Welding seam track determination method and device, computer device and storage medium

Abstract

The application relates to a vertical welding seam track determination method and device, computer equipment and a storage medium. The method comprises the following steps: scanning a to-be-welded piece to obtain first point cloud data of the to-be-welded piece, the to-be-welded piece comprising a rib plate and a bottom plate; first rib plate point cloud data and first bottom plate point cloud data are screened from the first point cloud data, and the welding seam type of the vertical welding seam of the to-be-welded piece is determined according to the first rib plate point cloud data; in the case that the welding seam type is a special-shaped welding seam type, a vertical welding seam straight line and a vertical welding seam curve are fitted based on the first point cloud data, and the welding seam track of the vertical welding seam of the to-be-welded piece is generated based on the vertical welding seam straight line and the vertical welding seam curve; in the case that the welding seam type is a normal welding seam type, the welding seam starting point and the welding seam ending point of the vertical welding seam of the to-be-welded piece are determined, and the welding seam track of the vertical welding seam of the to-be-welded piece is generated according to the welding seam starting point and the welding seam ending point. The method is beneficial to improving the robustness of the vertical welding seam track generation of diversified to-be-welded pieces.

Description

Technical Field

[0001] This application relates to the field of automated welding technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for determining the trajectory of a vertical weld seam. Background Technology

[0002] Sub-assembly welding is one of the key intermediate processes in the manufacturing of large equipment such as shipbuilding, steel structure engineering, and construction machinery. By breaking down large structures into multiple smaller substructures according to modular principles and completing some welding tasks in the sub-assembly stage, the overall manufacturing efficiency and assembly quality can be significantly improved.

[0003] With the improvement of industrial automation, robotic welding systems and weld seam recognition visual algorithms are gradually being introduced into the assembly process. In robotic automated welding systems, weld seam tracking technology based on vision or laser sensors is gradually becoming a core support. Its main functions include weld seam detection, tracking, path planning, and positioning of the start and end points, thereby driving the welding actuator to complete accurate operations.

[0004] In small-scale assembly scenarios, accurately identifying the start and end points of the weld seam is a crucial step in automated welding, especially when performing vertical welding. Traditional weld seam tracking systems typically have forward-looking scanning characteristics (i.e., the scanning point is located a certain distance in front of the welding torch). Therefore, in vertical welding scenarios, the start and end points of the weld seam often cannot be fully entered into the scanning field of view, resulting in missing identification or misjudgment, which cannot guarantee the integrity of the weld. Furthermore, small-scale assembly workpieces may contain diverse structures to be welded, increasing the difficulty of weld seam tracking path identification and easily leading to large weld seam tracking path errors, which affect the accuracy of robot trajectory control.

[0005] It is evident that the robustness of the related technologies in generating vertical weld trajectories for diverse workpieces to be welded is not high. Summary of the Invention

[0006] Therefore, it is necessary to provide a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for determining vertical weld seam trajectories that can improve the robustness of vertical weld seam trajectory generation, in response to the above-mentioned technical problems.

[0007] In a first aspect, this application provides a method for determining the trajectory of a vertical weld, including:

[0008] Scan the part to be welded to obtain the first point cloud data of the part to be welded, which includes stiffening plates and bottom plates;

[0009] The point cloud data of the first stiffening plate and the point cloud data of the first base plate are filtered out from the first point cloud data. Based on the point cloud data of the first stiffening plate, the weld type of the vertical weld of the part to be welded is determined.

[0010] When the weld type is an irregular weld type, the vertical weld line and vertical weld curve are fitted based on the first point cloud data. Based on the vertical weld line and vertical weld curve, the weld trajectory of the vertical weld of the workpiece to be welded is generated.

[0011] When the weld type is normal, determine the weld start and weld end points of the vertical weld of the workpiece to be welded, and generate the weld trajectory of the vertical weld of the workpiece to be welded based on the weld start and weld end points.

[0012] Secondly, this application also provides a vertical weld trajectory determination device, comprising:

[0013] A point cloud scanning module is used to scan the workpiece to be welded and obtain the first point cloud data of the workpiece to be welded, wherein the workpiece to be welded includes stiffening plates and a bottom plate.

[0014] The weld classification module is used to filter out the first stiffening plate point cloud data and the first base plate point cloud data from the first point cloud data, and determine the weld type of the vertical weld of the workpiece to be welded based on the first stiffening plate point cloud data.

[0015] The first weld trajectory generation module is used to, when the weld type is an irregular weld type, fit a vertical weld straight line and a vertical weld curve based on the first point cloud data, and generate the weld trajectory of the vertical weld of the workpiece to be welded based on the vertical weld straight line and the vertical weld curve.

[0016] The second weld trajectory generation module is used to determine the weld start point and weld end point of the vertical weld of the workpiece to be welded when the weld type is a normal weld type, and generate the weld trajectory of the vertical weld of the workpiece to be welded based on the weld start point and the weld end point.

[0017] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in any of the above embodiments of the vertical weld trajectory determination method.

[0018] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps in any of the above embodiments of the vertical weld trajectory determination method.

[0019] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps in any of the above embodiments of the vertical weld trajectory determination method.

[0020] The above-mentioned method, apparatus, computer equipment, computer-readable storage medium, and computer program product for determining vertical weld seam trajectories classify vertical weld seams into irregular weld seam types and normal weld seam types based on the shape of the vertical weld seam of the workpiece to be welded. In this way, vertical weld seams are scanned in a targeted manner according to the characteristics of different weld seam types to generate weld seam trajectories, so as to guide automated welding. In practical applications, the process begins by scanning the workpiece to be welded to obtain its point cloud data. This data is then categorized into point cloud data for the first stiffener plate and the first base plate. Next, the weld type of the vertical weld is determined based on the first stiffener plate point cloud data. If the vertical weld type is an irregular weld type, a straight line and curve are fitted based on the first point cloud data. The weld trajectory is then generated based on these lines, enabling the tracking and positioning of irregularly shaped welds in diverse group welding scenarios. If the vertical weld type is a normal weld type, the start and end points are determined, and a weld trajectory is generated. This improves the adaptability and robustness of vertical weld trajectory generation for diverse workpieces. Furthermore, it automatically generates the weld trajectory without manual instruction, increasing the efficiency of vertical weld trajectory generation, reducing labor costs, and enhancing the flexibility and stability of small-batch welding of irregular structures. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a diagram illustrating the application environment of an embodiment of the vertical weld trajectory determination method.

[0023] Figure 2 This is a flowchart illustrating a method for determining the trajectory of an upright weld in one embodiment;

[0024] Figure 3 This is a schematic diagram of the components to be welded in one embodiment;

[0025] Figure 4 This is a schematic diagram of the weld type in one embodiment;

[0026] Figure 5 This is a flowchart illustrating the method for determining the vertical weld trajectory in another embodiment;

[0027] Figure 6 This is a flowchart illustrating the method for determining the vertical weld trajectory in yet another embodiment;

[0028] Figure 7 This is a flowchart illustrating the method for determining the vertical weld trajectory in another embodiment;

[0029] Figure 8 This is a flowchart illustrating the method for determining the vertical weld trajectory in another embodiment;

[0030] Figure 9 This is a flowchart illustrating the method for determining the vertical weld trajectory in another embodiment;

[0031] Figure 10 A structural block diagram of an embodiment of an independent weld trajectory determination device;

[0032] Figure 11 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0034] The vertical weld trajectory determination method provided in this application embodiment can be applied to, for example, Figure 1 In the application environment shown, the control terminal 102 is connected to the weld seam tracking sensor 104 and the welding robot for communication.

[0035] Specifically, the control terminal 102 can control the weld seam tracking sensor to scan the workpiece to be welded, including the stiffener and the base plate, to obtain the first point cloud data of the workpiece. Next, the first stiffener point cloud data and the first base plate point cloud data are filtered from the point cloud data. Based on the first stiffener point cloud data, the weld seam type of the vertical weld seam of the workpiece to be welded is determined. Then, if the weld seam type is an irregular weld seam type, a vertical weld seam straight line and a vertical weld seam curve are fitted based on the first point cloud data. Based on the vertical weld seam straight line and the vertical weld seam curve, the weld seam trajectory of the vertical weld seam of the workpiece to be welded is generated. If the weld seam type is a normal weld seam type, the weld seam start point and weld seam end point of the vertical weld seam of the workpiece to be welded are determined. Based on the weld seam start point and weld seam end point, the weld seam trajectory of the vertical weld seam of the workpiece to be welded is generated. The generated weld seam trajectory can be used to instruct the welding robot to perform vertical welding on the workpiece to be welded.

[0036] In one exemplary embodiment, such as Figure 2 As shown, a method for determining the trajectory of a vertical weld is provided, which can be applied to... Figure 1 Taking control terminal 102 as an example, the explanation includes the following steps (hereinafter referred to as S): S100 to S400. Wherein:

[0037] S100, scan the part to be welded to obtain the first point cloud data of the part to be welded, which includes stiffening plates and bottom plates.

[0038] The first point cloud data can be point cloud data obtained by scanning the workpiece to be welded according to a preset scanning path. The first point cloud data includes the point cloud data of the stiffener plate and the point cloud data of the base plate.

[0039] In this embodiment, vertical welding during the assembly stage is used as an application scenario. The component to be welded may include a base plate and multiple stiffening plates, and the vertical weld seam is the location where the stiffening plates meet. The base plate may be a steel plate forming the bottom of the hull. The stiffening plates may be plates used for welding or fixing to the base plate, designed to enhance the rigidity and strength of the structure. For example, as... Figure 3 As shown, the part to be welded includes a structural component consisting of a base plate and two stiffening plates.

[0040] In practical applications, coarse positioning of the workpiece to be welded can be performed first to determine its location and the area where the weld seam is located. Specifically, coarse positioning can be achieved by using a fixture to attach the workpiece to a known reference surface, locating the approximate position of the workpiece based on the fixture coordinate system, and then using a 3D camera to detect the approximate position of the workpiece and locate the approximate position of the weld seam. 3D cameras include, but are not limited to, structured light cameras, binocular stereo vision cameras, and Time-of-Flight (ToF) cameras.

[0041] Subsequently, based on the position of the workpiece to be welded, a scanning path is generated according to a preset scanning path generation method to acquire the first point cloud data of the workpiece. Specifically, the scanning starting position can be the edge of the top of the stiffening plate in the workpiece to be welded, and the scanning direction can be the direction perpendicular to the bottom plate plane towards the length of the stiffening plate. For example, as shown... Figure 3 As shown, a scan path is generated based on scan position 1 and scan direction 1.

[0042] In practice, the control unit can perform coarse positioning of the workpiece to be welded using a structured light camera to obtain positioning information including the position of the workpiece and the weld seam. Based on the positioning information, a scanning path is generated. Subsequently, the weld seam tracking sensor is controlled to scan the workpiece to be welded according to the generated scanning path to obtain the first point cloud data P of the workpiece to be welded.

[0043]

[0044] The first point cloud data includes point cloud data of the rib plate and point cloud data of part of the base plate.

[0045] S200: Select the point cloud data of the first stiffening plate and the point cloud data of the first base plate from the first point cloud data, and determine the weld type of the vertical weld of the part to be welded based on the point cloud data of the first stiffening plate.

[0046] Among them, the first rib plate point cloud data represents the point cloud data of the rib plate in the first point cloud data. The first base plate point cloud data represents the point cloud data of the base plate in the first point cloud data.

[0047] In this embodiment, weld types are classified based on the shape of the welds in the assembled workpiece, dividing them into irregular welds and normal welds. Normal welds are characterized by their regular shape, while irregular welds are characterized by their irregular shape. Normal welds may include, but are not limited to, L-shaped, T-shaped, and I-shaped welds, while irregular welds may include welds whose shape changes due to factors such as the bending of stiffeners, resulting in a curved weld. For example,... Figure 4 As shown. Based on the weld type, the parts to be welded can be divided into L-shaped workpieces, T-shaped workpieces, I-shaped workpieces, and irregular-shaped workpieces. Among them, L-shaped workpieces can be composed of a base plate and L-shaped stiffeners, T-shaped workpieces can be composed of a base plate and T-shaped stiffeners, I-shaped workpieces can be composed of a base plate and I-shaped stiffeners, and irregular-shaped workpieces can be composed of a base plate and curved stiffeners or inclined stiffeners.

[0048] In practice, since there is a distance difference between the stiffening slab and the base plate, the distance between each point cloud data in the first point cloud data can be determined. Based on the distance, the first point cloud data can be divided to obtain the first stiffening slab point cloud data and the first base plate point cloud data. Specifically, the distances can be sorted in descending order, and the maximum number of distances can be selected to determine the average of the selected distances. Then, the first point cloud data can be divided using this average as the dividing line to obtain the first stiffening slab point cloud data and the first base plate point cloud data.

[0049] Based on the point cloud data of the first stiffener, the weld type of the vertical weld of the part to be welded can be determined as follows: First, determine the curvature of each point cloud data of the first stiffener. Then, determine the maximum curvature from each curvature. If the maximum curvature is greater than a preset maximum curvature threshold, it indicates that the stiffener contains a curved structure, and the weld type of the vertical weld of the part to be welded is determined to be an irregular weld. Otherwise, the weld type of the vertical weld of the part to be welded is determined to be a normal weld. The maximum curvature threshold can be set based on classification requirements. The method for determining the curvature of each point cloud data of the first stiffener can be as follows: For each point cloud data of the first stiffener, determine the neighborhood point cloud data of the first stiffener point cloud data. Construct a covariance matrix based on the neighborhood point cloud data of the first stiffener point cloud data. Perform eigenvalue decomposition on the covariance matrix to obtain three eigenvalues. Determine the curvature of the first stiffener point cloud data based on the decomposed eigenvalues.

[0050] S300, when the weld type is an irregular weld type, based on the first point cloud data, fit the vertical weld line and vertical weld curve, and based on the vertical weld line and vertical weld curve, generate the weld trajectory of the vertical weld of the workpiece to be welded.

[0051] In practice, the curvature of each first point cloud data point can be determined first. Based on the curvature, the first point cloud data points are divided into two parts. Then, a straight line is fitted to the first point cloud data points with lower curvature to obtain the vertical weld line, and a curve is fitted to the first point cloud data points with higher curvature to obtain the vertical weld curve. Next, the starting point of the vertical weld is determined based on the vertical weld line, and the ending point of the vertical weld is determined based on the vertical weld curve. Finally, the vertical weld line and the vertical weld curve are connected by tangents or arcs to obtain the weld trajectory of the vertical weld of the workpiece to be welded.

[0052] S400, when the weld type is normal weld type, determine the weld start point and weld end point of the vertical weld of the workpiece to be welded, and generate the weld trajectory of the vertical weld of the workpiece to be welded based on the weld start point and weld end point.

[0053] The weld start point of the vertical weld can characterize the location where the vertical weld between the stiffening plates intersects with the base plate. For example, such as... Figure 3 As shown, the point where the vertical weld intersects with the base plate is the weld start point. Correspondingly, the corner where the two stiffening plates meet is the weld end point.

[0054] In practice, the following steps can be taken: A straight line for the base plate can be fitted based on the point cloud data of the base plate; a straight line for the stiffening plate can be fitted based on the point cloud data of the stiffening plate; then, a straight line for the stiffening plate can be fitted based on the point cloud data of the stiffening plate; a straight line for the top plate can be fitted based on the endpoints of the stiffening plate lines (top endpoints); the plane containing the top plate can be determined based on the endpoints of the top plate lines and the stiffening plate lines; and the plane containing the base plate can be determined based on the straight line of the base plate and the point cloud data of the base plate. Next, the position of the vertical weld seam of the workpiece to be welded is scanned to obtain the point cloud data of the vertical weld seam. A straight line for the vertical weld seam is fitted based on the point cloud data of the vertical weld seam. The intersection of the vertical weld seam line and the plane containing the base plate is determined as the weld start point, and the intersection of the vertical weld seam line and the plane containing the top plate is determined as the weld end point. Finally, the point cloud data of the vertical weld seam is sampled between the weld end point and the weld start point to generate the weld seam trajectory of the vertical weld seam of the workpiece to be welded.

[0055] In other embodiments, the generated vertical weld seam trajectory may be sent to the welding robot to instruct the welding robot to perform vertical welding on the workpiece according to the vertical weld seam trajectory.

[0056] In the above method for determining the vertical weld trajectory, the vertical weld is divided into irregular weld type and normal weld type according to the shape of the vertical weld of the workpiece to be welded. Then, the vertical weld is scanned in a targeted manner according to the characteristics of different weld types to generate weld trajectory, so as to guide automated welding. In practical applications, the process begins by scanning the workpiece to be welded to obtain its point cloud data. This data is then categorized into point cloud data for the first stiffener plate and the first base plate. Next, the weld type of the vertical weld is determined based on the first stiffener plate point cloud data. If the vertical weld type is an irregular weld type, a straight line and curve are fitted based on the first point cloud data. The weld trajectory is then generated based on these lines, enabling the tracking and positioning of irregularly shaped welds in diverse group welding scenarios. If the vertical weld type is a normal weld type, the start and end points are determined, and a weld trajectory is generated. This improves the adaptability and robustness of vertical weld trajectory generation for diverse workpieces. Furthermore, it automatically generates the weld trajectory without manual instruction, increasing the efficiency of vertical weld trajectory generation, reducing labor costs, and enhancing the flexibility and stability of small-batch welding of irregular structures.

[0057] In one exemplary embodiment, such as Figure 5 As shown, the point cloud data of the first stiffening plate and the first base plate are selected from the first point cloud data. Based on the point cloud data of the first stiffening plate and the first base plate, the weld type of the vertical weld of the part to be welded is determined, including S210 to S290, wherein:

[0058] S210, cluster the first point cloud data to obtain the clustering results.

[0059] The clustering result may include two clusters.

[0060] In practice, since there is a distance between the stiffener and the base plate, Euclidean clustering can be used to separate the point cloud data of the stiffener and the base plate from the first point cloud data. Specifically, firstly, k initial cluster centers are randomly selected from the first point cloud data. Based on the Euclidean distance between the first point cloud data, each first point cloud data is assigned to the nearest center, resulting in multiple clusters. Subsequently, the cluster centers are re-determined based on the first point cloud data within each cluster. This process of allocation and cluster center update is iterated until the first point cloud data within each cluster no longer changes, yielding the clustering result.

[0061] S230, based on the clustering results, filter out the point cloud data of the first rib plate and the point cloud data of the first base plate from the first point cloud data.

[0062] In practice, the clustering results divide the first point cloud data into two clusters. This can be achieved by determining the average Z-value of the first point cloud data for each of the two clusters, identifying the cluster with the higher average Z-value as the cluster containing the first rib point cloud data, and the cluster with the lower average Z-value as the cluster containing the first base plate point cloud data, thus obtaining the first rib point cloud data and the first base plate point cloud data.

[0063] S250, perform linear fitting on the point cloud data of the first stiffener to obtain the straight line of the first stiffener.

[0064] In practical applications, line fitting methods include, but are not limited to, least squares method, RANSAC algorithm and Hough transform.

[0065] In practice, the control unit can perform linear fitting on the point cloud data of the first rib using the least squares method to obtain multiple straight lines. From these lines, the line with the most first rib point cloud data within it is selected and designated as the first rib line. The equation of the first rib line can be expressed as:

[0066]

[0067] in, Let (x0, y0, z0) be a point on the straight line of the first stiffener, and v be the direction vector of the straight line of the first stiffener.

[0068] S270: Select the preset number of point clouds with the highest height values ​​from the straight lines of the first stiffener, and determine the height difference between each point cloud.

[0069] The preset quantity can be set according to the detection requirements. In specific implementation, the point cloud in the first rib straight line can be sorted in descending order according to the Z value. Then, the first N point clouds are selected, where N is the preset quantity. After that, the height difference (Z value difference) between the selected point clouds is determined.

[0070] S290, if any of the height differences exceeds the preset height difference threshold, the vertical weld of the part to be welded is determined to be an irregular weld type; otherwise, the vertical weld to be welded is determined to be a normal weld type.

[0071] The preset height difference threshold can be set based on the height difference of the midpoint cloud of the first stiffener line of multiple different irregular workpieces.

[0072] In practice, if the height difference exceeds the preset height difference threshold, it indicates that there is a significant height change in the stiffener area, and the vertical weld of the part to be welded is determined to be an irregular weld type. Otherwise, it indicates that there is an insignificant height change in the stiffener area, and the vertical weld of the part to be welded is determined to be a normal weld type.

[0073] In this embodiment, by fitting the straight line of the stiffener, the vertical weld type is classified based on the height difference of the midpoint cloud of the straight line, which improves the efficiency and accuracy of weld type classification. Furthermore, it is beneficial to adopt different vertical weld trajectory recognition strategies for different weld types to improve the robustness of vertical weld trajectory determination.

[0074] In one exemplary embodiment, such as Figure 6 As shown, the first point cloud data is obtained by scanning the workpiece to be welded according to the preset first scanning path, determining the weld start and weld end points of the vertical weld of the workpiece to be welded, including S410 to S490, wherein:

[0075] S410: Scan the workpiece to be welded according to the preset second scanning path to obtain the second point cloud data of the workpiece to be welded.

[0076] In this embodiment, the preset first scanning path can be generated based on the height of the stiffening plate, the position of the bottom plate, and the position of the vertical weld in the workpiece to be welded, and the preset second scanning path can be determined based on the junction of the stiffening plate and the bottom plate and the position of the vertical weld. Continuing from the above example, as... Figure 3 As shown, the first scan path is generated based on scan position 1 and scan direction 1. The preset second scan path is generated based on scan position 2 and scan direction 2.

[0077] In practice, the control terminal can control the weld seam tracking sensor to scan the workpiece to be welded according to a preset second scanning path, thereby obtaining second point cloud data containing point cloud data of part of the stiffening plate and part of the bottom plate.

[0078] In other embodiments, to optimize the second scanning path, the angle between the two ends of the first stiffener line and the normal vector of the base plate (i.e., the Z-axis) can be determined based on the direction vector of the first stiffener line and the normal vector of the base plate. The end with the larger angle is then selected as the starting point of the second scanning path. The smaller the angle, the more inclined the stiffener; therefore, selecting the end with the larger angle as the starting point ensures entry from a more open area with less interference, which helps improve the stability and accuracy of weld start-up identification.

[0079] S430, Select the second stiffener point cloud data and the second base plate point cloud data from the second point cloud data, and perform linear fitting on the second stiffener point cloud data and the second base plate point cloud data respectively to obtain the second stiffener line and the second base plate line.

[0080] Among them, the second stiffener point cloud data represents the point cloud data of the stiffener in the second point cloud data. The second base plate point cloud data represents the point cloud data of the base plate in the second point cloud data.

[0081] In specific implementation, the method for selecting the second rib point cloud data and the second base plate point cloud data from the second point cloud data can refer to the implementation method for selecting the first rib point cloud data and the first base plate point cloud data from the first point cloud data in the above embodiment, and will not be repeated here. Here, the implementation method for performing linear fitting on the second rib point cloud data and the second base plate point cloud data to obtain the second rib straight line and the second base plate straight line is the same as the implementation method for performing linear fitting on the first rib point cloud data to obtain the first rib straight line in the above embodiment, and will not be repeated here.

[0082] S450, determine the direction vector of the bottom plate line based on the straight line of the second stiffener and the straight line of the second bottom plate.

[0083] Among them, the direction vector of the base plate line is a non-zero vector that coincides with the base plate line.

[0084] In practice, the intersection point of the second stiffening plate line and the second base plate line can be determined, and the direction vector of the base plate line can be determined based on the Z value of the intersection point. In other embodiments, determining the straight line direction vector of the base plate can also involve scanning the base plate to obtain point cloud data of the base plate, performing straight line fitting on the point cloud data of the base plate to fit a straight line of the base plate, and obtaining the direction vector of the straight line of the base plate. .

[0085] S470, scan the vertical weld seam of the workpiece to be welded, obtain the point cloud data of the vertical weld seam, perform linear fitting on the point cloud data of the vertical weld seam to obtain the vertical weld seam straight line, and determine the direction vector of the vertical weld seam straight line.

[0086] In practical implementation, since the workpiece to be welded has undergone coarse positioning and the approximate location of the vertical weld is known, the weld tracking sensor is controlled to scan the vertical weld to be welded based on the position of the vertical weld obtained from the coarse positioning, thereby obtaining the point cloud data of the vertical weld. For example, as shown... Figure 3 As shown, a scan is performed from scan position 3 along scan direction 3 to obtain the point cloud data of the vertical weld. Then, a straight line fit is performed on the point cloud data of the vertical weld to obtain the direction vector of the vertical weld line. Here, the method of straight line fitting is the same as in the above embodiment, which is to perform straight line fitting on the point cloud data of the first rib to obtain the straight line of the first rib, and will not be described again here.

[0087] S490, determine the start and end points of the vertical weld seam based on the direction vector of the vertical weld seam line and the direction vector of the base plate line.

[0088] In practice, the intersection point of the vertical weld line and the base plate line can be determined based on the direction vector of the vertical weld line and the direction vector of the base plate line, thus obtaining the weld start point of the vertical weld. :

[0089]

[0090]

[0091]

[0092] in, The parameter (scalar) on the vertical weld line indicates how much movement is required to reach the intersection point. These are the coordinates of a known point on the straight line of the base plate.

[0093] Referring to the method described above for determining the weld start point of a vertical weld, the intersection point of the vertical weld line and the top plate line is determined based on the direction vectors of the vertical weld line and the top plate line, thus yielding the weld end point of the vertical weld. The top plate straight line is obtained by fitting the starting point of multiple straight lines fitted from the point cloud data of the first stiffener.

[0094] In this embodiment, the direction vectors of the base plate straight line and the vertical weld line are determined based on different scanning paths. Based on the direction vectors of the two, the starting point and ending point of the vertical weld are determined. This helps to address the problem of inaccurate vertical weld identification caused by the forward scanning characteristics of the weld tracking system and improves the robustness of the vertical weld trajectory determination.

[0095] In one exemplary embodiment, such as Figure 7 As shown, based on the first point cloud data, the vertical weld line and vertical weld curve are fitted, including S310 to S390, where:

[0096] S310, for each first point cloud data, determine the neighboring point cloud data of the first point cloud data.

[0097] In specific implementation, each first point cloud data p can be defined as the neighborhood point cloud data within a preset radius centered on the first point cloud data p. The preset radius can be set according to the fitting accuracy requirements.

[0098] S330, determine the curvature of the first point cloud data based on the first point cloud data and the neighboring point cloud data of the first point cloud data.

[0099] Curvature is a measure of the local geometry of a point cloud surface, used to describe the degree of curvature of a curve. For any point in a point cloud, there exists a surface that approximates the neighboring point cloud. The curvature at a point can be characterized by the curvature of the local surface fitted to that point and its neighboring points.

[0100] In practice, for each first point cloud data p, assuming it has N neighboring point cloud data, a covariance matrix is ​​constructed:

[0101]

[0102] in, This represents the mean of the neighborhood point cloud data.

[0103] Subsequently, eigenvalue decomposition is performed on the covariance matrix C to obtain the eigenvalues. , , The three eigenvalues ​​represent the variance of the point cloud data distribution along three mutually orthogonal principal directions.

[0104] The curvature of the first point cloud data p is determined based on the decomposed eigenvalues. :

[0105]

[0106] S350, the first point cloud data with a curvature less than or equal to a preset curvature threshold is determined as bottom point cloud data, and the first point cloud data with a curvature greater than the preset curvature threshold is determined as top point cloud data.

[0107] The bottom point cloud data represents the point cloud data in the area near the bottom plate of the stiffening plate region, where there is an approximately straight structure. The top point cloud data represents the point cloud data in the top area of ​​the stiffening plate region, where there is a curved structure.

[0108] The curvature threshold can be set according to the detection requirements.

[0109] In practice, the control unit compares the curvature of each first point cloud data with a preset curvature threshold. The first point cloud data with a curvature less than or equal to the preset curvature threshold is determined as the bottom point cloud data, and the first point cloud data with a curvature greater than the preset curvature threshold is determined as the top point cloud data. Thus, the scanned rib plate area is divided into the top area and the bottom area.

[0110] S370, perform linear fitting on the bottom point cloud data to obtain the vertical weld line.

[0111] In specific implementation, the implementation steps of performing linear fitting on the bottom point cloud data to obtain the vertical weld line can be the same as the above embodiment, which involves performing linear fitting on the first stiffener point cloud data to obtain the first stiffener line, and will not be repeated here.

[0112] S390, curve fitting is performed on the top point cloud data to obtain the vertical weld curve.

[0113] In practice, the top point cloud data can be fitted with a curve using cubic spline interpolation or polynomial fitting to obtain the vertical weld curve. .

[0114] In this embodiment, by determining the curvature of the point cloud, the stiffening plate area is divided into two parts, and the vertical weld line and vertical weld curve are fitted respectively, which helps to improve the accuracy of weld identification and positioning for irregular weld types.

[0115] In one exemplary embodiment, such as Figure 8 As shown, based on the vertical weld line and vertical weld curve, the weld trajectory of the vertical weld of the workpiece to be welded is generated, including S320 to S380, wherein:

[0116] S320, obtain the spatial curvature and deflection of the vertical weld curve.

[0117] Torque characterizes the degree to which a curve deviates from its plane at each point, i.e., the degree of "twisting" of the curve in space. Spatial curvature characterizes the bending properties of a curve in space.

[0118] In practice, the spatial curvature of the vertical weld curve is obtained through the curvature calculation formula and the function of the vertical weld curve. The deflection of the vertical weld curve is then obtained through the deflection calculation formula.

[0119] S340, determine the intersection point of the direction vector of the vertical weld line and the direction vector of the base plate line, sample the point cloud between the intersection point and the vertical weld line, and obtain the extended vertical weld line.

[0120] In practice, the direction vector of the vertical weld line is determined. Subsequently, the direction vector of the vertical weld line is determined. Direction vector of the base plate line The intersection point is the starting point of the vertical weld. Then, the point cloud between the starting point of the weld and the vertical weld line is uniformly sampled and connected with the vertical weld line to obtain the extended vertical weld line.

[0121] S360 generates a spatial spiral curve extending from the endpoint of the vertical weld curve based on the spatial curvature and deflection, thus obtaining the extended vertical weld curve.

[0122] The endpoints of the vertical weld curve are the starting points for extension.

[0123] In practice, the endpoints of the vertical weld curve are first determined. Then, the positions, tangent vectors, and curvatures of the endpoints are determined, along with the radius of curvature and scaling factor. Subsequently, a spatial spiral curve is generated using the following formula until it intersects with the direction vector of the top plate straight line. The generated spatial spiral curve and the vertical weld curve together constitute the extended vertical weld curve. The parameterized expression of the spatial spiral curve is as follows:

[0124]

[0125]

[0126] Where R=1 / k is the radius of curvature, τ is the torsion, k is the curvature, and the scaling factor ensures the continuity of the tangent vector and curvature.

[0127] S380 generates the weld trajectory of the vertical weld based on the extended vertical weld line and the extended vertical weld curve.

[0128] In practice, the extended vertical weld line and the extended vertical weld curve can be connected by tangents or arcs to obtain the vertical weld trajectory.

[0129] In this embodiment, the fitted vertical weld line and vertical weld curve are extended to construct a complete vertical weld trajectory. The extension method of the vertical weld curve, compared with the method of directly fitting with a straight line or a low-order polynomial, is beneficial to make the extended trajectory transition smoothly in position, direction and curvature, thereby improving the accuracy of weld tracking and positioning.

[0130] In other embodiments, when the weld type is an irregular weld type, generating the weld trajectory of the vertical weld of the workpiece to be welded further includes: acquiring the point cloud data of the first stiffener and the second stiffener of the workpiece to be welded respectively; performing planar fitting on the point cloud data of the second stiffener to obtain the fitting plane of the second stiffener; projecting the point cloud data of the first stiffener onto the fitting plane of the second stiffener to obtain the projection points of the point cloud data of the first stiffener on the fitting plane; and smoothing each projection point to generate the weld trajectory. The point cloud data of the first stiffener can be obtained by scanning through a preset first scanning path, and the point cloud data of the second stiffener can be obtained by scanning through a preset fourth scanning path. The fourth scanning path is generated based on the positional relationship between the second stiffener and the base plate. For example, the fourth scanning path can be a scanning path starting from scanning position 4 and extending along scanning direction 4.

[0131] In one exemplary embodiment, such as Figure 9 As shown, after determining the weld trajectory of the vertical weld seam of the workpiece to be welded, the method further includes S510 to S590, wherein:

[0132] S510, when the workpiece to be welded still requires corner welding, obtains the thickness of the stiffening plate.

[0133] Corner welding can be a method of welding the corners of two metal plates together.

[0134] In practical applications, the control terminal can be pre-set to determine whether corner welding is required on the parts to be welded, based on welding requirements, and the specifications of the stiffeners can be pre-stored to determine the thickness of the stiffeners. These specifications include, but are not limited to, the dimensions, shape, and material of the stiffeners.

[0135] In practice, when the parts to be welded also require corner welding, the thickness of the stiffeners to be welded is selected from the pre-stored stiffener specifications.

[0136] S530, determine the normal vector of the vertical weld based on the direction vector of the vertical weld line and the direction vector of the base plate.

[0137] In practice, the direction vector of the vertical weld line is used as the basis. Direction vector of the base plate line Determine the normal vector of the vertical weld plane. :

[0138]

[0139] S550, offset the weld trajectory along the normal vector of the vertical weld to obtain the offset welding trajectory.

[0140] In practice, the offset welding trajectory is obtained by offsetting the point cloud data of each point along the normal vector of the vertical weld by the thickness of the stiffening plate.

[0141] S570, based on the weld trajectory and the offset weld trajectory, determine the normal vector of the wrap angle plane of the workpiece to be welded.

[0142] In practice, the wrap angle line between the weld trajectory and the offset weld trajectory can be determined based on the weld endpoint of the weld trajectory and the offset weld endpoint of the offset weld trajectory. A point on this wrap angle line is then selected. Based on the weld endpoint of the weld trajectory, the offset weld endpoint of the offset weld trajectory, and the point on the wrap angle line, the wrap angle plane is determined. Finally, based on these three points, the normal vector of the wrap angle plane of the workpiece to be welded is determined. The offset weld endpoint is the point obtained by offsetting the weld endpoint of the weld trajectory along the thickness of the stiffening plate from the normal vector of the vertical weld.

[0143] S590, generate the corner weld trajectory based on the weld endpoint of the weld trajectory, the offset weld endpoint of the offset weld trajectory, and the normal vector of the corner plane.

[0144] In practice, the weld seam trajectory can be generated by connecting the weld seam trajectory with an arc and offsetting the weld seam trajectory. Specifically, first, the weld seam endpoint is determined. and offset weld end point The midpoint C is then used to determine the weld end point. and offset weld end point First, determine the translation vector d. Then, determine the translation vector d and the normal vector of the wrap angle plane, and determine the direction w of the perpendicular bisector. Then, construct the midpoint M based on the midpoint C, the preset bending amount h, and the direction w of the perpendicular bisector. In this way, the weld endpoint is obtained. Offset weld end point After the midpoint M, the fillet weld trajectory is generated by interpolating the arc using the three-point method. The preset bending amount can be a parameter used to control the curvature of the arc. The methods for obtaining each parameter are as follows:

[0145]

[0146] In other embodiments, after obtaining the fillet weld trajectory, the fillet weld trajectory is connected with the vertical weld trajectory to obtain the complete target weld trajectory. The target weld trajectory is then sent to the welding robot to instruct the welding robot to perform welding operations on the workpiece to be welded.

[0147] In this embodiment, the weld trajectory is offset by the thickness of the stiffener plate, and the offset weld trajectory and the weld trajectory are connected by an arc to automatically generate the corner weld trajectory. In this way, there is no need for manual teaching to generate the corner welding trajectory, which reduces labor costs and improves the flexibility and stability of welding small batches of irregular structures.

[0148] To provide a clearer explanation of the vertical weld trajectory determination method provided in this application, a specific embodiment is described below, which includes the following steps:

[0149] S1, scan the workpiece to be welded according to the preset first scanning path to obtain the first point cloud data of the workpiece to be welded, the workpiece to be welded includes stiffening plate and bottom plate.

[0150] S2, cluster the first point cloud data to obtain the clustering results, and select the first rib plate point cloud data and the first base plate point cloud data from the first point cloud data based on the clustering results.

[0151] S3. Perform linear fitting on the point cloud data of the first stiffener to obtain the first stiffener straight line. Select the preset number of point clouds with the highest height value from the first stiffener straight line and determine the height difference between each point cloud. If there is a target height difference that exceeds the preset height difference threshold among the height differences, then the vertical weld of the part to be welded is determined to be an irregular weld type. Otherwise, the vertical weld to be welded is determined to be a normal weld type.

[0152] S4. When the weld type is an irregular weld type, for each first point cloud data, determine the neighboring point cloud data of the first point cloud data. Based on the first point cloud data and the neighboring point cloud data of the first point cloud data, determine the curvature of the first point cloud data. First point cloud data with curvature less than or equal to a preset curvature threshold are determined as bottom point cloud data, and first point cloud data with curvature greater than the preset curvature threshold are determined as top point cloud data. Perform straight line fitting on the bottom point cloud data to obtain the vertical weld straight line, and perform curve fitting on the top point cloud data to obtain the vertical weld curve.

[0153] S5. Obtain the spatial curvature and deflection of the vertical weld curve, determine the intersection point of the direction vector of the vertical weld line and the direction vector of the base plate line, sample the point cloud between the intersection point and the vertical weld line to obtain the extended vertical weld line, generate a spatial spiral curve extending from the endpoint of the vertical weld curve based on the spatial curvature and deflection, obtain the extended vertical weld curve, and generate the weld trajectory of the vertical weld based on the extended vertical weld line and the extended vertical weld curve.

[0154] S6. When the weld type is a normal weld type, scan the workpiece to be welded according to the preset second scanning path to obtain the second point cloud data of the workpiece to be welded. Select the second stiffener point cloud data and the second base plate point cloud data from the second point cloud data. Perform linear fitting on the second stiffener point cloud data and the second base plate point cloud data respectively to obtain the second stiffener line and the second base plate line. Determine the direction vector of the base plate line based on the second stiffener line and the second base plate line.

[0155] S7. Scan the vertical weld seam of the workpiece to be welded to obtain the point cloud data of the vertical weld seam. Perform linear fitting on the point cloud data of the vertical weld seam to obtain the vertical weld seam straight line. Determine the direction vector of the vertical weld seam straight line. Based on the direction vector of the vertical weld seam straight line and the direction vector of the base plate straight line, determine the weld seam start point and weld seam end point of the vertical weld seam.

[0156] S8. If the workpiece to be welded still requires corner welding, obtain the thickness of the stiffener plate. Based on the direction vector of the vertical weld line and the direction vector of the base plate, determine the normal vector of the vertical weld. Offset the weld trajectory along the normal vector of the vertical weld by the thickness to obtain the offset welding trajectory. Based on the weld trajectory and the offset weld trajectory, determine the normal vector of the corner plane of the workpiece to be welded. Based on the weld endpoint of the weld trajectory, the offset weld endpoint of the offset weld trajectory, and the normal vector of the corner plane, generate the corner weld trajectory.

[0157] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0158] In one exemplary embodiment, such as Figure 10 As shown, a vertical weld trajectory determination device 600 is provided, including: a point cloud scanning module 610, a weld classification module 620, a first weld trajectory generation module 630, and a second weld trajectory generation module 640, wherein:

[0159] The point cloud scanning module 610 is used to scan the workpiece to be welded and obtain the first point cloud data of the workpiece to be welded, which includes stiffening plates and bottom plates.

[0160] The weld classification module 620 is used to filter out the first stiffening plate point cloud data and the first bottom plate point cloud data from the first point cloud data, and determine the weld type of the vertical weld of the part to be welded based on the first stiffening plate point cloud data.

[0161] The first weld trajectory generation module 630 is used to fit the vertical weld line and vertical weld curve based on the first point cloud data when the weld type is an irregular weld type, and to generate the weld trajectory of the vertical weld of the workpiece to be welded based on the vertical weld line and vertical weld curve.

[0162] The second weld trajectory generation module 640 is used to determine the weld start point and weld end point of the vertical weld of the workpiece to be welded when the weld type is normal weld type, and generate the weld trajectory of the vertical weld of the workpiece to be welded based on the weld start point and weld end point.

[0163] In an exemplary embodiment, the weld classification module 620 is further configured to cluster the first point cloud data to obtain clustering results; based on the clustering results, filter out the first stiffening plate point cloud data and the first base plate point cloud data from the first point cloud data; perform linear fitting on the first stiffening plate point cloud data to obtain the first stiffening plate straight line; filter out a preset number of point clouds with the highest height values ​​from the first stiffening plate straight line to determine the height difference between each point cloud; if there is a target height difference value among the height differences that exceeds a preset height difference threshold, then the vertical weld of the part to be welded is determined to be an irregular weld type; otherwise, the vertical weld to be welded is determined to be a normal weld type.

[0164] In an exemplary embodiment, the point cloud scanning module 610 is further configured to scan the workpiece to be welded according to a preset second scanning path to obtain the second point cloud data of the workpiece to be welded.

[0165] The second weld trajectory generation module 640 is also used to filter out the second stiffener point cloud data and the second base plate point cloud data from the second point cloud data, and to perform linear fitting on the second stiffener point cloud data and the second base plate point cloud data respectively to obtain the second stiffener line and the second base plate line; to determine the direction vector of the base plate line based on the second stiffener line and the second base plate line; to scan the vertical weld of the workpiece to be welded to obtain the vertical weld point cloud data, to perform linear fitting on the vertical weld point cloud data to obtain the vertical weld line, and to determine the direction vector of the vertical weld line; and to determine the weld start point and weld end point of the vertical weld based on the direction vector of the vertical weld line and the direction vector of the base plate line.

[0166] In an exemplary embodiment, the first weld trajectory generation module 630 is further configured to: determine neighboring point cloud data for each first point cloud data; determine the curvature of the first point cloud data based on the first point cloud data and the neighboring point cloud data; determine the first point cloud data with a curvature less than or equal to a preset curvature threshold as bottom point cloud data, and determine the first point cloud data with a curvature greater than the preset curvature threshold as top point cloud data; perform straight line fitting on the bottom point cloud data to obtain a vertical weld straight line; and perform curve fitting on the top point cloud data to obtain a vertical weld curve.

[0167] In an exemplary embodiment, the first weld trajectory generation module 630 is further configured to acquire the spatial curvature and torsion of the vertical weld curve; determine the intersection point of the direction vector of the vertical weld line and the direction vector of the base plate line; sample the point cloud between the intersection point and the vertical weld line to obtain the extended vertical weld line; generate a spatial spiral curve extending from the endpoint of the vertical weld curve based on the spatial curvature and torsion to obtain the extended vertical weld curve; and generate the weld trajectory of the vertical weld based on the extended vertical weld line and the extended vertical weld curve.

[0168] In an exemplary embodiment, the vertical weld trajectory determination device 600 further includes a corner weld trajectory generation module 650, used to obtain the thickness of the stiffening plate when the workpiece to be welded still requires corner welding; determine the normal vector of the vertical weld based on the direction vector of the vertical weld line and the direction vector of the base plate; offset the weld trajectory along the normal vector of the vertical weld by the thickness to obtain an offset weld trajectory; determine the normal vector of the corner plane of the workpiece to be welded based on the weld trajectory and the offset weld trajectory; and generate a corner weld trajectory based on the weld endpoint of the weld trajectory, the offset weld endpoint of the offset weld trajectory, and the normal vector of the corner plane.

[0169] Each module in the aforementioned vertical weld seam trajectory determination device 600 can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0170] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 11 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network. When executed by the processor, the computer program implements a method for determining the trajectory of a vertical weld seam.

[0171] Those skilled in the art will understand that Figure 11 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0172] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in any of the above embodiments of the vertical weld trajectory determination method.

[0173] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps in any of the above embodiments of the vertical weld trajectory determination method.

[0174] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in any of the above embodiments of the vertical weld trajectory determination method.

[0175] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0176] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0177] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0178] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for determining the trajectory of a vertical weld, characterized in that, The method includes: Scan the workpiece to be welded to obtain the first point cloud data of the workpiece to be welded, the workpiece to be welded includes stiffening plates and a bottom plate; The first stiffener point cloud data and the first base plate point cloud data are filtered out from the first point cloud data. Based on the first stiffener point cloud data, the weld type of the vertical weld of the part to be welded is determined. When the weld type is an irregular weld type, based on the first point cloud data, a vertical weld line and a vertical weld curve are fitted, and based on the vertical weld line and the vertical weld curve, the weld trajectory of the vertical weld of the workpiece to be welded is generated. When the weld type is a normal weld type, the weld start point and weld end point of the vertical weld of the workpiece to be welded are determined, and the weld trajectory of the vertical weld of the workpiece to be welded is generated based on the weld start point and the weld end point. The step of filtering out the first stiffening plate point cloud data and the first base plate point cloud data from the first point cloud data, and determining the weld type of the vertical weld of the part to be welded based on the first stiffening plate point cloud data, includes: Cluster the cloud data of each of the first points to obtain the clustering results; Based on the clustering results, the first rib point cloud data and the first base plate point cloud data are selected from the first point cloud data. Linear fitting is performed on the point cloud data of the first stiffener to obtain the straight line of the first stiffener; Select a preset number of point clouds with the highest height values ​​from the first stiffener straight line, and determine the height difference between each point cloud; If any of the height differences exceeds a preset height difference threshold, the vertical weld of the part to be welded is determined to be an irregular weld type; otherwise, the vertical weld to be welded is determined to be a normal weld type.

2. The method according to claim 1, characterized in that, The first point cloud data is obtained by scanning the workpiece to be welded according to a preset first scanning path; determining the weld start point and weld end point of the vertical weld of the workpiece to be welded includes: The workpiece to be welded is scanned according to the preset second scanning path to obtain the second point cloud data of the workpiece to be welded; The second stiffener point cloud data and the second base plate point cloud data are selected from the second point cloud data. Straight line fitting is performed on the second stiffener point cloud data and the second base plate point cloud data respectively to obtain the second stiffener straight line and the second base plate straight line. Determine the direction vector of the base plate line based on the second stiffener line and the second base plate line; Scan the vertical weld seam of the workpiece to be welded to obtain the point cloud data of the vertical weld seam, perform linear fitting on the point cloud data of the vertical weld seam to obtain the vertical weld seam line, and determine the direction vector of the vertical weld seam line. The start and end points of the vertical weld are determined based on the direction vector of the vertical weld line and the direction vector of the base plate line.

3. The method according to claim 2, characterized in that, The process of fitting the vertical weld line and the vertical weld curve based on the first point cloud data includes: For each of the first point cloud data, determine the neighboring point cloud data of the first point cloud data; The curvature of the first point cloud data is determined based on the first point cloud data and the neighboring point cloud data of the first point cloud data. The first point cloud data with a curvature less than or equal to a preset curvature threshold is determined as the bottom point cloud data, and the first point cloud data with a curvature greater than the preset curvature threshold is determined as the top point cloud data. Linear fitting is performed on the bottom point cloud data to obtain the vertical weld line; Curve fitting is performed on the top point cloud data to obtain the vertical weld curve.

4. The method according to claim 3, characterized in that, The step of generating the weld trajectory of the vertical weld seam of the workpiece to be welded based on the vertical weld seam straight line and the vertical weld seam curve includes: Obtain the spatial curvature and deflection of the vertical weld curve; Determine the intersection point of the direction vector of the vertical weld line and the direction vector of the base plate line, sample the point cloud between the intersection point and the vertical weld line, and obtain the extended vertical weld line; Based on the spatial curvature and the deflection, a spatial spiral curve extending from the endpoint of the vertical weld curve is generated to obtain the extended vertical weld curve; The weld trajectory of the vertical weld is generated based on the extended vertical weld line and the extended vertical weld curve.

5. The method according to any one of claims 1 to 4, characterized in that, After determining the weld trajectory of the vertical weld seam of the workpiece to be welded, the method further includes: If the workpiece to be welded still requires corner welding, the thickness of the stiffening plate is obtained; The normal vector of the vertical weld is determined based on the direction vector of the vertical weld line and the direction vector of the base plate. The weld trajectory is offset by the thickness along the normal vector of the vertical weld to obtain the offset weld trajectory; Based on the weld trajectory and the offset weld trajectory, determine the normal vector of the wrap angle plane of the workpiece to be welded; The corner weld trajectory is generated based on the weld endpoint of the weld trajectory, the offset weld endpoint of the offset weld trajectory, and the normal vector of the corner plane.

6. A vertical weld seam trajectory determination device, characterized in that, The device includes: A point cloud scanning module is used to scan the workpiece to be welded and obtain the first point cloud data of the workpiece to be welded, wherein the workpiece to be welded includes stiffening plates and a bottom plate. The weld classification module is used to filter out the first stiffening plate point cloud data and the first base plate point cloud data from the first point cloud data, and determine the weld type of the vertical weld of the workpiece to be welded based on the first stiffening plate point cloud data. The first weld trajectory generation module is used to, when the weld type is an irregular weld type, fit a vertical weld straight line and a vertical weld curve based on the first point cloud data, and generate the weld trajectory of the vertical weld of the workpiece to be welded based on the vertical weld straight line and the vertical weld curve. The second weld trajectory generation module is used to determine the weld start point and weld end point of the vertical weld of the workpiece to be welded when the weld type is a normal weld type, and generate the weld trajectory of the vertical weld of the workpiece to be welded based on the weld start point and the weld end point. The weld classification module is also used to cluster the cloud data of each first point to obtain clustering results; Based on the clustering results, the first rib point cloud data and the first base plate point cloud data are selected from the first point cloud data. Linear fitting is performed on the point cloud data of the first stiffener to obtain the straight line of the first stiffener; Select a preset number of point clouds with the highest height values ​​from the first stiffener straight line, and determine the height difference between each point cloud; If any of the height differences exceeds a preset height difference threshold, the vertical weld of the part to be welded is determined to be an irregular weld type; otherwise, the vertical weld to be welded is determined to be a normal weld type.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images