flat part of the hemispherical magnet can fit smoothly with the welding surface of the fillet weld, and the hemispherical structure has a regular shape, which can enable the surface laser sensor to better identify the center of the hemispherical magnet, thereby improving the recognition accuracy.

[0072] In another possible implementation, the fillet weld starting point marker, fillet weld end point marker and auxiliary point marker may also be magnets with other regular shapes and clear centers, such as cube magnets or regular triangular prism magnets.

[0073] It should be noted that the point cloud information of the two positioning components collected by the point cloud acquisition device is the point cloud information of the six markers in space. In order to accurately distinguish each marker during subsequent data analysis, in this embodiment, the attribute characteristics of the fillet weld start marker, the fillet weld end marker, and the auxiliary point marker in the two positioning components are different.

[0074] In this embodiment, the attribute characteristics include one or more of the following:

[0075] Size characteristics, color characteristics.

[0076] It should be noted that in order to better distinguish the point cloud information of each marker, the size feature difference of the marker in each positioning component is greater than the set size threshold, and the color grayscale difference of the two positioning components is greater than the set grayscale threshold.

[0077] Specifically, the two positioning components of the present invention involve a total of 6 hemispherical magnets. These 6 hemispherical magnets are divided into two groups according to color, with 3 in each group, each in black and white. The diameters of the 3 hemispherical magnets in each color group, that is, in each positioning component, are different. Generally speaking, the larger the diameter of the hemispherical magnet, the more accurate the point cloud recognition, but the heavier it is, and the larger the area of the steel plate it needs to occupy. The diameter of the hemispherical magnet is generally around 15 mm, and the diameters of the three hemispherical magnets in a positioning component differ by 5 mm and are arranged in an arithmetic progression. Permanent magnets can be used as hemispherical magnets. However, after the permanent magnet is positioned, workers need a lot of force to break it off the steel plate; at the same time, when storing, the permanent magnets will stick together due to mutual attraction, making storage difficult. Therefore, in this embodiment, electromagnets are preferably used as hemispherical magnets.

[0078] like Figure 2As shown, six hemispherical magnets are placed on the steel plates on both sides of the fillet weld, i.e., on the two fillet weld surfaces C. The three black magnets are placed on one side and the white magnets on the other. The center of the black hemispherical magnets is denoted as B, and the center of the white hemispherical magnets is denoted as W. On the same side, the hemispherical magnet with the smallest diameter (centers B3 and W3, respectively) serves as the fillet weld starting point marker and is located at the weld starting point. The hemispherical magnet with an intermediate diameter (centers B2 and W2, respectively) serves as the fillet weld end point marker and is located at the weld end point. The hemispherical magnet with the largest diameter (centers B1 and W1, respectively) serves as an auxiliary point marker and is located on the steel plate away from the weld line. A surface laser sensor is used to collect laser point clouds of the two fillet weld surfaces and point clouds of the surfaces of the six hemispherical magnets.

[0079] During use, the six hemispherical magnets are placed on the steel plates on either side of the weld, according to the positioning relationship of the aforementioned positioning components. The surface laser sensor then scans the point cloud at these locations. Once the laser point cloud of the weld surface (i.e., the weld surface C of the two fillet welds) is captured, the center of each of the six hemispherical magnets is identified. Subsequent data processing then identifies the weld path, the weld starting point, and the welding gun's posture. This eliminates the need to analyze the laser point cloud of the entire weld, significantly improving the program's operational efficiency and stability.

[0080] In this embodiment, in order to improve the positioning accuracy of the initial coordinates of the starting point and the end point of the fillet weld, during the actual installation of the positioning component, the line connecting the two fillet weld starting point marking members and the starting point of the fillet weld is perpendicular to the fillet weld, and the line connecting the two fillet weld end point marking members and the end point of the fillet weld is perpendicular to the fillet weld.

[0081] Specifically, such as Figure 2 As shown, the line connecting the center B3 of the fillet weld starting point marker and the starting point of the fillet weld, and the line connecting W3 and the starting point of the fillet weld are perpendicular to the fillet weld, and the line connecting the center B2 of the fillet weld end point marker and the end point of the fillet weld, and the line connecting W2 and the end point of the fillet weld are perpendicular to the fillet weld.

[0082] Using a hemispherical vision-guided device, namely two positioning components and a point cloud acquisition device, the centers of six hemispherical magnets are obtained. This patent proposes a method using these six centers, using steps S1-S4, to determine the starting point of the weld and the posture of the welding gun.

[0083] In this embodiment, the above S1 may include:

[0084] S101, analyzing and obtaining the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning assemblies based on point cloud information of the two positioning assemblies;

[0085] In this embodiment, the above S101 may include:

[0086] Based on the attribute characteristics of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components, the point cloud information of the two positioning components is distinguished to obtain the point cloud information of each of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components;

[0087] Based on the point cloud information of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning components, the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning components are analyzed and obtained.

[0088] Wherein, the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning assemblies are the spatial coordinates of the sphere centers of the respective markers.

[0089] This invention utilizes point cloud recognition software such as PCL, building on existing point cloud recognition technology, to efficiently and accurately identify the centers of the six hemispherical magnets—that is, the spatial coordinates of the fillet weld start and end markers, as well as the auxiliary point markers in the two positioning assemblies. Formulas can then be used to calculate the weld start and end coordinates, as well as the proper welding gun posture. This data can be used to guide a robot to complete welding operations. This system offers advantages such as simple operation and a high recognition rate.

[0090] S102, based on the spatial coordinates of the fillet weld starting point marker and the fillet weld end point marker in any one of the positioning assemblies, calculating a unit vector from the fillet weld starting point marker to the fillet weld end point marker as a fillet weld direction;

[0091] Taking the positioning components as an example, all of them are black hemispherical magnets. The direction of the fillet weld Dirx is the direction from the middle ball, which is the fillet weld end point marker (the center of the ball is B2), to the small ball, which is the fillet weld start point marker (the center of the ball is B3), which can be expressed as:

[0092] Dirx=(B3-B2) / |B3-B2|;

[0093] S103, based on the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in each positioning assembly, using a spatial analytical geometry algorithm to calculate the plane equation of the fillet weld welding surface where each positioning assembly is located;

[0094] According to the formula, calculate the plane equation of the two sides of the steel plate, that is, the two fillet weld welding surface C. At this time, the steel plate on the left side of the weld, that is, Figure 2 The plane equation of the vertical fillet weld surface is expressed as:

[0095] a1x+b1y+c1z+d1=0;

[0096] a1=(B1 y -B2 y )(B3 z -B1 z )-(B2 z -B1 z )(B3 y -B1 y );

[0097] b1=(B3 x -B1 x )(B1 z -B1 x )-(B2 x -B1 x )(B3 x -B1 x );

[0098] c1=(B2 x -B2 x )(B3 y -B1 y )-(B3 x -B1 x )(B2 y -B1 y );

[0099] d1=-(a1B1 x +b1B1 y +c1B1 z );

[0100] Among them, a1, b1, c1 are the components of the plane normal vector of the steel plate on the left side of the weld in the x-axis, y-axis and z-axis directions respectively, d1 is a constant term; B1 x 、B1 y 、B1 z are the coordinates of the sphere center B1 in the x-axis, y-axis and z-axis directions, B2 x 、B2 y 、B2 z are the coordinates of the center B2 in the x-axis, y-axis and z-axis directions, B3 x 、B3 y 、B3 z are the coordinates of the sphere center B3 in the x-axis, y-axis and z-axis directions respectively;

[0101] The plane on the right is Figure 2 The calculation method of the plane equation of the horizontal fillet weld surface is similar to that of

[0102] a2x+b2y+c2z+d2=0;

[0103] Where a2, b2, and c2 are the components of the plane normal vector of the steel plate on the right side of the weld in the x-axis, y-axis, and z-axis directions, respectively, and d2 is a constant term;

[0104] S104. Based on the plane equations of the two fillet weld surfaces, a spatial analytical geometry algorithm is used to calculate a negative vector of the normal angle bisector of the two fillet weld surfaces as the posture direction of the welding gun;

[0105] Z-axis Dirz in welding gun TCP, Figure 3 The direction of the blue line at the tip of the middle welding wire is the negative direction of the bisector of the normal angles of the two fillet weld surfaces. The calculation process is expressed as:

[0106] u1=(a1,b1,c1);

[0107] u2=(a2,b2,c2);

[0108] Dirz=-(u1+u2) / |u1+u2|;

[0109] Among them, u1 and u2 are the normal vectors of the welding surfaces of the two fillet welds respectively;

[0110] In most cases, Figure 3 The direction of the middle red line can be opposite to Dirx.

[0111] S105. Based on the spatial coordinates of the two fillet weld starting point markers and the plane equations of the two fillet weld welding surfaces, a spatial analytic geometry algorithm is used to calculate the spatial coordinates of the projection point of the line connecting the midpoint of the two fillet weld starting point markers to the intersection line of the two fillet weld welding surfaces, and use these as the initial coordinates of the starting point of the fillet weld.

[0112] The rough position of the weld starting point, i.e., the initial coordinates of the starting point, is the coordinates of the projection point from the average value of the sphere centers of the two fillet weld starting point markers to the intersection line of the two fillet weld welding surfaces. The process of determining the initial coordinates of the end point is similar.

[0113] S106. Based on the spatial coordinates of the two fillet weld end point markers and the plane equations of the two fillet weld welding surfaces, a spatial analytic geometry algorithm is used to calculate the spatial coordinates of the projection point from the midpoint of the line connecting the two fillet weld end point markers to the intersection line of the two fillet weld welding surfaces, as the initial coordinates of the end point of the fillet weld.

[0114] In the above S3, two local areas are determined based on a preset range with the initial coordinates of the starting point and the initial coordinates of the end point as the center, the point cloud information of the two local areas is intercepted from the point cloud information of the structure to be welded, and point cloud data analysis is performed on the point cloud information of the two local areas to obtain the actual coordinates of the starting point and the actual coordinates of the end point;

[0115] Specifically, since the installation of the positioning assembly inevitably involves manual errors, once the rough locations of the weld's starting point (i.e., the initial coordinates of the start and end points) are determined, point cloud analysis software such as PCL can be used to capture a localized point cloud approximately 10 cm to the left and right of the rough starting point for precise analysis of the starting point coordinates. The algorithm for determining the end point position is essentially the same as that for the starting point. At this point, the weld's starting and end point coordinates, as well as the welding gun's Z-axis orientation (Dirz) and X-axis orientation (-Dirx) are calculated, thus completing the weld determination.

[0116] Example 2:

[0117] Based on the same inventive concept, the present invention also provides a fillet weld identification device, such as Figure 4 Shown, including:

[0118] Two positioning components, point cloud acquisition equipment and data processing modules;

[0119] The positioning assembly adopts the positioning assembly described above;

[0120] The point cloud acquisition device is used to acquire point cloud information of the structure to be welded and point cloud information of two positioning components respectively arranged on two welding surfaces of the fillet weld of the structure to be welded;

[0121] The data processing module is electrically connected to the point cloud acquisition device, and is used to convert the direction of the fillet weld, the posture direction of the welding gun, the initial coordinates of the starting point and the initial coordinates of the end point of the fillet weld based on the point cloud information of the two positioning components; based on the initial coordinates of the starting point and the initial coordinates of the end point as the center, based on a preset range, intercept the information of two analysis areas in the point cloud information of the structure to be welded, perform point cloud data analysis on the point cloud information of the two local areas, and obtain the actual coordinates of the starting point and the actual coordinates of the end point; based on the actual coordinates of the starting point, the actual coordinates of the end point, the direction of the fillet weld and the posture direction of the welding gun, complete the identification of the fillet weld.

[0122] In this embodiment, six hemispherical magnets are placed on the steel plates on both sides of the fillet weld, i.e., on the two fillet weld welding surfaces C. The three black magnets are placed on one side and the three white magnets are placed on the other side. The center of the black hemispherical magnet is denoted as B, and the center of the white hemispherical magnet is denoted as W. On the same side, the hemispherical magnet with the smallest diameter (with centers B3 and W3, respectively) is located at the starting point of the weld as the fillet weld starting point marker. The hemispherical magnet with an intermediate diameter (with centers B2 and W2, respectively) is located at the end point of the weld as the fillet weld end point marker. The hemispherical magnet with the largest diameter (with centers B1 and W1, respectively) is located at the position of the steel plate away from the weld line as an auxiliary point marker.

[0123] In this embodiment, the data processing module includes:

[0124] A first processing submodule is configured to analyze and obtain the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning assemblies based on the point cloud information of the two positioning assemblies;

[0125] a second processing submodule, configured to calculate, based on the spatial coordinates of the fillet weld starting point marker and the fillet weld ending point marker in any one of the positioning assemblies, a unit vector from the fillet weld starting point marker to the fillet weld ending point marker as a fillet weld direction;

[0126] a third processing submodule, configured to calculate, based on the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in each positioning assembly, a plane equation of the fillet weld welding surface where each positioning assembly is located using a spatial analytical geometry algorithm;

[0127] a fourth processing submodule, configured to calculate, based on the plane equations of the two fillet weld surfaces, a negative vector of the normal angle bisector of the two fillet weld surfaces using a spatial analytical geometry algorithm, as a posture direction of the welding gun;

[0128] a fifth processing submodule, configured to calculate, based on the spatial coordinates of the two fillet weld starting point markers and the plane equations of the two fillet weld welding surfaces, using a spatial analytic geometry algorithm to obtain the spatial coordinates of a projection point from a midpoint of a line connecting the two fillet weld starting point markers to an intersection line of the two fillet weld welding surfaces, as the initial coordinates of the starting point of the fillet weld;

[0129] The sixth processing submodule is used to calculate the spatial coordinates of the projection point from the midpoint of the line connecting the two fillet weld end point markers to the intersection line of the two fillet weld welding surfaces based on the spatial coordinates of the two fillet weld end point markers and the plane equations of the two fillet weld welding surfaces, using a spatial analytical geometry algorithm, as the initial coordinates of the end point of the fillet weld.

[0130] In this embodiment, the first processing submodule is specifically configured to:

[0131] Based on the attribute characteristics of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components, the point cloud information of the two positioning components is distinguished to obtain the point cloud information of each of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components; wherein the attribute characteristics of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components are different;

[0132] Based on the point cloud information of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning components, the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning components are analyzed and obtained.

[0133] In this embodiment, the attribute characteristics include one or more of the following:

[0134] Size characteristics, color characteristics.

[0135] In this embodiment, the fillet weld starting point marker, fillet weld end point marker, and auxiliary point marker of the positioning assembly all include:

[0136] Hemispherical magnet.

[0137] In this embodiment, Figure 5 As shown, the hemispherical magnet is a hemispherical electromagnet 1 and a switch 2 arranged on the outer wall of the hemispherical electromagnet 1. The switch 2 is connected in series to the power supply circuit of the hemispherical electromagnet 1. The iron core, coil, and power line of the hemispherical electromagnet 1 are not shown in the figure. The existing electromagnet structure is adopted. The hemispherical electromagnet 1 has a certain cavity inside to ensure heat dissipation when the electromagnet is working.

[0138] In this embodiment, the diameter of the hemispherical magnet ranges from 10 mm to 20 mm.

[0139] Example 3

[0140] like Figure 6 As shown, the present invention also provides an electronic device, which may be a computer, a single-chip microcomputer, a smart mobile device, or the like. The electronic device in this embodiment may include a processor, a memory, a transceiver component, and the like. The memory, processor, and transceiver component are connected via a bus; the memory may be used to store an execution program, which may include instructions; and the processor may be used to execute the instructions stored in the memory. The memory may also be used to store data, which may be accessed and / or modified during the execution of the instructions.

[0141] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in a storage medium to implement the corresponding method flow or corresponding function, so as to implement the steps of a fillet weld identification method in the above embodiment.

[0142] Example 4

[0143] Based on the same inventive concept, the present invention also provides a readable storage medium, specifically an electronic device readable storage medium (Memory), which is a memory device in the electronic device for storing programs and data. It can be understood that the storage medium here can include both the built-in storage medium in the electronic device and, of course, the extended storage medium supported by the electronic device. The storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space. These instructions can be one or more execution programs (including program codes). It should be noted that the storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor loads and executes one or more instructions stored in the storage medium, which can implement the steps of a fillet weld identification method in the above embodiment.

[0144] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0145] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0146] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0147] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0148] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit its scope of protection. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that after reading the present invention, those skilled in the art may still make various changes, modifications or equivalent substitutions to the specific implementation methods of the application, but these changes, modifications or equivalent substitutions are all within the scope of protection of the claims of the present invention.

Claims

1. A fillet weld identification method, characterized in that: include: Based on the point cloud acquisition device, point cloud information of the structure to be welded and point cloud information of two positioning components respectively arranged on two fillet weld surfaces of the structure to be welded are acquired; Based on the point cloud information of the two positioning components, the direction of the fillet weld, the posture direction of the welding gun, the initial coordinates of the starting point of the fillet weld, and the initial coordinates of the end point are converted; Taking the initial coordinates of the starting point and the initial coordinates of the end point as centers, respectively, determining two local areas based on a preset range, intercepting point cloud information of the two local areas from the point cloud information of the structure to be welded, performing point cloud data analysis on the point cloud information of the two local areas, and obtaining actual coordinates of the starting point and the end point; The fillet weld is identified based on the actual coordinates of the starting point, the actual coordinates of the end point, the direction of the fillet weld, and the posture direction of the welding gun.

2. A fillet weld identification method according to claim 1, characterized in that: Each positioning assembly includes a fillet weld starting point marker, a fillet weld end point marker, and an auxiliary point marker that are not arranged in a collinear manner.

3. A fillet weld identification method according to claim 2, characterized in that: A line connecting the two fillet weld starting point marking members and the starting point of the fillet weld is perpendicular to the fillet weld, and a line connecting the two fillet weld ending point marking members and the ending point of the fillet weld is perpendicular to the fillet weld.

4. A fillet weld identification method according to claim 2 or 3, characterized in that: The point cloud information based on the two positioning components is converted to obtain the direction of the fillet weld, the posture direction of the welding gun, the initial coordinates of the starting point of the fillet weld, and the initial coordinates of the end point, including: Based on the point cloud information of the two positioning components, the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning components are analyzed and obtained; Based on the spatial coordinates of the fillet weld starting point marker and the fillet weld end point marker in any one of the positioning assemblies, a unit vector from the fillet weld starting point marker to the fillet weld end point marker is calculated as a fillet weld direction; Based on the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in each positioning assembly, a spatial analytical geometry algorithm is used to calculate the plane equation of the fillet weld welding surface where each positioning assembly is located; Based on the plane equations of the two fillet weld surfaces, a spatial analytical geometry algorithm is used to calculate the negative vector of the normal angle bisector of the two fillet weld surfaces as the posture direction of the welding gun; Based on the spatial coordinates of the two fillet weld starting point markers and the plane equations of the two fillet weld welding surfaces, a spatial analytic geometry algorithm is used to calculate the spatial coordinates of the projection point of the midpoint of the line connecting the two fillet weld starting point markers to the intersection line of the two fillet weld welding surfaces, and use them as the initial coordinates of the starting point of the fillet weld; Based on the spatial coordinates of the two fillet weld end point markers and the plane equations of the two fillet weld welding surfaces, a spatial analytical geometry algorithm is used to calculate the spatial coordinates of the projection point from the midpoint of the line connecting the two fillet weld end point markers to the intersection of the two fillet weld welding surfaces as the initial coordinates of the end point of the fillet weld.

5. A fillet weld identification method according to claim 4, characterized in that: The step of analyzing the point cloud information of the two positioning components to obtain the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components includes: Based on the attribute characteristics of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components, the point cloud information of the two positioning components is distinguished to obtain the point cloud information of each of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components; wherein the attribute characteristics of the fillet weld starting point marker, the fillet weld end point marker, and the auxiliary point marker in the two positioning components are different; Based on the point cloud information of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning components, the spatial coordinates of the fillet weld starting point marker, the fillet weld end point marker and the auxiliary point marker in the two positioning components are analyzed and obtained.

6. A fillet weld identification method according to claim 5, characterized in that: The attribute characteristics include one or more of the following: Size characteristics, color characteristics.

7. A fillet weld identification device, characterized in that: include: Two positioning components, point cloud acquisition equipment and data processing modules; The positioning assembly adopts the positioning assembly according to any one of claims 1 to 6; The point cloud acquisition device is used to acquire point cloud information of the structure to be welded and point cloud information of two positioning components respectively arranged on two welding surfaces of the fillet weld of the structure to be welded; The data processing module is electrically connected to the point cloud acquisition device, and is used to convert the direction of the fillet weld, the posture direction of the welding gun, the initial coordinates of the starting point and the initial coordinates of the end point of the fillet weld based on the point cloud information of the two positioning components; based on the initial coordinates of the starting point and the initial coordinates of the end point as the center, based on a preset range, intercept the information of two analysis areas in the point cloud information of the structure to be welded, perform point cloud data analysis on the point cloud information of the two local areas, and obtain the actual coordinates of the starting point and the actual coordinates of the end point; based on the actual coordinates of the starting point, the actual coordinates of the end point, the direction of the fillet weld and the posture direction of the welding gun, complete the identification of the fillet weld.

8. A fillet weld identification device according to claim 6, characterized in that: The fillet weld starting point marker, fillet weld end point marker and auxiliary point marker of the positioning assembly all include: Hemispherical magnet.

9. A fillet weld identification device according to claim 8, characterized in that: The hemispherical magnet is a hemispherical electromagnet.

10. A fillet weld identification device according to claim 8, characterized in that: The diameter of the hemispherical magnet ranges from 10 mm to 20 mm.

Images