Method for generating workpiece model based on point cloud data

Abstract

The invention discloses a method for generating a workpiece model based on point cloud data, and aims to automatically generate a high-precision workpiece model through reverse modeling. The method comprises the following steps: dynamically adjusting a data acquisition amount through a line structure light sensor, and obtaining workpiece surface point cloud data; preprocessing the point cloud, including three-dimensional effective area cutting, uniform downsampling and radius filtering to remove interference data and reduce data volume; carrying out clustering segmentation on the preprocessed data by utilizing a point cloud region growing method, separating a bottom surface part from a non-bottom surface part, extracting contour points of each plane region, calculating normal lines, generating vertexes by fitting contour edges and solving intersection points, and forcibly closing unclosed contours to construct complete geometric features; and finally, generating a workpiece model by combining the point, line and surface information packaged by the open source library. According to the method, the low-error and high-precision model can be directly generated for the model-free non-standard workpiece, limitation of manual modeling or model presetting is avoided, and programming time consumption is remarkably reduced.

Description

Technical Field

[0001] The present application relates to the field of workpiece model construction, and in particular to a method for generating a workpiece model based on point cloud data. Background Art

[0002] In the field of robotic welding, the two main methods currently used are teaching welding and scanning welding. Teaching welding requires manual pre-setting of the welding path, which is suitable for the assembly line production of standardized workpieces. However, when faced with non-standard workpieces, the path setting is time-consuming and lacks flexibility. Scanning welding uses sensors to obtain the weld position and generate a path. Although it reduces manual intervention, it is highly dependent on the existing three-dimensional model of the workpiece. However, non-standard workpieces often lack ready-made models, and the existing models are mostly integral structures that cannot adapt to local welding needs. In existing technologies, reliance on manual programming or limited models leads to low welding efficiency and poor adaptability, making it difficult to meet the intelligent welding needs of complex non-standard workpieces. Summary of the Invention

[0003] The purpose of this application is to provide a method for generating a workpiece model based on point cloud data, and to automatically generate a high-precision workpiece model through reverse modeling.

[0004] To solve the above technical problems, the present application provides a method for generating a workpiece model based on point cloud data, comprising:

[0005] Use visual sensors to collect point cloud data on the workpiece surface;

[0006] Preprocessing the point cloud data;

[0007] Extract point, line and surface information from preprocessed point cloud data;

[0008] Based on the extracted point, line, and surface information, the workpiece model is generated using the open source library.

[0009] Preferably, the point cloud data is preprocessed, including: performing data cropping, downsampling and filtering on the point cloud data to remove interference data and reduce the data volume.

[0010] Preferably, the point cloud data is preprocessed, including:

[0011] Step S201: Set the valid area of the workpiece point cloud [Range X ,Range Y ,Range Z ], where Range X ∈[X min ,X max ] indicates the effective range of the X axis, Range Y ∈[Y min ,Y max] indicates the effective range of the Y axis, Range Z ∈[Z min ,Z max ] indicates the effective range on the Z axis. X ,Range Y ,Range Z ]Crop the point cloud data;

[0012] Step S202: uniformly downsampling the cropped point cloud data to reduce the data volume while ensuring that key workpiece information is not lost;

[0013] Step S203: performing radius filtering on the downsampled point cloud data to filter out outlier noise points.

[0014] Preferably, extracting point, line, and surface information from the pre-processed point cloud data includes:

[0015] Step S301: performing cluster segmentation on the point cloud data based on the point cloud region growing method to separate the point cloud data of the bottom surface part and the point cloud data of the non-bottom surface part;

[0016] Step S302: performing secondary clustering segmentation on the point cloud data of the bottom surface, extracting contour points of each plane area of the bottom surface after the secondary clustering segmentation, and calculating the normal of each plane;

[0017] Step S303: performing contour edge fitting on the contour points of each region of the bottom plane, determining the starting point and end point of each contour edge, and generating vertices by calculating the intersection points between the contour edges;

[0018] Step S304: performing secondary clustering segmentation on the point cloud data of the non-bottom surface portion, extracting the contour lines of each area of the non-bottom surface portion after the secondary clustering segmentation, fitting each contour line to find the starting point and the end point, and finally obtaining the intersection points of the fitted contour lines, and using the obtained intersection points as the vertices of each contour;

[0019] Step S305: Construct complete point, line, and surface geometric features based on the contour edge, vertex, and normal information.

[0020] Preferably, based on the contour edge, vertex and normal information, complete point, line and surface geometric features are constructed, including: forcing an unclosed contour to be closed by extending the contour edge.

[0021] Preferably, the visual sensor is a line structured light sensor, and the data collection amount of the line structured light sensor is dynamically adjusted according to the length of the workpiece.

[0022] Preferably, generating the workpiece model using the open source library based on the extracted point, line and surface information includes: encapsulating the extracted point, line and surface information in the form of the Open CASCADE Technology open source library and generating the corresponding workpiece model.

[0023] Preferably, after the workpiece model is generated, a scanning path of the weld is generated according to the workpiece model.

[0024] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0025] 1. The method for generating a workpiece model based on point cloud data provided in this application can automatically generate a high-precision workpiece model through reverse modeling, thereby assisting in the intelligent planning of subsequent welding paths, thereby solving the problems of non-standard workpieces having no model dependence and excessive manual intervention;

[0026] 2. Use line structured light sensors to dynamically collect point cloud data on the workpiece surface. Combined with 3D effective area cropping, downsampling, and radius filtering, interference data is removed while retaining key features. This allows for direct generation of low-error, high-precision models for model-free, non-standard workpieces, eliminating the limitations of manual modeling or preset models.

[0027] 3. Based on the workpiece model generated by reverse modeling, the bottom surface and non-bottom surface data are separated by the point cloud region growing method, contour points are extracted and contour edges are fitted. Combined with the Open CASCADE Technology open source library, the corresponding workpiece model is automatically generated to reduce programming time. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0029] Figure 1 is a flow chart of a method for generating a workpiece model based on point cloud data provided by the present application;

[0030] Figure 2 is a schematic diagram of the entire point cloud data collected and provided in this application;

[0031] Figure 3 is a schematic diagram of the cropped point cloud data provided in this application;

[0032] Figure 4 is a schematic diagram of point cloud data after clustering and segmentation provided by this application;

[0033] Figure 5is a schematic diagram of the outline of the bottom plane cluster provided by this application;

[0034] Figure 6 is a schematic diagram of the vertices of the bottom plane outline provided by this application;

[0035] Figure 7 is a schematic diagram of the vertices of the non-bottom plane contour provided by this application;

[0036] Figure 8 It is a schematic diagram of the generated workpiece model provided by this application. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described in this application are only some embodiments of this application, not all embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0038] Please refer to Figure 1 , a method for generating a workpiece model based on point cloud data, comprising:

[0039] Use visual sensors to collect point cloud data on the workpiece surface. Please refer to Figure 2 , Figure 2 It is a schematic diagram of the entire point cloud data collected;

[0040] Preprocessing the point cloud data;

[0041] Extract point, line and surface information from preprocessed point cloud data;

[0042] Based on the extracted point, line, and surface information, the workpiece model is generated using the open source library.

[0043] Specifically, the point cloud data is preprocessed, including: performing data cropping, downsampling and filtering on the point cloud data to remove interference data and reduce the data volume.

[0044] Specifically, the point cloud data is preprocessed, including:

[0045] Step S201: Set the valid area of the workpiece point cloud [Range X ,Range Y ,Range Z ], where Range X ∈[X min ,X max ] indicates the effective range of the X axis, Range Y ∈[Y min ,Ymax ] indicates the effective range of the Y axis, Range Z ∈[Z min ,Z max ] indicates the effective range on the Z axis. X ,Range Y ,Range Z ] To crop point cloud data, please refer to Figure 3 , Figure 3 It is a schematic diagram of the cropped point cloud data;

[0046] Step S202: uniformly downsampling the cropped point cloud data. The purpose of downsampling is to effectively reduce the data volume without losing key workpiece information. In this embodiment, the sampling radius is set to 3 mm.

[0047] Step S203: Radius filtering is performed on the downsampled point cloud data to filter out obvious outlier noise points. The filter radius set in this embodiment is 7 mm. The value of the filter radius is set based on the actual trigger period, scanning speed and downsampling radius of the visual sensor in the previous step.

[0048] Specifically, point, line, and surface information are extracted from the preprocessed point cloud data, including:

[0049] Step S301: Clustering and segmenting the point cloud data based on the point cloud region growing method to separate the point cloud data of the bottom surface and the point cloud data of the non-bottom surface. Figure 4 , Figure 4 It is a schematic diagram of point cloud data after clustering and segmentation;

[0050] Step S302: Perform secondary clustering segmentation on the point cloud data of the bottom surface, extract the contour points of each plane area of the bottom surface after secondary clustering segmentation, and calculate the normal of each plane. The purpose of calculating the plane normal is to provide a direction reference and geometric constraint for subsequent contour fitting, vertex generation, and model construction, thereby ensuring the high accuracy and structural rationality of the generated workpiece model. Please refer to Figure 5 , Figure 5 is a schematic diagram of the outline of the bottom plane cluster;

[0051] Step S303: Fit the contour points of each area of the bottom plane to the contour edge, determine the starting point and end point of each contour edge, and generate vertices by calculating the intersection points between the contour edges. Figure 6 , Figure 6 is a schematic diagram of the vertices of the bottom plane contour;

[0052] Step S304: Perform secondary clustering segmentation on the point cloud data of the non-bottom part, extract the contour lines of each area of the non-bottom part after secondary clustering segmentation, and fit each contour line to find the starting point and end point. Finally, find the intersection point of the fitted contour line, and use the intersection point as the vertex of each contour, as shown in the figure. Figure 7 As shown, Figure 7 It is a schematic diagram of the vertices of the non-bottom plane contour;

[0053] Step S305: Construct complete point, line, and surface geometric features based on the contour edge, vertex, and normal information.

[0054] Specifically, based on the contour edge, vertex and normal information, complete point, line and surface geometric features are constructed, including: forcing the unclosed contour to be closed by extending the contour edge.

[0055] Specifically, the visual sensor is a line structured light sensor, and the data collection amount of the line structured light sensor is dynamically adjusted according to the length of the workpiece.

[0056] Specifically, based on the extracted point, line, and surface information, generating the workpiece model using the open source library includes: based on the extracted point, line, and surface information, encapsulating them in the form of the Open CASCADE Technology open source library and generating the corresponding workpiece model, such as Figure 8 As shown, Figure 8 This is a schematic diagram of the generated workpiece model. Open CASCADE Technology is a powerful open-source geometric modeling library designed specifically for engineering software development and widely used in CAD, CAM, CAE, and other fields. OpenCASCADE Technology is free and open-source, offering high-precision modeling capabilities that can adapt to complex industrial needs.

[0057] Specifically, after the workpiece model is generated, a scanning path of the weld is generated according to the workpiece model.

[0058] It should also be noted that, in this specification, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

Claims

1. A method for generating a workpiece model based on point cloud data, characterized in that: include: Use visual sensors to collect point cloud data on the workpiece surface; Preprocessing the point cloud data; Extract point, line and surface information from preprocessed point cloud data; Based on the extracted point, line, and surface information, the workpiece model is generated using the open source library.

2. The method for generating a workpiece model based on point cloud data according to claim 1, characterized in that: The point cloud data is preprocessed, including: data cropping, downsampling and filtering the point cloud data to remove interference data and reduce data volume.

3. The method for generating a workpiece model based on point cloud data according to claim 1, characterized in that: Preprocessing the point cloud data includes: Step S201: Set the valid area of the workpiece point cloud [Range X ,Range Y ,Range Z ], where Range X ∈[X min ,X max ] indicates the effective range of the X axis, Range Y ∈[Y min ,Y max ] indicates the effective range of the Y axis, Range Z ∈[Z min ,Z max ] indicates the effective range on the Z axis. X ,Range Y ,Range Z ]Crop the point cloud data; Step S202: uniformly downsampling the cropped point cloud data to reduce the data volume while ensuring that key workpiece information is not lost; Step S203: performing radius filtering on the downsampled point cloud data to filter out outlier noise points.

4. The method for generating a workpiece model based on point cloud data according to claim 1, characterized in that: Extract point, line, and surface information from preprocessed point cloud data, including: Step S301: performing cluster segmentation on the point cloud data based on the point cloud region growing method to separate the point cloud data of the bottom surface part and the point cloud data of the non-bottom surface part; Step S302: performing secondary clustering segmentation on the point cloud data of the bottom surface, extracting contour points of each plane area of the bottom surface after the secondary clustering segmentation, and calculating the normal of each plane; Step S303: performing contour edge fitting on the contour points of each region of the bottom plane, determining the starting point and end point of each contour edge, and generating vertices by calculating the intersection points between the contour edges; Step S304: performing secondary clustering segmentation on the point cloud data of the non-bottom surface portion, extracting the contour lines of each area of the non-bottom surface portion after the secondary clustering segmentation, fitting each contour line to find the starting point and the end point, and finally obtaining the intersection points of the fitted contour lines, and using the obtained intersection points as the vertices of each contour; Step S305: Construct complete point, line, and surface geometric features based on the contour edge, vertex, and normal information.

5. The method for generating a workpiece model based on point cloud data according to claim 4, characterized in that: Based on the contour edge, vertex and normal information, complete point, line and surface geometric features are constructed, including: forcing an unclosed contour to be closed by extending the contour edge.

6. The method for generating a workpiece model based on point cloud data according to claim 1, characterized in that: The visual sensor is a line structured light sensor, and the data collection amount of the line structured light sensor is dynamically adjusted according to the length of the workpiece.

7. The method for generating a workpiece model based on point cloud data according to claim 1, characterized in that: Generating a workpiece model based on the extracted point, line, and surface information using an open source library includes: based on the extracted point, line, and surface information, encapsulating the extracted point, line, and surface information in the form of an OpenCASCADE Technology open source library and generating a corresponding workpiece model.

8. The method for generating a workpiece model based on point cloud data according to any one of claims 1 to 7, characterized in that: Also includes: After the workpiece model is generated, a scanning path of the weld is generated according to the workpiece model.

Images