Industrial robot automatic welding method and control equipment

Abstract

The invention provides an industrial robot automatic welding method and control equipment, and relates to the technical field of machine control. The method comprises the following steps: acquiring three-dimensional point cloud data of a workpiece, and adjusting a pre-constructed BIM model of the workpiece according to the three-dimensional point cloud data, so that the adjusted BIM model is aligned with an actual workpiece position; obtaining a theoretical welding seam path from the adjusted BIM model; acquiring an actual welding seam image of the workpiece; according to the actual welding seam image and the theoretical welding seam path, calibration parameters of the robot are determined, and a final welding path is determined according to the calibration parameters; and according to the final welding path and the material and thickness of the workpiece, a final control instruction set is determined, and the robot is controlled to conduct automatic welding according to the final control instruction set. The welding precision can be improved, meanwhile, manual intervention is not needed, the welding efficiency can be greatly improved, and the method can adapt to complex and changeable welding scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of machine control, and particularly to an automatic welding method and control device for an industrial robot. Background Art

[0002] The automatic welding technology of industrial robots is of irreplaceable significance in improving production efficiency, ensuring product quality, etc. With the rapid development of intelligent manufacturing, welding robots are widely used in fields such as steel structure, automobile manufacturing, shipbuilding, etc., and their automation level directly affects the competitiveness of industrial production.

[0003] In related technologies, robot welding methods mostly rely on on-line teaching programming or fixed fixture positioning. This not only requires a large amount of preparatory work in the early stage, but also has extremely high requirements for the placement accuracy of workpieces. Once the workpiece position deviates or the model changes, the robot often cannot adapt flexibly and can only mechanically execute the preset program, resulting in poor welding accuracy and low welding efficiency. Especially in the complex and changeable steel structure welding scenarios, the problem is particularly prominent. Summary of the Invention

[0004] Embodiments of the present invention provide an automatic welding method and control device for an industrial robot to solve the problems of poor welding accuracy and low welding efficiency of existing robots in complex and changeable welding scenarios.

[0005] In a first aspect, embodiments of the present invention provide an automatic welding method for an industrial robot, including: Obtaining three-dimensional point cloud data of a workpiece, and adjusting a pre-constructed BIM model of the workpiece according to the three-dimensional point cloud data to align the adjusted BIM model with the actual workpiece position; Obtaining a theoretical weld path from the adjusted BIM model; Obtaining an actual weld image of the workpiece; Determining calibration parameters of the robot according to the actual weld image and the theoretical weld path, and determining a final welding path according to the calibration parameters; Determining a final control instruction set according to the final welding path and the material and thickness of the workpiece, and controlling the robot to perform automatic welding according to the final control instruction set.

[0006] In a possible implementation manner, obtaining a theoretical weld path from the adjusted BIM model includes: Extracting weld features from the adjusted BIM model to determine theoretical weld position information; Using a spatial alignment method to correct the theoretical weld position information to obtain corrected theoretical weld position information; Generating a theoretical initial weld path based on the corrected theoretical weld position information; Perform path integrity verification and optimization on the theoretical initial weld path, optimize it to adapt to the undulations of the actual workpiece surface, and optimize it to meet the operating requirements of the robot, obtaining the theoretical weld path.

[0007] In a possible implementation, obtain the actual weld image of the workpiece, including: Based on a scanning device, obtain the scanning data stream of the workpiece, and based on a vision system, preprocess the scanning data stream to obtain preliminary scanning data; Extract the surface features of the workpiece based on the preliminary scanning data, determine the dynamic change trend of the workpiece surface features, and based on the dynamic change trend, determine the dynamic imaging range including the weld; Based on the dynamic imaging range and optical features, separate to obtain the weld area; Perform image enhancement and noise processing on the weld area to obtain a stable weld image; Extract the weld contour according to the stable weld image to generate an image containing the weld contour; Perform contour verification on the image containing the weld contour, adjust the image generation parameters affected by environmental changes, and generate the actual weld image.

[0008] In a possible implementation, determine the calibration parameters of the robot according to the actual weld image and the theoretical weld path, including: Identify the weld edge positions in the actual weld image to obtain edge distribution data, and perform coordinate transformation on the edge distribution data to obtain the actual weld path; Calculate the offset between the actual weld path and the theoretical weld path to obtain a preliminary deviation value; For the actual area where the preliminary deviation value is greater than the preset deviation threshold, perform interpolation adjustment to obtain the adjusted edge distribution data; Based on the adjusted edge distribution data, use the least squares method to determine the calibration parameters of the robot.

[0009] In a possible implementation, determine the final welding path according to the calibration parameters, including: According to the calibration parameters, correct the attitude data of the welding head of the robot to obtain the corrected attitude data; According to the corrected attitude data, adjust and smooth the movement trajectory to obtain the smoothed trajectory data; Generate an optimized welding path according to the smoothed trajectory data, and extract the feature distribution data of the optimized welding path; If the feature distribution data does not match the preset feature threshold, then based on the least squares method, adjust the optimized welding path to obtain the adjusted path parameters; Generate a candidate control instruction set according to the adjusted path parameters, and optimize the smoothed trajectory data according to the candidate control instruction set to obtain stable trajectory data; Generate the final welding path according to the stable trajectory data.

[0010] In a possible implementation, determine the final control instruction set according to the final welding path and the material and thickness of the workpiece, including: Extract the key point coordinates in the final welding path and determine the coordinate feature data of the key point coordinates; Based on a preset process parameter database, determine the initial process parameters according to the coordinate feature data and the material and thickness of the workpiece; Filter the initial process parameters to obtain the smoothed process parameters; If the smoothed process parameters exceed the preset process parameter threshold, correct the smoothed process parameters to obtain the corrected process parameters; Determine the stability characteristics of the corrected process parameters, and adjust the final welding path according to the stability characteristics and the key point coordinates to obtain the adjusted welding path; Generate the final control instruction set according to the adjusted welding path and the corrected process parameters.

[0011] In a possible implementation, adjust the pre-constructed BIM model of the workpiece according to the three-dimensional point cloud data, including: According to the three-dimensional point cloud data, determine the optimized point cloud data and the target coordinate system reflecting the actual posture of the workpiece; Based on the target coordinate system, determine the deviation data between the optimized point cloud data and the BIM model, and determine the position transformation matrix according to the deviation data; Adjust the BIM model according to the position transformation matrix to obtain the adjusted BIM model.

[0012] In a possible implementation, determine the optimized point cloud data and the target coordinate system reflecting the actual posture of the workpiece according to the three-dimensional point cloud data, including: Remove the noise from the three-dimensional point cloud data to obtain the effective point cloud data on the surface of the workpiece; Perform clustering segmentation on the effective point cloud data to determine the position information of the workpiece in the effective point cloud data; Based on the position information of the workpiece, determine the boundary points of the contour features in the effective point cloud data; Based on the boundary points, determine the preliminary coordinate system reflecting the actual posture of the workpiece; Based on the preliminary coordinate system, perform coordinate transformation on the effective point cloud data to obtain the optimized point cloud data; Verify the contour features of the optimized point cloud data, and use the preliminary coordinate system that passes the verification as the target coordinate system.

[0013] In a possible implementation, after controlling the robot for automatic welding, it further includes: Obtain the feedback data of the robot; Based on the feedback data, determine whether the welding quality of the robot is qualified; If the welding quality of the robot is unqualified, update the final control instruction set according to the feedback data, and control the robot for automatic welding according to the updated final control instruction set.

[0014] In a second aspect, an embodiment of the present invention provides a control device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the industrial robot automatic welding method in the first aspect above or any possible implementation of the first aspect.

[0015] In the embodiment of the present invention, the BIM model of the workpiece pre-constructed is adjusted based on the actual three-dimensional point cloud data of the workpiece, so that the adjusted BIM model is completely aligned with the actual workpiece point cloud in spatial position, which can avoid welding deviation caused by the misalignment of the BIM model, improve welding accuracy, then obtain the theoretical weld path from the adjusted BIM model, and obtain the actual weld image of the workpiece. Furthermore, the difference between the theoretical weld path and the actual weld path can be determined according to the actual weld image and the theoretical weld path, so that the calibration parameters of the robot can be determined, and the final welding path can be determined according to the calibration parameters. Finally, according to the final welding path and the material and thickness of the workpiece, the final control instruction set is determined, and the robot is controlled for automatic welding according to the final control instruction set. The final control instruction set of the robot can be automatically adjusted based on the difference between theory and practice, the welding accuracy can be improved, and at the same time, no manual intervention is required, the welding efficiency can be greatly improved, and complex and changeable welding scenarios can be adapted. Description of the Drawings

[0016] Figure 1 is the implementation flowchart of the industrial robot automatic welding method provided by the embodiment of the present invention; Figure 2 is the structural schematic diagram of the industrial robot automatic welding device provided by the embodiment of the present invention. Detailed Embodiments

[0017] The embodiments of the present invention will be described in detail below with reference to the drawings.

[0018] See Figure 1, which shows the implementation flowchart of the industrial robot automatic welding method provided by the embodiments of the present invention. The execution subject of this method can be a control device. For example, it can be the controller of an industrial robot, or other control devices, etc., and no specific limitation is made here.

[0019] The above industrial robot automatic welding method is described in detail as follows: In S101, obtain the three-dimensional point cloud data of the workpiece, and adjust the pre-constructed BIM (Building Information Modeling) model of the workpiece according to the three-dimensional point cloud data, so that the adjusted BIM model is aligned with the actual position of the workpiece.

[0020] The pre-constructed BIM model of the workpiece can be understood as the theoretical model or design model of the workpiece, and can store the three-dimensional design data of the workpiece, such as geometric shape, weld position, material and thickness attributes, etc.

[0021] In the embodiments of the present application, a laser scanner can be used to collect the three-dimensional point cloud of the workpiece. This process is the basis for obtaining the surface space information of the workpiece, and the three-dimensional point cloud data of the workpiece can be obtained through the three-dimensional point cloud collection. This three-dimensional point cloud data is the actual point cloud data of the workpiece. For example, in automobile manufacturing, when scanning the door weld, the laser scanner captures the light points reflected by the workpiece surface at a speed of tens of thousands of points per second, forming a dense three-dimensional point cloud data. These data contain information such as the height, width and depth of the workpiece. The advantage of this method is non-contact measurement, which avoids potential damage to the workpiece caused by physical contact and improves efficiency at the same time.

[0022] The above BIM model can be adjusted through the three-dimensional point cloud data of the workpiece to obtain the adjusted BIM model. Among them, the spatial position of the adjusted BIM model is completely aligned (coincident) with the point cloud data of the actual workpiece. The embodiments of the present application can avoid welding deviation caused by the misalignment of the BIM model and improve welding accuracy.

[0023] In S102, obtain the theoretical weld path from the adjusted BIM model.

[0024] In the embodiments of the present application, the weld of the workpiece in the adjusted BIM model is called the theoretical weld, and the corresponding weld path is called the theoretical weld path. The theoretical weld path can be represented in the form of coordinates from the starting point to the ending point of the weld, or in other forms, and no specific limitation is made here.

[0025] The above theoretical weld path can include the paths of at least one theoretical weld of the workpiece.

[0026] In S103, obtain the actual weld image of the workpiece.

[0027] In the embodiment of the present application, an actual weld image of the workpiece can be obtained based on a scanning device and a vision system. Exemplarily, an actual weld image of the workpiece can be obtained in real time.

[0028] In S104, calibration parameters of the robot are determined according to the actual weld image and the theoretical weld path, and a final welding path is determined according to the calibration parameters.

[0029] In the embodiment of the present application, the actual weld path of the workpiece can be determined according to the actual weld image. Based on the difference between the actual weld path and the theoretical weld path, calibration parameters of the robot can be determined. The calibration parameters of the robot can be used to adjust the posture of the welding head of the robot. A final welding path can be determined according to the calibration parameters of the robot. The final welding path can be understood as the path along which the robot welds the weld of the workpiece.

[0030] In S105, a final control instruction set is determined according to the final welding path, the material and thickness of the workpiece, and the robot is controlled to perform automatic welding according to the final control instruction set.

[0031] The final control instruction set can include the path for welding the workpiece and the corresponding process parameters.

[0032] In the embodiment of the present application, the material and thickness of the workpiece can be obtained. According to the material and thickness of the workpiece and the final welding path determined in the foregoing steps, a final control instruction set of the robot is generated, so as to control the robot to perform automatic welding on the workpiece according to the final control instruction set, which can improve the welding accuracy and efficiency.

[0033] In the embodiment of the present application, the pre-constructed BIM model of the workpiece is adjusted based on the actual three-dimensional point cloud data of the workpiece, so that the adjusted BIM model is completely aligned with the actual workpiece point cloud in terms of spatial position, which can avoid welding deviation caused by the misalignment of the BIM model, improve the welding accuracy, then obtain the theoretical weld path from the adjusted BIM model, and obtain the actual weld image of the workpiece. Furthermore, the difference between the theoretical weld path and the actual weld path can be determined according to the actual weld image and the theoretical weld path, so that the calibration parameters of the robot can be determined, and the final welding path can be determined according to the calibration parameters. Finally, a final control instruction set is determined according to the final welding path, the material and thickness of the workpiece, and the robot is controlled to perform automatic welding according to the final control instruction set. The final control instruction set of the robot can be automatically adjusted based on the difference between theory and practice, which can improve the welding accuracy. At the same time, without manual intervention, the welding efficiency can be greatly improved, and complex and changeable welding scenarios can be adapted.

[0034] The above embodiments introduce the overall implementation process of the automatic welding method for industrial robots. Next, each step will be introduced in detail.

[0035] In some embodiments, in S101, the adjustment of the BIM model of the workpiece pre-constructed according to the three-dimensional point cloud data may include: Determine the optimized point cloud data and the target coordinate system reflecting the actual posture of the workpiece according to the three-dimensional point cloud data; Based on the target coordinate system, determine the deviation data between the optimized point cloud data and the BIM model, and determine the position transformation matrix according to the deviation data; Adjust the BIM model according to the position transformation matrix to obtain the adjusted BIM model.

[0036] Among them, the optimized point cloud data is the point cloud data after denoising, clustering segmentation, contour recognition, and coordinate transformation of the three-dimensional point cloud data.

[0037] The target coordinate system is a coordinate system that can reflect the actual posture of the workpiece. Exemplarily, the origin of the target coordinate system may be the centroid of the point cloud data of the workpiece main body, the X-axis is the eigenvector of the main direction of the workpiece main body, the Y-axis is the eigenvector of the secondary direction of the workpiece main body, and the Z-axis is the axis perpendicular to the XY plane; or, the origin of the target coordinate system may be the centroid of the point cloud data of the workpiece main body, the X-axis is the axis parallel to the fitting line of the weld seam of the workpiece, the Y-axis is the axis perpendicular to the X-axis, and the Z-axis is the axis perpendicular to the XY plane; or, the origin of the target coordinate system may be the starting point of the weld seam of the workpiece, the X-axis is the direction of the fitting line of the weld seam of the workpiece, the Y-axis is the axis perpendicular to the X-axis, and the Z-axis is the axis perpendicular to the XY plane.

[0038] The optimized point cloud data can also be understood as the point cloud data transformed into the target coordinate system.

[0039] In the embodiments of the present application, based on the target coordinate system, a registration algorithm can be used to preliminarily match the optimized point cloud data with the BIM model to determine the deviation data. Among them, the core of the registration algorithm is to align the optimized point cloud data with the BIM model.

[0040] Exemplarily, in the scenario of welding a workpiece, assuming that the optimized point cloud data comes from the weld seam area of the car door, and the BIM model is the design blueprint of the car door, the centroid or center of the two can be preliminarily aligned through rough registration. For example, the geometric center of the optimized point cloud data is coincided with the center of the BIM model, and then fine-tuning is performed using key feature points.

[0041] Among them, during rough registration, due to uneven point cloud density, there may be a deviation of about 5 millimeters in the initial alignment, and this deviation needs further analysis. After preliminary matching (rough registration), the determination of the deviation data can be achieved by comparing the optimized point cloud data with the key features of the BIM model. In one possible implementation, the starting point and the ending point of the weld can be selected as the key features, and the deviation data can be the deviation data of the key features. For example, the deviation data of the starting point and the ending point of the weld.

[0042] Suppose in the target coordinate system, the position coordinates of the starting point and the ending point of the weld in the optimized point cloud data are (102 mm, 52 mm, 10 mm) and (150 mm, 53 mm, 10 mm) respectively, while in the theoretical coordinate system of the BIM model, the position coordinates of the starting point and the ending point of the corresponding weld in the BIM model are (100 mm, 50 mm, 10 mm) and (148 mm, 51 mm, 10 mm) respectively. Through comparison, the deviation data is 2 mm in the X direction, 2 mm in the Y direction, and 0 mm in the Z direction. This method is relatively intuitive, can quickly locate the problem area, and provides a basis for subsequent adjustment.

[0043] Among them, the deviation data includes not only the magnitude of the deviation data, but also the direction of the deviation data. Through the deviation data, the translation and rotation requirements of the optimized point cloud data relative to the BIM model can be determined. For example, if the optimized point cloud data is shifted to the upper right as a whole, the calculation shows that it needs to be translated 2 mm in the positive X-axis direction and 2 mm in the positive Y-axis direction. It can be understood that this geometric calculation is based on spatial vector analysis, and the position transformation matrix essentially describes how the optimized point cloud data "moves" to the ideal position. Exemplarily, suppose it is also detected that there is a 3-degree clockwise rotation deviation in the optimized point cloud data, then the position transformation matrix needs to include both rotation and translation information. This method ensures the comprehensiveness of the transformation. Determining the final coordinate transformation result according to the position transformation matrix is an embodiment of applying theoretical calculation to actual data. In one possible implementation, after the position transformation matrix is applied to the point cloud, the coordinate of the starting point of the weld is adjusted from (102 mm, 52 mm, 10 mm) to (100 mm, 50 mm, 10 mm), which is completely consistent with the BIM model.

[0044] The position transformation matrix of the embodiment of the present application can not only correct the position, but also maintain the geometric characteristics of the point cloud. For example, the length and width of the weld are not stretched or compressed. Exemplarily, if the original point cloud causes the weld to be inclined due to the scanning angle, it can be restored to the horizontal state in the design after transformation. This adjustment provides a high-precision basis for subsequent workpiece positioning.

[0045] It should be noted that the above description is given by the method of adjusting the optimized point cloud data to align with the BIM model. However, the position transformation matrix in the embodiments of the present application is a position transformation matrix including translation and rotation, which is subsequently applied to the BIM model to align the BIM model with the optimized point cloud data, so that the two are accurately aligned in the same coordinate system, so that the theoretical weld path extracted from the model can directly guide actual welding.

[0046] It should be noted that the choice of registration algorithm will affect the matching efficiency. For example, if registration based on key features is adopted, it is suitable for scenarios with obvious linear features such as welds; if the surface of the workpiece is complex, it can be extended to a global registration scheme. Preferably, combining coarse registration and fine registration can improve the processing speed while ensuring accuracy.

[0047] According to the position transformation matrix in the embodiments of the present application, the BIM model is adjusted to obtain an adjusted BIM model, so that the adjusted BIM model is aligned with the actual position of the workpiece.

[0048] Exemplarily, the BIM model and the optimized point cloud data can be aligned. Assume that the deviation between the BIM model and the optimized point cloud data is 5 cm in the X-axis, 3 cm in the Y-axis, and 0 cm in the Z-axis. After adjustment by the position transformation matrix, the coordinates of each part of the BIM model can match the actual position of the workpiece. By adjusting the BIM model, the accuracy of subsequent feature extraction can be ensured.

[0049] Preferably, during the process of adjusting the BIM model, spatial grid division can be introduced to divide the BIM model into multiple sub-regions for separate processing, avoiding the cumulative error caused by excessive overall deviation.

[0050] In some embodiments, the above-mentioned determining the optimized point cloud data and the target coordinate system reflecting the actual posture of the workpiece according to the three-dimensional point cloud data includes: Removing noise from the three-dimensional point cloud data to obtain effective point cloud data on the surface of the workpiece; Performing clustering segmentation on the effective point cloud data to determine the position information of the workpiece in the effective point cloud data; Based on the position information of the workpiece, determining the boundary points of the contour features in the effective point cloud data; Based on the boundary points, determining a preliminary coordinate system reflecting the actual posture of the workpiece; Based on the preliminary coordinate system, performing coordinate transformation on the effective point cloud data to obtain the optimized point cloud data; Performing contour feature verification on the optimized point cloud data and using the verified preliminary coordinate system as the target coordinate system.

[0051] In addition to containing information such as the height, width, and depth of the workpiece, the three-dimensional point cloud data is also mixed with environmental noise. The collected three-dimensional point cloud data may show an irregular distribution and needs to be further processed to extract useful information. Specifically, the collected data on the surface of the workpiece can be separated from the three-dimensional point cloud data, and the noise can be removed to obtain effective point cloud data, which can be achieved through filtering techniques. In one possible implementation, a statistical filter can be used to remove outliers (noise). For example, assuming that the average distance of the neighborhood of a certain point in the three-dimensional point cloud data is 0.5 mm, and the distance of a certain noise point exceeds 1.5 mm, it is regarded as noise and removed. After separation, the distribution of the effective point cloud data on the surface of the workpiece is clearer. For example, the point cloud in the weld area shows a strip-like feature. Removing noise can effectively reduce the interference in subsequent analysis and improve the data quality.

[0052] Using a data extraction method, the effective point cloud data is clustered and segmented, and the effective point cloud data is grouped according to the spatial distance, and two groups, namely the workpiece body and the weld, can be separated, and then the position information of the workpiece in the effective point cloud data can be determined. Among them, the position information of the workpiece may include the position information of the weld and / or the position information of the workpiece body.

[0053] In one possible implementation, the Euclidean clustering method can be used to group the effective point cloud data according to the spatial distance. For example, it can be divided into two clusters, namely the weld and the workpiece body. The position information of the weld can be determined by the geometric center of the cluster corresponding to the weld. For example, it may be (100 mm, 50 mm, 10 mm). The position information of the workpiece body can be determined by the geometric center of the cluster corresponding to the workpiece body. If there is more than one weld, each weld can be clustered into a cluster, and each weld corresponds to a position information.

[0054] The method for determining the position information of the workpiece provided by the embodiments of the present application is intuitive and efficient, and can provide a reliable position reference for subsequent feature extraction.

[0055] In the embodiments of the present application, based on the position information of the workpiece, the boundary points of the contour features in the effective point cloud data are determined, which is the key to point cloud analysis. In a possible implementation manner, the above boundary points can be determined by the method of normal vector analysis. Exemplarily, the direction of the normal vector of the boundary points of the contour features (such as the weld boundary points) changes greatly, while the plane area is relatively stable. Set the normal vector angle threshold. Based on the position information of the weld, if the angle between the normal vector of a certain point on the weld and the normal vectors of at least one of its target adjacent points is greater than the normal vector angle threshold, then determine that certain point as the boundary point of the weld; or, if the maximum angle among the angles between the normal vector of a certain point and the normal vectors of its respective target adjacent points is greater than the normal vector angle threshold, then determine that certain point as the boundary point of the weld. Among them, the target adjacent point of a certain point on the weld is the point adjacent to that certain point on the contour feature of the weld. Similarly, based on the position information of the workpiece body, using the same method, the boundary points of the contour of the workpiece body can be determined.

[0056] Among them, the normal vector angle threshold can be set according to actual needs. For example, it can be 30 degrees. The first preset quantity can be set according to actual needs and will not be specifically limited here.

[0057] This method can accurately outline the weld contour and is helpful for subsequent spatial positioning.

[0058] Based on the determined boundary points above, spatial coordinates can be generated through geometric calculations, and a preliminary coordinate system reflecting the actual posture of the workpiece can be determined, which can also be called the coordinate system framework. Exemplarily, the least squares method can be used to fit the boundary points of the weld to obtain a line direction vector, such as (1, 0, 0). This vector can be used as the X-axis direction, and at the same time, taking the center of the workpiece as the origin, a preliminary coordinate system is established. The preliminary coordinate system reflects the actual posture of the workpiece, provides a reference framework for point cloud adjustment, and reduces the deviation caused by the scanning angle.

[0059] Through the preliminary coordinate system, the distribution of the effective point cloud data can be adjusted to obtain optimized point cloud data, that is, the effective point cloud data is transformed into the preliminary coordinate system to obtain the optimized point cloud data in the preliminary coordinate system. This process can be achieved through rigid transformation. Exemplarily, if the workpiece is tilted by 15 degrees, the effective point cloud data can be aligned to the standard coordinate system, that is, the aforementioned preliminary coordinate system, through a rotation matrix. The distribution of the optimized point cloud data is more regular, and the weld feature changes from a side view to a front view, which is convenient for subsequent verification. This adjustment can significantly improve the visualization effect and analysis accuracy of the point cloud.

[0060] Verify the contour features using the optimized point cloud data to determine the final spatial coordinate system, i.e., the target coordinate system. Among them, the iterative closest point algorithm can be used for verification. Exemplarily, compare the optimized point cloud data with the standard model (also called the ideal model) of the workpiece. If the deviation is less than 0.1 mm, determine the preliminary coordinate system as the target coordinate system; otherwise, return to the previous steps to re-determine the preliminary coordinate system.

[0061] The above verification ensures the accuracy of the weld position and provides a reliable basis for subsequent robot path planning. The overall process progresses step by step, not only improving the usability of the point cloud data but also bringing higher precision and stability to automated welding.

[0062] In some embodiments, the above S102 may include: Extract the weld features from the adjusted BIM model to determine the theoretical weld position information; Use a spatial alignment method to correct the theoretical weld position information to obtain the corrected theoretical weld position information; Generate a theoretical initial weld path based on the corrected theoretical weld position information; Perform path integrity verification and optimization, optimization for adapting to the surface undulation of the actual workpiece, and optimization for adapting to the operation requirements of the robot on the theoretical initial weld path to obtain the theoretical weld path.

[0063] The embodiment of the present application can extract the weld features from the adjusted BIM model to determine the theoretical weld position information. Among them, the contour of the theoretical weld in the adjusted BIM model can be identified through edge detection technology to obtain the theoretical weld position information. Exemplarily, on a steel member with a length of 2 meters, a weld feature with a width of 1 cm and a depth of 0.5 cm is detected through edge detection technology, and thus the weld position information can be determined. When extracting the weld features, the analysis of the illumination angle can also be combined to avoid misjudgment caused by surface reflection. This method relies on the accuracy of the adjusted BIM model and can provide a reliable basis for subsequent correction.

[0064] The embodiments of the present application can adopt a spatial alignment method to verify or correct the theoretical weld position information and obtain the corrected theoretical weld position information. In a possible implementation, the theoretical weld position information can be compared with the actual weld position information. If the distance between the theoretical weld position information and the actual weld position information exceeds a preset distance, for example, the preset distance can be 2 millimeters, then the theoretical weld position information needs to be corrected to obtain the corrected theoretical weld position information. Among them, the corrected theoretical weld position information can be obtained through spatial projection correction. The above-mentioned actual weld position information can be the actual weld position information in the optimized point cloud data, or the actual weld position information determined by scanning the weld position of the workpiece with a laser scanner, etc.

[0065] In some possible implementations, during the process of correcting the above-mentioned theoretical weld position information, multi-angle data fusion can also be combined to further improve the accuracy.

[0066] By correcting the theoretical weld position information, the reliability of the weld distribution can be significantly improved, laying a foundation for path planning.

[0067] Based on the corrected theoretical weld position information, a theoretical initial weld path can be generated. For example, for a theoretical weld of a curve, a theoretical initial weld path including 100 nodes can be generated based on the theoretical weld position information of the curve. The adjacent nodes can be spaced 2 centimeters apart to initially form a continuous welding trajectory. The theoretical initial weld path directly reflects the spatial characteristics of the weld, which is helpful for subsequent optimization. Exemplarily, if there is a corner in the weld, the node density can be increased at the corner to ensure a smooth transition of the path.

[0068] Perform path integrity verification on the theoretical initial weld path. If the theoretical initial weld path does not cover the entire length of the theoretical weld, then interpolate to supplement the weld nodes to form a path that covers the entire length of the theoretical weld. For example, in a 3-meter-long weld, it is detected that the theoretical initial weld path misses a 5-centimeter area at the end, then nodes can be interpolated to supplement to form a complete path.

[0069] The coordinates of each node in the theoretical initial weld path can be finely adjusted. For example, the Z-axis coordinate of a certain node can be adjusted from 10 centimeters to 10.2 centimeters, etc., to adapt to the undulation of the actual workpiece surface.

[0070] The operating requirements of the robot can also be considered to determine whether the path curvature in the theoretical initial weld path exceeds the capabilities of the robot. If so, the path curvature in the theoretical initial weld path is adjusted so that the path curvature is within the capabilities of the robot, thereby effectively avoiding path interruption or redundancy and enhancing practicality. It is determined whether the theoretical initial weld path can ensure that the welding head of the robot can always maintain the best angle with the workpiece. If not, the theoretical initial weld path is optimized so that the optimized path can ensure that the welding head of the robot can always maintain the best angle with the workpiece, thereby improving welding efficiency and quality.

[0071] The path obtained by optimizing the theoretical initial weld path is called the theoretical weld path. The optimized theoretical weld path not only conforms to the weld distribution, but also adapts to the operating requirements of the robot, and can provide an accurate theoretical basis for automated welding, reducing manual intervention.

[0072] In some embodiments, the above S103 may include: Based on a scanning device, obtain the scanning data stream of the workpiece, and based on a vision system, preprocess the scanning data stream to obtain preliminary scanning data; Extract the surface features of the workpiece based on the preliminary scanning data, determine the dynamic change trend of the surface features of the workpiece, and determine the dynamic imaging range including the weld according to the dynamic change trend; Based on the dynamic imaging range and optical features, separate the weld area; Perform image enhancement and noise processing on the weld area to obtain a stable weld image; Extract the weld contour according to the stable weld image to generate an image including the weld contour; Perform contour verification on the image including the weld contour, adjust the image generation parameters affected by environmental changes, and generate the actual weld image.

[0073] Among them, the scanning device can be a laser scanner, a high-resolution camera, etc.

[0074] The real-time scanning data stream of the workpiece can be obtained through the scanning device, and the purpose is to capture the original information on the surface of the workpiece. Exemplarily, assume that a laser scanner is used to scan a 1-meter-long steel member at a speed of 100,000 points per second, and a point cloud data stream including information such as the surface height and texture of the steel member can be generated.

[0075] The vision system can be a real-time image acquisition and processing module integrating hardware and software, mainly used for dynamically capturing, analyzing, and matching the weld features on the surface of workpieces. Its hardware part usually includes a high-resolution industrial camera (such as a CCD (Charge-Coupled Device) / CMOS (Complementary Metal Oxide Semiconductor) sensor), an anti-interference active light source (such as structured light or laser-assisted illumination), and an optical filter component, which is used to achieve stable imaging in complex welding environments (such as strong arc light, soot, etc.); the software part is based on image processing algorithms (such as filtering, edge enhancement) or deep learning models (such as convolutional neural networks) to extract the weld contour in real time, obtain the actual weld path, and perform high-precision matching with the theoretical weld path of the BIM model, so as to calculate the position offset of the actual weld. In addition, the vision system needs to cooperate synchronously with the laser scanner and control equipment to ensure consistent data timing.

[0076] Based on the vision system, preprocess the scanned data stream to obtain preliminary scanned data, which can include: based on the vision system, perform downsampling and filtering of ambient light interference on the scanned data stream to obtain preliminary scanned data. Downsampling can reduce the data volume. For example, the point cloud density can be reduced from 100 points per square centimeter to 20 points per square centimeter, reducing the subsequent processing burden while retaining the main features. Filtering out ambient light interference can ensure that the data reflects the true surface state of the workpiece. The above-mentioned preliminary scanned data can also be referred to as the preliminary surface distribution.

[0077] Extract the surface features of the workpiece based on the preliminary scanned data. The key to determining the dynamic change trend of the workpiece surface features lies in identifying the key geometric information. Exemplarily, through curvature analysis, it can be judged whether the surface is a plane or a curved surface. If it is a curved surface, it indicates that there may be an edge or a depression here, and there may be a weld. For example, on a curved steel plate, it is detected that the curvature of a certain area changes from 0 to 0.5, indicating that there may be an edge or a depression here; it is also possible to track the change in height values during the scanning process to determine whether the workpiece moves due to vibration. For example, if the height of a certain point fluctuates from 5 cm to 5.2 cm in the preliminary scanned data corresponding to 10 consecutive frames, it indicates that the workpiece may have a small displacement due to vibration.

[0078] According to the dynamic change trend of the workpiece surface features, the dynamic imaging range including the weld can be further determined. For example, an area where the height value change exceeds a preset height change threshold can be locked as the dynamic imaging range. Or, an area where the curvature is greater than a preset curvature threshold and the height value change exceeds the preset height change threshold can be used as the dynamic imaging range, which helps to focus on the weld in the subsequent process. The preset height change threshold and the preset curvature threshold can be determined according to actual needs. For example, the preset height change threshold can be 0.1 cm, and the preset curvature threshold can be 0, 0.1 or 0.2, etc.

[0079] For the dynamic imaging range, the weld area and the surrounding area can be distinguished based on optical features. Specifically, the weld area and the surrounding area can be distinguished based on the reflection value. The reflection value of the weld area is lower than that of the surrounding area by a certain value. For example, in a weld area with a width of 8 mm, its reflection value is 20% lower than that of the base material. Based on this, the weld area can be separated.

[0080] Image enhancement of the weld area can make the weld details more prominent. For example, the contrast stretching algorithm can be used to expand the gray value range from 50 - 100 to 0 - 255, etc. If there is noise in the weld area after image enhancement, such as random bright spots caused by dust, then Gaussian filtering can be used for smoothing processing to obtain a stable image structure, that is, a stable weld image.

[0081] For the stable weld image, edge detection technology can be used to extract the weld contour and generate an image containing the weld contour. For example, on a weld with a length of 50 cm, the coordinates of the boundary points are detected to change continuously, thus forming a clear trajectory, that is, the weld contour.

[0082] For the image containing the weld contour, contour verification can be performed to verify the impact caused by environmental changes based on the contour data, so as to adjust the image generation parameters affected by environmental changes and generate the actual weld image. For example, if it is found through contour verification that the light intensity has decreased by 10%, the exposure parameters can be adjusted to regenerate the image to ensure the integrity of the contour, thus generating the actual weld image. The above actual weld image can provide a reliable basis for subsequent welding path planning.

[0083] In some possible implementation manners, during the process of generating the actual weld image, if it is found that the weld area appears blurred due to surface corrosion or other reasons, multi - spectral data can be combined to enhance the recognition accuracy. Thus, it can effectively cope with complex environments and ensure image quality.

[0084] In some embodiments, in the above S104, according to the actual weld image and the theoretical weld path, the calibration parameters of the robot are determined, including: Identify the weld edge positions in the actual weld image, obtain the edge distribution data, and perform coordinate transformation on the edge distribution data to obtain the actual weld path; Calculate the offset between the actual weld path and the theoretical weld path to obtain the preliminary deviation value; For the actual area where the preliminary deviation value is greater than the preset deviation threshold, perform interpolation adjustment to obtain the adjusted edge distribution data; Based on the adjusted edge distribution data, use the least squares method to determine the calibration parameters of the robot.

[0085] In the embodiment of the present application, the details in the actual weld image can be captured based on the first convolutional neural network model, the weld edge positions in the actual weld image can be identified, and the edge distribution data can be obtained. The first convolutional neural network model can extract the texture and shape features of the weld edge through multiple convolutional operations to obtain the edge distribution data. For example, in an image containing a 50 - centimeter - long weld, the first convolutional neural network model can identify the gray - scale change pattern in the edge area and output an edge probability distribution map, that is, obtain the edge distribution data. By this method, the interference caused by uneven illumination in the image can be effectively dealt with, and relatively accurate edge distribution data can be obtained.

[0086] Perform coordinate transformation on the edge distribution data to convert the edge probability distribution map into specific coordinate points, so as to obtain the actual weld path represented by coordinates. These coordinate points can be pixel coordinate points or coordinate points in the target coordinate system, which can be determined according to actual needs. Exemplarily, assume that in the coordinate points corresponding to the edge distribution data of a weld, it is detected that the abscissa of a certain continuous boundary point changes from 100 pixels to 120 pixels, indicating that there may be a slight bend here.

[0087] By comparing each coordinate point in the actual weld path with the corresponding coordinate point in the theoretical weld path, determine the offset between the actual weld path and the theoretical weld path to obtain the preliminary deviation value. For example, if the coordinate of a certain point in the theoretical weld path is (100 pixels, 50 pixels, 10 pixels), and the coordinate of the corresponding point in the actual weld path is (102 pixels, 51 pixels, 10 pixels), then the offset is a lateral offset of 2 pixels, a longitudinal offset of 1 pixel, and a depth - direction offset of 0 pixels, and the preliminary deviation value includes this offset. The quantization of the preliminary deviation value helps with subsequent adjustments.

[0088] It should be noted that if the offsets between the respective coordinate points in the actual weld path and the corresponding coordinate points in the theoretical weld path are all the same, the preliminary deviation value may include this offset; if the offsets between the respective coordinate points in the actual weld path and the corresponding coordinate points in the theoretical weld path are not completely the same, the preliminary deviation value may include the corresponding relationship between the coordinate points and the offsets, and the coordinate points may be the respective coordinate points in the actual weld path or the respective coordinate points in the theoretical weld path.

[0089] For the actual area where the preliminary deviation value is greater than the preset deviation threshold, specifically, the preliminary deviation value being greater than the preset deviation threshold means that the lateral offset, longitudinal offset, or depthwise offset in the offsets included in the preliminary deviation value is greater than or equal to the preset deviation threshold. The above-mentioned actual area refers to the area in the actual weld path corresponding to the preliminary deviation value being greater than the preset deviation threshold. The preset deviation threshold can be set according to actual needs. For example, it can be 3 pixels, etc.

[0090] For the above-mentioned actual area, correction is required. The preferred method is to use the interpolation method to adjust the edge position to obtain the adjusted edge distribution data. Exemplarily, in a weld with a length of 10 cm, it is detected that the lateral offset in the preliminary deviation value of a certain area is 3 pixels. The edge point coordinates are smoothly adjusted from (104, 50, 10) to (101, 50, 10) through linear interpolation to generate the adjusted edge distribution data. Exemplarily, the function corresponding to this linear interpolation can be: x(k)=104 - k, y(k)=50, z(k)=10, k = 0, 1, 2, 3, where k is the index of the point, x(k) is the coordinate of the X-axis, y(k) is the coordinate of the Y-axis, and z(k) is the coordinate of the Z-axis; when k = 0, the coordinate is the above-mentioned (104, 50, 10), when k = 1, the coordinate is (103, 50, 10), when k = 2, the coordinate is (102, 50, 10), and when k = 3, the coordinate is the above-mentioned (101, 50, 10); that is, points (103, 50, 10), (102, 50, 10), and (101, 50, 10) are added after (104, 50, 10) to smoothly adjust the edge point coordinates from (104, 50, 10) to (101, 50, 10). This processing method can maintain the continuity of the edge.

[0091] When determining the calibration parameters of the robot based on the adjusted edge distribution data, the least squares method can be used to optimize the adjustment process. Specifically, the actual weld path can be re-determined based on the adjusted edge distribution data, and the relationship between the actual weld path and the theoretical weld path can be fitted using the least squares method to calculate the adjustment coefficient, which can be used as the calibration parameter of the robot. The calibration parameter can include at least one of a scaling factor, a translation amount (lateral and / or longitudinal translation amount), and a rotation angle. For example, on a 20-cm-long weld, the scaling factor is determined to be 0.95 by the least squares method, indicating that the actual weld path needs to be slightly scaled to fit the theoretical design. Correspondingly, when the robot performs welding, it needs to adjust its corresponding parameters to fit this adjustment; if the lateral translation amount is determined to be 2 mm by the least squares method, the robot needs to move 2 mm and / or rotate the corresponding angle during welding to align with the actual weld; if the rotation angle is determined to be 5 degrees by the least squares method, the robot needs to rotate 5 degrees and / or move the corresponding distance during welding to align with the actual weld; and so on. The above method can significantly improve the calibration accuracy.

[0092] In some possible implementation manners, a coordinate transformation can be performed on the actual weld image by the above adjustment coefficient. When performing the coordinate transformation, each pixel point in the image can be remapped according to the coefficient.

[0093] In one possible implementation manner, the resolution of an original image is 1920x1080, and after adjustment, a calibrated actual weld path is generated, in which the weld trajectory is closer to the theoretical expectation. When extracting boundary information from the calibrated actual weld path, it can be determined whether the path is complete by detecting continuity. For example, if the distance between certain boundary points suddenly jumps from 1 pixel to 5 pixels, it may indicate a break here and further inspection is needed. When generating a complete weld trajectory (actual weld path) according to the stable path structure, small jitters can be eliminated by smoothing filtering.

[0094] In one possible implementation manner, on a 30-cm-long weld, after applying mean filtering, the fluctuation between trajectory points is reduced from 0.2 cm to 0.05 cm, forming a smooth final actual weld path. This smoothing process can provide a more stable reference basis for subsequent processes.

[0095] It should be noted that if the weld edge is blurred due to surface roughness, the matching accuracy can be enhanced by combining multi-frame image data. For example, 5 frames of images are continuously acquired, and the reliability of the distribution data is improved by averaging the edge positions. This method is particularly useful in a dynamic environment and can effectively cope with the challenges brought by workpiece displacement or light changes.

[0096] In some embodiments, in the above S104, determining the final welding path according to the calibration parameter includes: According to the calibration parameters, correct the attitude data of the welding head of the robot to obtain the corrected attitude data; According to the corrected attitude data, adjust and smooth the movement trajectory to obtain the smoothed trajectory data; Generate an optimized welding path according to the smoothed trajectory data, and extract the feature distribution data of the optimized welding path; If the feature distribution data does not match the preset feature threshold, then based on the least squares method, adjust the optimized welding path to obtain the adjusted path parameters; Generate a candidate control instruction set according to the adjusted path parameters, and optimize the smoothed trajectory data according to the candidate control instruction set to obtain stable trajectory data; Generate the final welding path according to the stable trajectory data.

[0097] According to the calibration parameters determined above, the embodiments of the present application can correct the attitude data of the welding head of the robot, so that the welding head corresponding to the corrected attitude data is aligned with the actual weld starting point. Exemplarily, in a straight weld with a length of 40 cm, if a lateral offset of 2 mm is detected, the angle of the welding head can be rotated, such as adjusted by 5 degrees, to align it with the actual weld. Such adjustment depends on the accurate input of the calibration parameters and can ensure that the attitude of the welding head is consistent with the actual requirements.

[0098] According to the corrected attitude data, the interpolation method can be used to adjust and smooth the movement trajectory of the welding head to obtain the smoothed trajectory data. Exemplarily, in a curved weld, if the distance between data points is 5 cm, intermediate points can be generated by cubic spline interpolation to change the trajectory from a discrete point set to a continuous curve. For example, when a certain section of the trajectory is adjusted from the starting coordinate of 10 cm to 15 cm, the interpolation can fill in the smooth transition points to avoid sudden changes.

[0099] An optimized welding path can be generated according to the smoothed trajectory data, and the feature distribution data of the optimized welding path can be extracted by using the second convolutional neural network model. Among them, the curvature change and key turning points can be extracted as the feature distribution data. For example, in a weld, an arc feature with a radius of 3 cm is detected, which can be used as the feature distribution data. The feature distribution data extracted here can provide a basis for subsequent optimization.

[0100] The preset feature threshold can be determined based on the theoretical weld path, and different feature distribution data can correspond to different preset feature thresholds. The mismatch between the feature distribution data and the preset feature threshold can be understood as the difference between the feature distribution data and the preset feature threshold being greater than a certain value. If the feature distribution data does not match the preset feature threshold, then based on the least squares method, the optimized welding path is adjusted to obtain the adjusted path parameters, so that the adjusted path parameters are consistent with the theoretical weld path or have a small difference. Exemplarily, in a 30-cm-long weld, if a certain segment of feature points in the feature distribution data deviates by more than 2 mm compared to the corresponding preset feature threshold, then the actual points (optimized welding path) and the theoretical weld path can be fitted by the least squares method to obtain the adjustment parameters, and the optimized welding path is adjusted by the adjustment parameters to obtain the adjusted path parameters, and the deviation of the adjusted path parameters can be reduced to within 0.5 mm. This method improves the matching degree of the optimized welding path through data optimization.

[0101] According to the adjusted path parameters, a candidate control instruction set for the robot can be generated. The candidate control instruction set can be a control instruction set obtained by optimizing the initial control instruction set according to the adjusted path parameters. Exemplarily, in a complex weld, the initial control instruction set may contain 100 control instructions, and 120 control instructions are generated through parameter optimization after adjustment. The instruction distribution is finer and the adaptability is stronger. The update of this instruction set directly affects the welding accuracy.

[0102] When optimizing and smoothing the trajectory data according to the candidate control instruction set, the filtering method can further stabilize the data, thereby obtaining stable trajectory data. Exemplarily, on a 50-cm-long weld, the smoothed trajectory data is processed by Gaussian filtering to remove the 0.3-mm fluctuation caused by noise, making the trajectory smoother. This processing provides a reliable basis for subsequent execution. Finally, the final welding path can be generated according to the stable trajectory data.

[0103] It should be noted that if the distance between certain trajectory points suddenly increases from 1 mm to 4 mm, it can be judged as a potential break point, and intermediate points need to be supplemented. For example, in a 25-cm-long weld, the integrity of the final welding path is significantly improved after supplementing the points, providing a stable reference for the welding process.

[0104] In the embodiments of the present application, the gradual advancement of these steps, from attitude adjustment to final path generation, forms a rigorous logical chain. The optimization of each link provides support for the next step. For example, smoothing the trajectory reduces the risk of jitter of the welding head, while feature extraction enhances the adaptability of the path. This method can effectively improve the stability and consistency of the welding process in practical applications, especially in complex trajectory scenarios.

[0105] In a possible implementation, if the surface conditions of the weld seam are variable, multiple sets of trajectory data can be combined to enhance stability. For example, by collecting the trajectory distribution three times continuously and taking the average value to generate the final path. This extended solution further enriches the robustness of the solution and ensures its applicability in a dynamic environment.

[0106] In some embodiments, in S105, according to the final welding path and the material and thickness of the workpiece, determine the final control instruction set, including: Extract the key point coordinates in the final welding path and determine the coordinate feature data of the key point coordinates; Based on a preset process parameter database, according to the coordinate feature data and the material and thickness of the workpiece, determine the initial process parameters; Filter the initial process parameters to obtain the smoothed process parameters; If the smoothed process parameters exceed the preset process parameter threshold, correct the smoothed process parameters to obtain the corrected process parameters; Determine the stability characteristics of the corrected process parameters, and according to the stability characteristics and the key point coordinates, adjust the final welding path to obtain the adjusted welding path; Generate the final control instruction set according to the adjusted welding path and the corrected process parameters.

[0107] When extracting the key point coordinates in the final welding path, it can be achieved by analyzing the geometric characteristics of the final welding path. Exemplarily, in a straight weld seam with a length of 30 cm, assuming a key point coordinate is set every 5 cm, 6 key point coordinates can be extracted. The distribution of these key point coordinates reflects the linear characteristics of the final welding path. If the final welding path is a curve, the turning points can be selected according to the curvature change. For example, for an arc region with a radius of 4 cm, its vertex coordinates can be used as the key point coordinates.

[0108] When determining the coordinate feature data of the key point coordinates, statistical methods can be used to calculate data such as point spacing and angle change, and then determine the coordinate feature data. The coordinate feature data can be stable or unstable, etc. Exemplarily, in a 20-cm-long weld seam, if the average point spacing is 3 cm and the angle change is less than 10 degrees, it indicates a uniform distribution, and the coordinate feature data can be marked as stable, otherwise it is unstable.

[0109] In the preset process parameter database, the corresponding relationships among coordinate feature data, workpiece material, workpiece thickness, and the process parameters of the robot can be stored. Therefore, based on the preset process parameter database, the coordinate feature data of the key point coordinates and the process parameters corresponding to the material and thickness of the workpiece can be queried as the initial process parameters. The initial process parameters can include current parameters, speed parameters, etc. Exemplarily, if the coordinate feature data is stable, the material is stainless steel, and the thickness is 3 mm, the preset process parameter database can return the corresponding initial process parameters. For example, the initial process parameters can be a current range of 100 - 150 amperes, a speed range of 20 - 30 cm / minute, etc. For steel with a thickness increased to 5 mm, the database may suggest increasing the current to 180 amperes and decreasing the speed to 15 cm / minute to meet the heat input requirements.

[0110] Filter the initial process parameters to process data fluctuations, and smoothed process parameters can be obtained. For example, in a 40 - cm long weld seam, if a certain section of current data jumps between 120 - 130 amperes, it can be stabilized at 125 amperes after filtering, reducing the instability during the robot's execution.

[0111] The preset process parameter threshold can be the upper limit value of the process parameter or the normal range of the process parameter. If the smoothed process parameter exceeds the preset process parameter threshold, the smoothed process parameter is corrected to obtain the corrected process parameter, and the corrected process parameter does not exceed the preset process parameter threshold. Among them, the least - squares method can be used to correct the smoothed process parameter. Exemplarily, assuming the current upper limit is 140 amperes, if the current process parameter exceeds 140 amperes, it needs to be corrected to be less than 140 amperes; assuming the speed is 32 cm / minute and the corresponding threshold is 28 cm / minute, it is adjusted to 26 cm / minute through the least - squares method to ensure that the parameter is reasonable and does not exceed the limit.

[0112] To determine the stability characteristics of the corrected process parameters, the continuity of the data can be concerned. The stability characteristics can be high stability, low stability, instability, etc. Exemplarily, in a 25 - cm long weld seam, if the current distribution changes smoothly from beginning to end, the stability characteristic can be marked as high stability; if the number of non - smooth points in the current distribution is greater than 0 and less than the second preset quantity, the stability characteristic can be low stability; if the number of non - smooth points in the current distribution is more than the second preset quantity, the stability characteristic can be instability. The second preset quantity is greater than 0 and can be set according to actual needs without specific limitation here. For example, it can be 2, 3, etc. The stability characteristics can also be determined according to the speed distribution.

[0113] For the stability characteristics of low stability and instability, the final welding path can be adjusted based on the key point coordinates to obtain an adjusted welding path, so that the stability characteristic finally determined based on the adjusted welding path is high stability.

[0114] According to the adjusted welding path and the corrected process parameters determined through the adjusted welding path, a final control instruction set can be generated. The instruction density in the final control instruction set can be relatively large to refine the control of the robot. For example, if there was originally one control instruction every 5 centimeters, it can be refined to one control instruction every 3 centimeters. The corrected process parameters can be bound to the adjusted welding path in the final control instruction set, and each control instruction can include the corresponding path coordinates and process parameters.

[0115] The control instructions in the final control instruction set can be arranged in chronological order. Exemplarily, in a 50 - centimeter - long weld seam, the final control instruction set may include 150 control instructions, covering all actions from the starting point to the ending point, to obtain stable execution data. This method forms a complete process from key point coordinate extraction to parameter adjustment and then to instruction generation, ensuring the reliability and adaptability of the welding process.

[0116] In some possible implementation manners, controlling the robot to perform automatic welding according to the final control instruction set may include: generating a robot execution program according to the final control instruction set, and transmitting the program to the corresponding robot through a general interface protocol, so as to control the robot to perform automatic welding according to the program.

[0117] Exemplarily, the robot execution program is a program recognizable by the robot, and a robot execution program including start and end points, process parameters, etc. can be generated through a preset template.

[0118] Through a general interface protocol, such as TCP / IP, etc., the program transmission is completed, and the program can be transmitted to the corresponding robot in the form of data packets. It should be noted that the protocol needs to ensure that there is no packet loss during transmission, and the robot returns a confirmation signal after receiving, such as a "data complete" status code.

[0119] In some embodiments, in the above - mentioned S105, after controlling the robot to perform automatic welding, it further includes: Obtaining the feedback data of the robot; Based on the feedback data, determining whether the welding quality of the robot is qualified; If the welding quality of the robot is unqualified, updating the final control instruction set according to the feedback data, and controlling the robot to perform automatic welding according to the updated final control instruction set.

[0120] During the automatic welding process, the robot can collect the real-time current and speed values of the robot through sensors as feedback data and feed it back to the control device. For example, during the welding process, a certain piece of feedback data shows that the current fluctuates between 118 and 122 amperes, and the speed is stable at 24 - 26 centimeters per minute.

[0121] In some possible implementation manners, determining whether the welding quality of the robot is qualified based on the feedback data may include: Determine whether there is abnormal fluctuation in the feedback data. If there is abnormal fluctuation, determine the distribution of the fluctuation range; If in the distribution of the fluctuation range, the proportion of abnormal fluctuations exceeds a preset proportion, adjust the feedback data through a filtering algorithm to obtain the smoothed feedback data; Based on the smoothed feedback data, determine whether the welding quality of the robot is qualified.

[0122] Exemplarily, it can be determined whether there is abnormal fluctuation according to whether the fluctuation range of the feedback data exceeds the normal fluctuation range. For example, if the fluctuation range of the current in a certain piece of feedback data exceeds the normal fluctuation range of ±5 amperes, for example, the current is 115 - 125 amperes, it is determined that there is abnormal fluctuation in this piece of feedback data. It can also be determined whether there is abnormal fluctuation through speed fluctuation, which will not be elaborated here.

[0123] If there is abnormal fluctuation, a statistical method can be used to draw the distribution of the fluctuation range. If in the distribution of the fluctuation range, the proportion of abnormal fluctuations exceeds the preset proportion, a filtering algorithm is adopted. For example, the moving average method is used to adjust the feedback data to obtain the smoothed feedback data. The preset proportion can be set according to actual needs. For example, the preset proportion is 8%, etc.

[0124] Exemplarily, if in the distribution of the fluctuation range, the current fluctuations of 90% of the points are within ±5 amperes, and the current fluctuations of 10% of the points exceed ±5 amperes, the moving average method is adopted to adjust the current data with fluctuations exceeding ±5 amperes to within ±5 amperes to improve the consistency of the data stream.

[0125] After that, based on the smoothed feedback data, determine whether the welding quality of the robot is qualified, that is, whether it meets the standard.

[0126] If there is no abnormal data in the feedback data, directly determine whether the welding quality of the robot is qualified based on the feedback data.

[0127] When judging whether the welding quality of the robot is qualified, it is necessary to pay attention to the matching degree of the current and speed, compare it with the speed and current in the preset process parameter database, and determine whether the deviation between the two is within a certain ratio. Exemplarily, if the speed of a certain section is 25 cm / min and the current is stable at 121 A, and the deviation from the reference value in the preset process parameter database is less than 5%, the welding quality can be determined to be qualified; otherwise, the welding quality is determined to be unqualified.

[0128] For the feedback data with unqualified welding quality, it can be marked as to be optimized for subsequent optimization and adjustment.

[0129] If the welding quality of the robot is unqualified, the final control instruction set can be updated according to the feedback data, so that the welding quality is qualified when the updated final control instruction set controls the robot for automatic welding. Specifically, the final control instruction set can be adjusted according to the deviation between the two. For example, the speed can be increased from 25 cm / min to 27 cm / min, etc.

[0130] The execution program corresponding to the updated final control instruction set is sent to the corresponding robot by using the transmission protocol, and the status feedback of "parameters have been updated" is received from the robot.

[0131] The feedback data of the robot regarding the current and speed is obtained again. If there is no abnormal fluctuation in the feedback data, it is determined that the compatibility of the robot is good, and it can always operate stably in the long term with the updated final control instruction set. This method optimizes the parameters through feedback and improves the welding consistency.

[0132] In a possible implementation manner, for complex welds such as arc paths, the feedback data may show uneven local speed. For example, the speed of a certain section drops from 25 cm / min to 22 cm / min. After filtering, smoothing and updating the program, the speed distribution tends to be uniform, reducing the frequent pauses during equipment adjustment. It can be understood that this method enhances the adaptability of the robot to dynamic processes and effectively guarantees the execution accuracy.

[0133] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0134] The following is the device embodiment of the present invention. For the details not described in detail, reference can be made to the corresponding method embodiment above.

[0135] Figure 2 The structural schematic diagram of the industrial robot automatic welding device provided by the embodiment of the present invention is shown. For the convenience of description, only the parts related to the embodiment of the present invention are shown and are described in detail as follows: As Figure 2As shown in the figure, the industrial robot automatic welding device 2 includes: an alignment module 21, a theoretical weld path acquisition module 22, an actual weld image acquisition module 23, a final welding path determination module 24, and a welding module 25.

[0136] Among them, the alignment module 21 is used to obtain the three-dimensional point cloud data of the workpiece, and adjust the pre-constructed BIM model of the workpiece according to the three-dimensional point cloud data, so that the adjusted BIM model is aligned with the actual position of the workpiece; The theoretical weld path acquisition module 22 is used to obtain the theoretical weld path from the adjusted BIM model; The actual weld image acquisition module 23 is used to obtain the actual weld image of the workpiece; The final welding path determination module 24 is used to determine the calibration parameters of the robot according to the actual weld image and the theoretical weld path, and determine the final welding path according to the calibration parameters; The welding module 25 is used to determine the final control instruction set according to the final welding path, the material and thickness of the workpiece, and control the robot to perform automatic welding according to the final control instruction set.

[0137] In a possible implementation manner, the theoretical weld path acquisition module 22 is specifically used for: Extract the weld features from the adjusted BIM model to determine the theoretical weld position information; Adopt a spatial alignment method to correct the theoretical weld position information to obtain the corrected theoretical weld position information; Generate a theoretical initial weld path based on the corrected theoretical weld position information; Perform path integrity verification and optimization, optimization to adapt to the surface undulation of the actual workpiece, and optimization to adapt to the operation requirements of the robot on the theoretical initial weld path to obtain the theoretical weld path.

[0138] In a possible implementation manner, the actual weld image acquisition module 23 is specifically used for: Based on the scanning device, obtain the scanning data stream of the workpiece, and based on the vision system, preprocess the scanning data stream to obtain the preliminary scanning data; Extract the surface features of the workpiece based on the preliminary scanning data, determine the dynamic change trend of the workpiece surface features, and determine the dynamic imaging range including the weld according to the dynamic change trend; Based on the dynamic imaging range and optical features, separate the weld area; Perform image enhancement and noise processing on the weld area to obtain a stable weld image; Extract the weld contour according to the stable weld image to generate an image including the weld contour; Perform contour verification on the image containing the weld contour, adjust the image generation parameters affected by environmental changes, and generate the actual weld image.

[0139] In a possible implementation, in the final welding path determination module 24, determine the calibration parameters of the robot according to the actual weld image and the theoretical weld path, including: Identify the weld edge positions in the actual weld image to obtain edge distribution data, and perform coordinate transformation on the edge distribution data to obtain the actual weld path; Calculate the offset between the actual weld path and the theoretical weld path to obtain the preliminary deviation value; For the actual area where the preliminary deviation value is greater than the preset deviation threshold, perform interpolation adjustment to obtain the adjusted edge distribution data; Based on the adjusted edge distribution data, use the least squares method to determine the calibration parameters of the robot.

[0140] In a possible implementation, in the final welding path determination module 24, determine the final welding path according to the calibration parameters, including: According to the calibration parameters, correct the attitude data of the welding head of the robot to obtain the corrected attitude data; According to the corrected attitude data, adjust and smooth the movement trajectory to obtain the smoothed trajectory data; Generate an optimized welding path according to the smoothed trajectory data, and extract the feature distribution data of the optimized welding path; If the feature distribution data does not match the preset feature threshold, then based on the least squares method, adjust the optimized welding path to obtain the adjusted path parameters; According to the adjusted path parameters, generate a candidate control instruction set, and optimize the smoothed trajectory data according to the candidate control instruction set to obtain stable trajectory data; Generate the final welding path according to the stable trajectory data.

[0141] In a possible implementation, in the welding module 25, determine the final control instruction set according to the final welding path and the material and thickness of the workpiece, including: Extract the key point coordinates in the final welding path, and determine the coordinate feature data of the key point coordinates; Based on the preset process parameter database, determine the initial process parameters according to the coordinate feature data and the material and thickness of the workpiece; Filter the initial process parameters to obtain the smoothed process parameters; If the smoothed process parameters exceed the preset process parameter threshold, then correct the smoothed process parameters to obtain the corrected process parameters; Determine the stability characteristics of the corrected process parameters, and adjust the final welding path according to the stability characteristics and the key point coordinates to obtain the adjusted welding path; Generate a final control instruction set according to the adjusted welding path and the corrected process parameters.

[0142] In a possible implementation manner, in the alignment module 21, adjust the BIM model of the pre-constructed workpiece according to the three-dimensional point cloud data, including: Determine the optimized point cloud data and the target coordinate system reflecting the actual posture of the workpiece according to the three-dimensional point cloud data; Based on the target coordinate system, determine the deviation data between the optimized point cloud data and the BIM model, and determine the position transformation matrix according to the deviation data; Adjust the BIM model according to the position transformation matrix to obtain the adjusted BIM model.

[0143] In a possible implementation manner, in the alignment module 21, determine the optimized point cloud data and the target coordinate system reflecting the actual posture of the workpiece according to the three-dimensional point cloud data, including: Remove the noise from the three-dimensional point cloud data to obtain the effective point cloud data on the surface of the workpiece; Perform clustering segmentation on the effective point cloud data to determine the position information of the workpiece in the effective point cloud data; Based on the position information of the workpiece, determine the boundary points of the contour features in the effective point cloud data; Based on the boundary points, determine the preliminary coordinate system reflecting the actual posture of the workpiece; Based on the preliminary coordinate system, perform coordinate transformation on the effective point cloud data to obtain the optimized point cloud data; Verify the contour features of the optimized point cloud data, and use the verified preliminary coordinate system as the target coordinate system.

[0144] In a possible implementation manner, in the welding module 25, after controlling the robot to perform automatic welding, it further includes: Obtain the feedback data of the robot; Based on the feedback data, judge whether the welding quality of the robot is qualified; If the welding quality of the robot is unqualified, update the final control instruction set according to the feedback data, and control the robot to perform automatic welding according to the updated final control instruction set.

[0145] An embodiment of the present invention further provides a control device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the industrial robot automatic welding method in the above method embodiment. Exemplarily, the control device may be a controller or the like, which is not limited herein.

[0146] An embodiment of the present invention further provides a robot, including the above control device.

[0147] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the industrial robot automatic welding method in the above method embodiment.

[0148] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For parts not detailed or recorded in a certain embodiment, reference may be made to the relevant descriptions of other embodiments. Without special instructions and logical conflicts, the terms and / or descriptions between different embodiments are consistent and can be mutually referred to. The technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0149] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements for some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention and should all be included in the protection scope of the present invention.

Claims

1. An industrial robot automatic welding method, characterized in that: include: Acquire three-dimensional point cloud data of the workpiece, and adjust a pre-built BIM model of the workpiece according to the three-dimensional point cloud data so that the adjusted BIM model is aligned with the actual workpiece position; Obtaining a theoretical weld path from the adjusted BIM model; Acquire an actual weld image of the workpiece; Determining calibration parameters of the robot according to the actual weld image and the theoretical weld path, and determining a final welding path according to the calibration parameters; A final control instruction set is determined according to the final welding path and the material and thickness of the workpiece, and the robot is controlled to perform automatic welding according to the final control instruction set.

2. The industrial robot automatic welding method according to claim 1, characterized in that: The obtaining of a theoretical weld path from the adjusted BIM model comprises: Extract weld features from the adjusted BIM model to determine theoretical weld position information; The theoretical weld position information is corrected by using a spatial alignment method to obtain corrected theoretical weld position information; Based on the corrected theoretical weld position information, generating a theoretical initial weld path; The theoretical initial weld path is verified and optimized for path integrity, optimized to adapt to the actual surface undulations of the workpiece, and optimized to adapt to the operation requirements of the robot, to obtain a theoretical weld path.

3. The industrial robot automatic welding method according to claim 1, characterized in that: The step of obtaining an actual weld image of the workpiece includes: Based on the scanning device, a scanning data stream of the workpiece is acquired, and based on the visual system, the scanning data stream is preprocessed to obtain preliminary scanning data; Extracting workpiece surface features based on the preliminary scanning data, determining a dynamic change trend of the workpiece surface features, and determining a dynamic imaging range including a weld according to the dynamic change trend; Separating and obtaining the weld area based on the dynamic imaging range and the optical characteristics; Performing image enhancement and noise processing on the weld area to obtain a stable weld image; Extracting the weld contour according to the stable weld image to generate an image containing the weld contour; The image containing the weld contour is subjected to contour verification, and image generation parameters affected by environmental changes are adjusted to generate an actual weld image.

4. The industrial robot automatic welding method according to claim 1, characterized in that: Determining the calibration parameters of the robot according to the actual weld image and the theoretical weld path includes: Identify the weld edge position in the actual weld image to obtain edge distribution data, and perform coordinate transformation on the edge distribution data to obtain an actual weld path; Calculating the offset between the actual weld path and the theoretical weld path to obtain a preliminary deviation value; For actual areas where the preliminary deviation value is greater than the preset deviation threshold, interpolation adjustment is performed to obtain adjusted edge distribution data; Based on the adjusted marginal distribution data, the calibration parameters of the robot are determined using the least squares method.

5. The industrial robot automatic welding method according to claim 1, characterized in that: Determining the final welding path according to the calibration parameters includes: According to the calibration parameters, the posture data of the welding head of the robot is corrected to obtain corrected posture data; According to the corrected posture data, the moving trajectory is adjusted and smoothed to obtain smoothed trajectory data; generating an optimized welding path according to the smoothed trajectory data, and extracting characteristic distribution data of the optimized welding path; If the characteristic distribution data does not match the preset characteristic threshold, adjusting the optimized welding path based on the least square method to obtain an adjusted path parameter; generating a candidate control instruction set according to the adjusted path parameters, and optimizing the smoothed trajectory data according to the candidate control instruction set to obtain stable trajectory data; A final welding path is generated according to the stable trajectory data.

6. The industrial robot automatic welding method according to claim 1, characterized in that: Determining a final control instruction set according to the final welding path and the material and thickness of the workpiece includes: Extracting the coordinates of key points in the final welding path, and determining the coordinate feature data of the key point coordinates; Based on a preset process parameter database, determining initial process parameters according to the coordinate feature data and the material and thickness of the workpiece; Filtering the initial process parameters to obtain smoothed process parameters; If the smoothed process parameter exceeds a preset process parameter threshold, the smoothed process parameter is corrected to obtain a corrected process parameter; Determining the stability characteristics of the corrected process parameters, and adjusting the final welding path according to the stability characteristics and the key point coordinates to obtain an adjusted welding path; A final control instruction set is generated according to the adjusted welding path and the corrected process parameters.

7. The industrial robot automatic welding method according to claim 1, characterized in that: The adjusting the pre-built BIM model of the workpiece according to the three-dimensional point cloud data includes: Determining optimized point cloud data and a target coordinate system reflecting an actual posture of the workpiece according to the three-dimensional point cloud data; Based on the target coordinate system, determining deviation data between the optimized point cloud data and the BIM model, and determining a position transformation matrix according to the deviation data; The BIM model is adjusted according to the position transformation matrix to obtain an adjusted BIM model.

8. The industrial robot automatic welding method according to claim 7, characterized in that: Determining the optimized point cloud data and a target coordinate system reflecting the actual posture of the workpiece according to the three-dimensional point cloud data includes: removing noise from the three-dimensional point cloud data to obtain valid point cloud data of the workpiece surface; Performing clustering and segmentation on the valid point cloud data to determine the position information of the workpiece in the valid point cloud data; Based on the position information of the workpiece, determining the boundary points of the contour features in the valid point cloud data; Based on the boundary points, determining a preliminary coordinate system reflecting the actual posture of the workpiece; Based on the preliminary coordinate system, coordinate transformation is performed on the effective point cloud data to obtain optimized point cloud data; The optimized point cloud data is subjected to contour feature verification, and the verified preliminary coordinate system is used as the target coordinate system.

9. The industrial robot automatic welding method according to any one of claims 1 to 8, characterized in that: After the controlling robot performs automatic welding, the method further comprises: Obtaining feedback data of the robot; Based on the feedback data, judging whether the welding quality of the robot is qualified; If the welding quality of the robot is unqualified, the final control instruction set is updated according to the feedback data, and the robot is controlled to perform automatic welding according to the updated final control instruction set.

10. A control device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the industrial robot automatic welding method according to any one of claims 1 to 9 when executing the computer program.

Images