A fastening functional parts assembly method and system based on intelligent identification

Abstract

The present invention discloses a method and system for assembling fastening functional parts based on intelligent identification, aiming to improve assembly accuracy and efficiency. The present invention comprises the following steps: acquiring laser scanning data of each component to construct a three-dimensional model, and establishing a component identification system; extracting and adjusting the assembly position and direction of the component based on the assembly process and process data, while controlling the assembly force and generating an assembly plan; performing assembly operations according to the plan, and performing quality assessment on the assembled functional parts; if the assessment result is lower than expected, calibrating the assembly parameters. The present invention effectively improves the assembly accuracy, efficiency and quality stability.

Description

Technical Field

[0001] The present invention relates to the technical field of assembly control, and in particular to an assembly method and system of fastening functional parts based on intelligent identification. Background Art

[0002] Fastening functional parts are binding components in outdoor sportswear. The quality of their assembly directly affects the overall performance and service life, and in severe cases may even endanger outdoor safety. Traditional fastening functional parts assembly methods mainly rely on manual operation and experience judgment, which are easily affected by human factors, resulting in insufficient assembly accuracy and low efficiency. In addition, the complex assembly process has high technical requirements for operators, and is prone to problems such as uneven assembly force and assembly position deviation, which affects the final assembly quality of fastening functional parts.

[0003] With the development of industrial automation and intelligent manufacturing technology, it has become a trend to use advanced intelligent recognition and control technology to improve the assembly accuracy and efficiency of fastening functional parts. In the existing technology, although some systems have adopted visual recognition and mechanical automated assembly technology, their recognition accuracy and assembly flexibility are still insufficient, especially when facing components with complex shapes or large differences in surface features, it is easy to make recognition errors or inadequate assembly. In addition, post-assembly quality assessment mostly relies on post-sampling inspection, which lacks real-time and comprehensiveness.

[0004] Therefore, there is an urgent need for a fastening functional parts assembly method based on intelligent identification, which can identify and locate components with high precision during the assembly process, and adjust assembly parameters in real time through intelligent control means, thereby improving assembly accuracy and quality. At the same time, through real-time quality evaluation and feedback mechanisms, problems in the assembly process can be corrected in a timely manner to ensure that the final assembly quality meets the expected standards. Summary of the invention

[0005] In order to solve at least one of the above technical problems, the present invention proposes a method and system for assembling fastening functional parts based on intelligent identification.

[0006] A first aspect of the present invention provides a method for assembling a fastening functional part based on intelligent identification, comprising:

[0007] Acquire laser scanning data of each component of the fastening functional part to build a three-dimensional model of the component, and build a component recognition system based on the three-dimensional model of the component;

[0008] Acquire assembly process data and assembly process data of the fastening functional part, extract components according to the assembly process data and the component identification system, adjust the component assembly position and assembly direction, control the assembly force of each component according to the assembly process data, and obtain the assembly plan of the fastening functional part;

[0009] Performing an assembly operation on the fastening functional component assembly according to the assembly scheme, and performing an assembly quality assessment on the fastening functional component after the assembly is completed to obtain an assembly quality assessment result;

[0010] According to the assembly quality evaluation result, if the assembly quality of the fastening functional component is lower than the expected quality, the assembly parameters of the fastening functional component are calibrated.

[0011] In this solution, the laser scanning data of each component of the fastening functional part is obtained to build a component three-dimensional model, and a component identification system is built according to the component three-dimensional model, specifically:

[0012] Obtaining a standard production sample of each component of the fastening functional part, and obtaining point cloud data of the standard production sample of each component based on a laser scanner;

[0013] The outliers and noise points of the point cloud data are removed according to the statistical filtering method to obtain the preprocessed point cloud data;

[0014] Projecting the preprocessed point cloud data into a three-dimensional grid, gridding the preprocessed point cloud data using an octree space partitioning structure, selecting the centroid of all points in each grid cell as a representative point in each grid cell, integrating the representative points of each grid cell to obtain compressed point cloud data;

[0015] Constructing a three-dimensional model of the surface of each component based on a triangular mesh reconstruction algorithm and the compressed point cloud data, and performing a surface fitting operation on the three-dimensional model;

[0016] Calculating the Gaussian curvature of the surface of the three-dimensional model of each component, extracting the surface features of each component according to the Gaussian curvature, and obtaining surface feature data of each component;

[0017] A component recognition system is constructed based on a convolutional neural network, and the surface feature data of each component is imported into the component recognition system for learning and training to obtain a component recognition system with component recognition capability.

[0018] In this solution, the Gaussian curvature of the surface of the three-dimensional model of each component is calculated, and the surface features of each component are extracted according to the Gaussian curvature to obtain the surface feature data of each component, which is specifically:

[0019] For each component's three-dimensional model, extract a triangular mesh of the three-dimensional model surface, wherein the triangular mesh is composed of a vertex set V, an edge set E, and a triangular facet F;

[0020] Performing Gaussian curvature calculation on each triangular mesh vertex in the three-dimensional model, and marking the triangular mesh vertex entering the Gaussian curvature calculation as a triangular mesh vertex in a calculation state;

[0021] Determine adjacent triangle patches to a vertex of a triangular mesh in a computing state, wherein the adjacent triangle patches share the vertex of the triangular mesh in a computing state, determine adjacent vertices to the vertex of the triangular mesh in a computing state according to the adjacent triangle patches, and obtain a vertex neighborhood of the vertex of the triangular mesh in a computing state, wherein the adjacent vertices are directly connected to the vertex of the triangular mesh in a computing state through a triangle edge;

[0022] Determine the coordinates of each adjacent vertex in the vertex neighborhood according to the point cloud data, calculate each internal angle of the adjacent triangular facets based on the cosine theorem, calculate the sum of the internal angles of all triangular facets, and calculate the angle defect of the triangular mesh vertex in the calculated state according to the sum of the internal angles;

[0023] Calculate the area of ​​the adjacent triangular facets, take a preset equal-division area from each adjacent triangular facet and add them up to obtain the associated local area area of ​​the triangular mesh vertex in the calculated state, divide the angle defect by the associated local area area to obtain the Gaussian curvature of the triangular mesh vertex in the calculated state, integrate and visualize the Gaussian curvatures of all triangular mesh vertices to obtain a Gaussian curvature distribution map of the three-dimensional model;

[0024] The surface features of each component are determined according to the Gaussian curvature distribution diagram, wherein the surface features include feature points and geometric feature edges. The surface features are depicted to obtain surface feature data of each component.

[0025] In this solution, the assembly process data and assembly process data of the fastening functional parts are obtained, and components are extracted according to the assembly process data and the component identification system, and the assembly position and assembly direction of the components are adjusted. The assembly force of each component is controlled according to the assembly process data to obtain the assembly solution of the fastening functional parts, which is specifically:

[0026] Acquiring assembly process data and assembly process data of the fastening functional parts;

[0027] Acquire laser scanning data of the top view surface of each component in the component conveyor belt in real time, and determine the surface features of the top view surface of each component according to the laser scanning data;

[0028] Importing the surface features of the top view of each component in the conveyor belt into the component recognition system for feature matching, and determining the degree of matching between the top view features of each component in the conveyor belt and the surface features of the component standard production sample;

[0029] The component standard production samples with a matching degree higher than a preset value are used as the matching result of each component in the conveyor belt to obtain the component recognition result in the conveyor belt;

[0030] Determine the assembly component to be assembled that needs to be assembled in the current assembly process according to the assembly process data of the fastening functional part, and randomly grab one of the assembly components in the conveyor belt according to the component identification result;

[0031] The current grasping angle information of the grasped component to be assembled is obtained, and the current grasping angle information includes vertical angle information and horizontal angle information. The assembly position and assembly direction of the component to be assembled are adjusted according to the assembly process data and the current grasping angle information, and the assembly force of each component is controlled to obtain an assembly plan for the fastening functional parts.

[0032] In this solution, the assembly process data and assembly process data of the fastening functional parts are obtained, components are extracted according to the assembly process data and the component identification system, and the assembly position and assembly direction of the components are adjusted, and the assembly force of each component is controlled according to the assembly process data to obtain the assembly solution of the fastening functional parts, which also includes:

[0033] According to the component recognition result in the conveyor belt, marking a plurality of components to be assembled identified in the conveyor belt to obtain marked components;

[0034] According to the matching degree between the top view feature of each component in the conveyor belt and the surface feature of the component standard production sample, analyzing whether there are multiple components whose surface features of the standard production sample and the top view surface feature of the marked component have a matching degree higher than a preset value;

[0035] If so, the marked components whose surface feature matching degree with multiple component standard production samples is higher than the preset value are marked as pending components for which the component identification results are to be further confirmed, and the other marked components are marked as non-pending components for which the component identification results do not need to be further confirmed;

[0036] When grabbing the components to be assembled, the non-pending components are grabbed and assembled first. When all the components to be assembled identified in the conveyor belt are pending components, the multiple components to which the pending components are identified are determined according to the component identification results, and the top view surface of the pending components is determined in the belonging components.

[0037] According to the surface feature data of each component and the surface features of the top view surface of each component, the feature points of the top view surface of the component to be determined and the surface of each component are connected in pairs to form a feature connection graph;

[0038] The characteristic connection diagram of the undetermined component is compared with the characteristic connection diagram of each of its belonging components, and the belonging components of the undetermined component are secondary identified to obtain a secondary identification result of the undetermined component, and the component identification result in the conveyor belt is corrected according to the secondary identification result.

[0039] In this solution, the fastening functional component assembly is assembled according to the assembly solution, and the assembly quality of the fastening functional component after assembly is evaluated to obtain the assembly quality evaluation result, which is specifically:

[0040] Assembling the fastening functional component assembly according to the assembly scheme, performing mechanical testing on the assembled fastening functional component, evaluating the connection tightness of the component connection parts, determining the structural strength of the component connection parts according to the connection tightness, and obtaining a structural quality evaluation result of the fastening functional component;

[0041] Performing a functional test on the movable parts of the fastening functional parts to obtain the friction between the movable parts, analyzing the degree of jamming between the movable parts according to the friction, and obtaining the functional test result of the fastening functional parts;

[0042] An assembly quality assessment of the fastening functional part is performed according to the structural quality assessment result and the functional test result of the fastening functional part to obtain an assembly quality assessment result.

[0043] In this solution, according to the assembly quality evaluation result, if the assembly quality of the fastening functional part is lower than the expected quality, the assembly parameters of the fastening functional part are calibrated, specifically:

[0044] Acquire preset quality standard data of the fastening functional part, compare the assembly quality evaluation result with the preset quality standard data, and if the assembly quality of the fastening functional part is lower than the preset quality standard, mark the item corresponding to the assembly quality lower than the preset quality standard as the assembly item to be optimized;

[0045] Conducting source tracing analysis on the assembly items to be optimized, determining assembly control parameters that cause quality problems on the assembly items to be optimized, and determining calibration coefficients of the assembly control parameters according to the quality severity of the assembly items to be optimized;

[0046] The assembly parameters of the fastening functional parts are calibrated according to the calibration coefficients, and the calibration operation includes calibration of the position of the robot arm and calibration of the torque parameters of the press-fitting equipment.

[0047] The second aspect of the present invention further provides a fastening functional component assembly system based on intelligent identification, the system comprising: a memory, a processor, the memory comprising a fastening functional component assembly method program based on intelligent identification, the fastening functional component assembly method program based on intelligent identification, when executed by the processor, implements the following steps:

[0048] Acquire laser scanning data of each component of the fastening functional part to build a three-dimensional model of the component, and build a component recognition system based on the three-dimensional model of the component;

[0049] Acquire assembly process data and assembly process data of the fastening functional part, extract components according to the assembly process data and the component identification system, adjust the component assembly position and assembly direction, control the assembly force of each component according to the assembly process data, and obtain the assembly plan of the fastening functional part;

[0050] Performing an assembly operation on the fastening functional component assembly according to the assembly scheme, and performing an assembly quality assessment on the fastening functional component after assembly to obtain an assembly quality assessment result;

[0051] According to the assembly quality evaluation result, if the assembly quality of the fastening functional component is lower than the expected quality, the assembly parameters of the fastening functional component are calibrated.

[0052] The present invention discloses a method and system for assembling fastening functional parts based on intelligent identification, aiming to improve assembly accuracy and efficiency. The present invention comprises the following steps: acquiring laser scanning data of each component to construct a three-dimensional model, and establishing a component identification system; extracting and adjusting the assembly position and direction of the component based on the assembly process and process data, while controlling the assembly force and generating an assembly plan; performing assembly operations according to the plan, and performing quality assessment on the assembled functional parts; if the assessment result is lower than expected, calibrating the assembly parameters. The present invention effectively improves the assembly accuracy, efficiency and quality stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A flow chart of a fastening functional component assembly method based on intelligent identification according to the present invention is shown;

[0054] Figure 2 A flow chart showing the assembly quality evaluation result obtained by the present invention is shown;

[0055] Figure 3 A flow chart showing the calibration operation of the assembly parameters of the fastening functional parts according to the present invention is shown;

[0056] Figure 4 A block diagram of a fastening functional parts assembly system based on intelligent identification of the present invention is shown. DETAILED DESCRIPTION

[0057] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0058] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0059] Figure 1 A flow chart of a method for assembling fastening functional parts based on intelligent identification of the present invention is shown.

[0060] like Figure 1 As shown, the first aspect of the present invention provides a method for assembling fastening functional parts based on intelligent identification, comprising:

[0061] S102, acquiring laser scanning data of each component of the fastening functional part to construct a component three-dimensional model, and constructing a component recognition system according to the component three-dimensional model;

[0062] S104, obtaining assembly process data and assembly process data of the fastening functional part, extracting components according to the assembly process data and the component recognition system, adjusting the component assembly position and assembly direction, controlling the assembly force of each component according to the assembly process data, and obtaining an assembly plan of the fastening functional part;

[0063] S106, performing an assembly operation on the fastening functional component assembly according to the assembly scheme, and performing an assembly quality assessment on the fastening functional component after the assembly is completed to obtain an assembly quality assessment result;

[0064] S108, according to the assembly quality evaluation result, if the assembly quality of the fastening functional component is lower than the expected quality, calibrate the assembly parameters of the fastening functional component.

[0065] It should be noted that by acquiring the three-dimensional point cloud data of components through laser scanning, the complex geometric shapes and subtle features of each component can be accurately captured, thereby generating a high-precision three-dimensional model. The recognition system based on this three-dimensional model can accurately distinguish components with similar shapes or complex structures, and achieve high-precision recognition; the component recognition system can efficiently and accurately identify and extract components to be assembled, avoiding errors in manual screening and judgment, especially when dealing with components with complex shapes or similar sizes, greatly improving the accuracy and efficiency of recognition and extraction, and can automatically adjust the assembly position and direction of the component according to the three-dimensional model data of the component, ensuring that each component is in the optimal position during the assembly process and avoiding assembly deviation. This reduces assembly failures or component damage caused by position and direction errors, controls the assembly force of each component according to the assembly process data, ensures that the force during the assembly process is accurate and moderate, avoids component damage caused by excessive force or looseness caused by insufficient force, and improves the stability of assembly quality; a comprehensive quality assessment of the fastening functional parts after assembly can accurately detect various problems that may occur during the assembly process, including loose connections, insufficient fastening, component misalignment, etc., to ensure that each product meets the expected quality standards. When the assembly quality is lower than expected, the specific assembly parameters that cause the problem can be traced and identified by analyzing the assembly quality assessment results. This precise positioning helps to quickly discover the root cause of the problem, avoid blind adjustments, and improve the efficiency of troubleshooting.

[0066] According to an embodiment of the present invention, the laser scanning data of each component of the fastening functional part is acquired to construct a component three-dimensional model, and a component identification system is constructed according to the component three-dimensional model, specifically:

[0067] Obtaining a standard production sample of each component of the fastening functional part, and obtaining point cloud data of the standard production sample of each component based on a laser scanner;

[0068] The outliers and noise points of the point cloud data are removed according to the statistical filtering method to obtain the preprocessed point cloud data;

[0069] Projecting the preprocessed point cloud data into a three-dimensional grid, gridding the preprocessed point cloud data using an octree space partitioning structure, selecting the centroid of all points in each grid cell as a representative point in each grid cell, integrating the representative points of each grid cell to obtain compressed point cloud data;

[0070] Constructing a three-dimensional model of the surface of each component based on a triangular mesh reconstruction algorithm and the compressed point cloud data, and performing a surface fitting operation on the three-dimensional model;

[0071] Calculating the Gaussian curvature of the surface of the three-dimensional model of each component, extracting the surface features of each component according to the Gaussian curvature, and obtaining surface feature data of each component;

[0072] A component recognition system is constructed based on a convolutional neural network, and the surface feature data of each component is imported into the component recognition system for learning and training to obtain a component recognition system with component recognition capability.

[0073] It should be noted that the compression of point cloud data significantly reduces the amount of data, thereby reducing the demand for computing resources. In the process of point cloud data compression, octree space partitioning and centroid selection technology are used to ensure that while reducing the amount of data, key geometric information and surface features are retained; through surface fitting operations, the surface of the three-dimensional model can be smoothed to eliminate the discreteness and discontinuity in the point cloud data; the Gaussian curvature of the surface of each component three-dimensional model is calculated, so that the system can accurately extract the surface features of the component, including feature points, geometric edges, etc. As a geometric invariant, Gaussian curvature can effectively capture the changes in surface morphology, thereby extracting feature information that is crucial to component identification; by extracting accurate surface feature data and importing it into a convolutional neural network for learning and training, an efficient and accurate component identification system is constructed. Due to the accuracy of the surface features, the recognition system can quickly and accurately classify and identify complex and diverse components, greatly improving the recognition accuracy; the triangular mesh reconstruction algorithm converts the compressed point cloud data into a complete three-dimensional surface model, and the formed three-dimensional surface model consists of multiple triangular meshes, which can accurately describe the complex geometric shape and surface details of the component.

[0074] According to an embodiment of the present invention, the Gaussian curvature of the surface of the three-dimensional model of each component is calculated, and the surface features of each component are extracted according to the Gaussian curvature to obtain the surface feature data of each component, specifically:

[0075] For each component's three-dimensional model, extract a triangular mesh of the three-dimensional model surface, wherein the triangular mesh is composed of a vertex set V, an edge set E, and a triangular facet F;

[0076] Performing Gaussian curvature calculation on each triangular mesh vertex in the three-dimensional model, and marking the triangular mesh vertex entering the Gaussian curvature calculation as a triangular mesh vertex in a calculation state;

[0077] Determine adjacent triangle patches to a vertex of a triangular mesh in a computing state, wherein the adjacent triangle patches share the vertex of the triangular mesh in a computing state, determine adjacent vertices to the vertex of the triangular mesh in a computing state according to the adjacent triangle patches, and obtain a vertex neighborhood of the vertex of the triangular mesh in a computing state, wherein the adjacent vertices are directly connected to the vertex of the triangular mesh in a computing state through a triangle edge;

[0078] Determine the coordinates of each adjacent vertex in the vertex neighborhood according to the point cloud data, calculate each internal angle of the adjacent triangular facets based on the cosine theorem, calculate the sum of the internal angles of all triangular facets, and calculate the angle defect of the triangular mesh vertex in the calculated state according to the sum of the internal angles;

[0079] Calculate the area of ​​the adjacent triangular facets, take a preset equal-division area from each adjacent triangular facet and add them up to obtain the associated local area area of ​​the triangular mesh vertex in the calculated state, divide the angle defect by the associated local area area to obtain the Gaussian curvature of the triangular mesh vertex in the calculated state, integrate and visualize the Gaussian curvatures of all triangular mesh vertices to obtain a Gaussian curvature distribution map of the three-dimensional model;

[0080] The surface features of each component are determined according to the Gaussian curvature distribution diagram, wherein the surface features include feature points and geometric feature edges. The surface features are depicted to obtain surface feature data of each component.

[0081] It should be noted that Gaussian curvature is a geometric quantity that describes the local shape characteristics of a surface. It measures the curvature properties of a surface at a certain point. A three-dimensional model is usually composed of a series of triangular meshes, and the vertices, edges, and patches of the triangular meshes constitute the basic structure of the three-dimensional surface. By calculating the Gaussian curvature, the local geometric characteristics at each triangular mesh vertex can be analyzed. At a triangular mesh vertex, the sum of the internal angles of all adjacent triangles should be close to the angle of a plane (i.e., 360 degrees). An angle that deviates from this value is called an "angle defect." This angle defect is a manifestation of curvature and directly affects the size of the Gaussian curvature. By dividing the angle defect by the area of ​​the local area at the vertex, the Gaussian curvature of the point can be obtained. This curvature describes the degree of curvature of the vertex in three-dimensional space. Points with larger Gaussian curvature usually correspond to key geometric features of the model, such as sharp points, protrusions, or depressions. These points can be used to identify and distinguish different parts of the model. In areas where the Gaussian curvature changes dramatically, there are usually obvious geometric boundaries (such as edges and transition areas). These regions are important structural features of the model, indicating the transition between different surfaces; the angular defect is equal to 2π minus the sum of the internal angles.

[0082] According to an embodiment of the present invention, the assembly process data and assembly process data of the fastening functional parts are obtained, components are extracted according to the assembly process data and the component identification system, and the assembly position and assembly direction of the components are adjusted, and the assembly force of each component is controlled according to the assembly process data to obtain the assembly scheme of the fastening functional parts, specifically:

[0083] Acquiring assembly process data and assembly process data of the fastening functional parts;

[0084] Acquire laser scanning data of the top view surface of each component in the component conveyor belt in real time, and determine the surface features of the top view surface of each component according to the laser scanning data;

[0085] Importing the surface features of the top view of each component in the conveyor belt into the component recognition system for feature matching, and determining the degree of matching between the top view features of each component in the conveyor belt and the surface features of the component standard production sample;

[0086] The component standard production samples with a matching degree higher than a preset value are used as the matching result of each component in the conveyor belt to obtain the component recognition result in the conveyor belt;

[0087] Determine the assembly component to be assembled that needs to be assembled in the current assembly process according to the assembly process data of the fastening functional part, and randomly grab one of the assembly components in the conveyor belt according to the component identification result;

[0088] The current grasping angle information of the grasped component to be assembled is obtained, and the current grasping angle information includes vertical angle information and horizontal angle information. The assembly position and assembly direction of the component to be assembled are adjusted according to the assembly process data and the current grasping angle information, and the assembly force of each component is controlled to obtain an assembly plan for the fastening functional parts.

[0089] It should be noted that the assembly process data includes the assembly sequence of each component of the fastening functional part, and the assembly process data includes the assembly pressing force, insertion force, tightening torque, positional relationship between assembly components, and assembly horizontal and vertical angles of each component; the scenario problem determined by the present invention is that when there are multiple components mixed in the material box, it is necessary to identify the components so that the robot arm has the ability to accurately pick up the components to be assembled from the mixed components. When the components are mixed in the material box, the conveyor belt will convey the materials from the material box with mixed components. When the conveyor belt is conveyed to the preset position, it stops for the robot arm to automatically pick up the identified components to be assembled for assembly. When the materials are mixed, the present invention avoids manual intervention to classify the materials, realizes the full automatic assembly of the fastening functional parts, and reduces the human resource cost; by obtaining the top view laser scanning data of the component in real time and matching it with the surface features of the standard production sample, each component can be accurately identified and ensured. On this basis, the assembly position and direction of the component can be automatically adjusted to ensure that each component is accurately assembled according to the established assembly plan. This precise adjustment and identification mechanism greatly improves the accuracy and consistency of the overall assembly and reduces errors and deviations during the assembly process.

[0090] According to an embodiment of the present invention, the method of obtaining assembly process data and assembly process data of the fastening functional part, extracting components according to the assembly process data and the component identification system, adjusting the component assembly position and assembly direction, controlling the assembly force of each component according to the assembly process data, and obtaining the assembly scheme of the fastening functional part also includes:

[0091] According to the component recognition result in the conveyor belt, marking a plurality of components to be assembled identified in the conveyor belt to obtain marked components;

[0092] According to the matching degree between the top view feature of each component in the conveyor belt and the surface feature of the component standard production sample, analyzing whether there are multiple components whose surface features of the standard production sample and the top view surface feature of the marked component have a matching degree higher than a preset value;

[0093] If so, the marked components whose surface feature matching degree with multiple component standard production samples is higher than the preset value are marked as pending components for which the component identification results are to be further confirmed, and the other marked components are marked as non-pending components for which the component identification results do not need to be further confirmed;

[0094] When grabbing the components to be assembled, the non-pending components are grabbed and assembled first. When all the components to be assembled identified in the conveyor belt are pending components, the multiple components to which the pending components are identified are determined according to the component identification results, and the top view surface of the pending components is determined in the belonging components.

[0095] According to the surface feature data of each component and the surface features of the top view surface of each component, the feature points of the top view surface of the component to be determined and the surface of each component are connected in pairs to form a feature connection graph;

[0096] The characteristic connection diagram of the undetermined component is compared with the characteristic connection diagram of each of its belonging components, and the belonging components of the undetermined component are secondary identified to obtain a secondary identification result of the undetermined component, and the component identification result in the conveyor belt is corrected according to the secondary identification result.

[0097] It should be noted that if there are multiple component standard production samples whose surface features match the surface features of the top view of the marked component at a higher degree than the preset value, it means that the component to be assembled identified in the conveyor belt has surface features higher than the preset value for both component A and component B. This means that there are multiple matching results, and it is necessary to further determine whether the component with multiple matching results is the component to be assembled; for example, assuming that component A is the component to be assembled, there are N components in the conveyor belt whose surface features match the surface features of component A at a higher degree than the preset value, but there may be a component among these N components whose surface features of component B are also higher than the preset value, and there are multiple matching results for the components in the conveyor belt. At this time, it is necessary to further determine which component it is to avoid the robot arm mistakenly grabbing the component for wrong assembly, resulting in assembly errors of the fastening functional parts and causing material loss; the pending component is a component that has multiple matching results and is identified as the component to be assembled in the conveyor belt, while the non-pending component has only one matching result with a matching degree higher than the preset value; by comparing the feature connection diagram, the secondary recognition can more accurately distinguish these similar components to ensure that each component is correctly identified. This greatly improves the accuracy of the overall recognition system and avoids misidentification; by performing secondary recognition on the components to be identified, assembly errors caused by component recognition errors during the assembly process can be effectively reduced. This secondary confirmation mechanism ensures that even if there is ambiguity or uncertainty in the initial recognition process, the system can still correctly identify the component through further analysis, thereby reducing the assembly error rate and improving assembly quality.

[0098] Figure 2 A flow chart of obtaining assembly quality evaluation results according to the present invention is shown.

[0099] According to an embodiment of the present invention, the assembly operation of the fastening functional component assembly is performed according to the assembly scheme, and the assembly quality evaluation of the fastening functional component after the assembly is completed is performed to obtain the assembly quality evaluation result, which is specifically:

[0100] S202, assembling the fastening functional component assembly according to the assembly scheme, performing mechanical testing on the assembled fastening functional component, evaluating the connection tightness of the component connection parts, determining the structural strength of the component connection parts according to the connection tightness, and obtaining a structural quality evaluation result of the fastening functional component;

[0101] S204, performing a functional test on the movable parts of the fastening functional parts, obtaining the friction between the movable parts, analyzing the degree of jamming between the movable parts according to the friction, and obtaining the functional test result of the fastening functional parts;

[0102] S206, performing assembly quality assessment on the fastening functional component according to the structural quality assessment result and the functional test result of the fastening functional component to obtain an assembly quality assessment result.

[0103] It should be noted that by conducting mechanical tests on the connection parts of the fastening functional parts assembly and evaluating the tightness of the connection, it is possible to effectively ensure that the assembled functional parts have sufficient structural strength and stability; functional testing of the movable parts of the fastening functional parts, especially analyzing the friction and jamming between the movable parts, can effectively ensure the functionality of the product. For example, by evaluating the friction between the movable parts, problems that may cause poor operation of the parts during assembly can be discovered and corrected in a timely manner, thereby ensuring smooth operation of the product in actual use.

[0104] Figure 3 A flow chart of the calibration operation of the assembly parameters of the fastening functional parts of the present invention is shown.

[0105] According to an embodiment of the present invention, according to the assembly quality evaluation result, if the assembly quality of the fastening functional component is lower than the expected quality, the assembly parameters of the fastening functional component are calibrated, specifically:

[0106] S302, obtaining preset quality standard data of the fastening functional part, comparing the assembly quality evaluation result with the preset quality standard data, and if the assembly quality of the fastening functional part is lower than the preset quality standard, marking the item corresponding to the assembly quality lower than the preset quality standard as the assembly item to be optimized;

[0107] S304, performing a traceability analysis on the assembly item to be optimized, determining the assembly control parameters that cause quality problems on the assembly item to be optimized, and determining the calibration coefficients of the assembly control parameters according to the quality severity of the assembly item to be optimized;

[0108] S306, performing a calibration operation on the assembly parameters of the fastening functional parts according to the calibration coefficients, wherein the calibration operation includes calibrating the position of the robot arm and calibrating the torque parameters of the press-fitting equipment.

[0109] It should be noted that by calibrating the position of the robot arm and the torque parameters of the press-fitting equipment, the accuracy of the assembly process can be significantly improved. This improvement in accuracy helps ensure that each component can be accurately positioned and fixed, thereby reducing error accumulation, avoiding deviations and defects in the assembly process, and ultimately improving assembly quality; by marking and analyzing assembly items that are below the preset quality standards, assembly parameters that cause quality problems can be accurately located and calibrated in a targeted manner. This calibration operation can effectively reduce common defects in the assembly process, such as component misalignment, looseness, or improper tightening, thereby improving product consistency and reliability.

[0110] According to an embodiment of the present invention, it also includes:

[0111] Acquire laser scanning data of the top view of the fastening functional component under different environmental conditions in the assembly workshop to obtain environmental-laser scanning data;

[0112] Extracting surface feature data of the top view of the component obtained under different environmental conditions according to the environment-laser scanning data to obtain environment-surface feature data;

[0113] Acquire surface feature data extracted from the top view of the component under an ideal environment to obtain ideal environment-surface feature data;

[0114] Comparing the environment-surface feature data with the ideal environment-surface feature data to determine the accuracy of surface feature extraction of the top view of the component using laser scanning data acquired under different environmental conditions;

[0115] Acquire accuracy threshold data of surface feature extraction accuracy, mark environmental conditions whose accuracy is lower than the accuracy threshold, and obtain environmental conditions to be optimized;

[0116] Compare the laser scanning data obtained under the environmental conditions to be optimized with the laser scanning data obtained under the ideal environmental conditions, analyze the differences between the laser scanning data obtained under the environmental conditions to be optimized and the laser scanning data obtained under the ideal environmental conditions, determine the difference items of the laser scanning data according to the differences, and determine the repair parameters of the difference items, so as to obtain the repair parameters of the laser scanning data of the top view surface of the component under different environmental conditions;

[0117] The current environmental condition data of the assembly workshop is obtained in real time. If the current environmental condition is to be optimized, the laser scanning data obtained under the current environmental condition is repaired according to the repair parameters of the component top view laser scanning data under different environmental conditions.

[0118] It should be noted that the environmental conditions of the assembly workshop (such as humidity, light intensity, dust level, etc.) will affect the accuracy of the laser scanning data, resulting in deviations in the surface feature data obtained under different environmental conditions. This deviation may lead to component recognition errors and reduced assembly accuracy, which in turn affects the quality of the final product; therefore, by acquiring and analyzing the laser scanning data under different environmental conditions and comparing it with the data under the ideal environment, the environmental conditions that cause the scanning data deviation can be identified and marked. Subsequently, by repairing the scanned data under these environmental conditions, the stability and accuracy of the data can be significantly improved, ensuring that high-quality surface feature data can be obtained under various environmental conditions. By repairing the laser scanning data under the current environmental conditions in real time, the assembly accuracy problems caused by environmental fluctuations can be effectively reduced. Even under undesirable environmental conditions, high-precision operations can be maintained during the assembly process to ensure accurate positioning and assembly of components; the ideal environmental conditions are environmental conditions that will not cause laser data quality degradation during laser data acquisition.

[0119] According to an embodiment of the present invention, it also includes:

[0120] Acquire location information of laser scanning devices of multiple assembly production lines in a fastening functional parts assembly workshop, and construct a distribution map of laser scanning devices in the assembly workshop according to the location information;

[0121] Obtain data on the impact of data transmission distance on laser scanning data collection delay and the minimum data collection delay requirement for laser scanning data in assembly workshops;

[0122] Introducing a genetic algorithm to generate a set of combinations of positions and quantities of laser scanning data acquisition devices in the laser scanning device distribution map as an initial population, calculating the delay information of each laser scanning device in each combination through a delay model, and using the delay information as the value of the fitness function;

[0123] The next generation of population is generated through the selection, crossover and mutation operations of the genetic algorithm. After multiple population iterations, until the data transmission delay of the laser scanning equipment in the assembly workshop meets the minimum data acquisition delay requirement, the location information and quantity information of the laser scanning data acquisition equipment are output, and the layout plan of the laser scanning data acquisition equipment in the assembly workshop is obtained.

[0124] It should be noted that due to the complex layout of the assembly workshop, the number and location of laser scanning equipment, and the data transmission distance between equipment, data collection and transmission may face significant delays. This delay may affect the real-time monitoring, fault detection and data analysis of the production line, reducing production efficiency and product quality; therefore, by introducing genetic algorithms to optimize the location and number of laser scanning equipment, the delay in data collection and transmission can be effectively reduced. The optimized layout plan enables each device to collect data at the optimal location, reducing the distance of data transmission, thereby reducing the overall delay.

[0125] Figure 4 A block diagram of a fastening functional parts assembly system based on intelligent identification of the present invention is shown.

[0126] The second aspect of the present invention further provides a fastening functional component assembly system 4 based on intelligent identification, the system comprising: a memory 41, a processor 42, the memory comprising a fastening functional component assembly method program based on intelligent identification, the fastening functional component assembly method program based on intelligent identification being executed by the processor to implement the following steps:

[0127] Acquire laser scanning data of each component of the fastening functional part to build a three-dimensional model of the component, and build a component recognition system based on the three-dimensional model of the component;

[0128] Acquire assembly process data and assembly process data of the fastening functional part, extract components according to the assembly process data and the component identification system, adjust the component assembly position and assembly direction, control the assembly force of each component according to the assembly process data, and obtain the assembly plan of the fastening functional part;

[0129] Performing an assembly operation on the fastening functional component assembly according to the assembly scheme, and performing an assembly quality assessment on the fastening functional component after the assembly is completed to obtain an assembly quality assessment result;

[0130] According to the assembly quality evaluation result, if the assembly quality of the fastening functional component is lower than the expected quality, the assembly parameters of the fastening functional component are calibrated.

[0131] The present invention discloses a method and system for assembling fastening functional parts based on intelligent identification, aiming to improve assembly accuracy and efficiency. The present invention comprises the following steps: acquiring laser scanning data of each component to construct a three-dimensional model, and establishing a component identification system; extracting and adjusting the assembly position and direction of the component based on the assembly process and process data, while controlling the assembly force and generating an assembly plan; performing assembly operations according to the plan, and performing quality assessment on the assembled functional parts; if the assessment result is lower than expected, calibrating the assembly parameters. The present invention effectively improves the assembly accuracy, efficiency and quality stability.

[0132] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0133] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0134] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0135] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0136] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0137] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for assembling fastening functional parts based on intelligent identification, characterized in that: The following steps are involved: Acquire laser scanning data of each component of the fastening functional part to build a three-dimensional model of the component, and build a component recognition system based on the three-dimensional model of the component; Acquire assembly process data and assembly process data of the fastening functional part, extract components according to the assembly process data and the component identification system, adjust the component assembly position and assembly direction, control the assembly force of each component according to the assembly process data, and obtain the assembly plan of the fastening functional part; Performing an assembly operation on the fastening functional component assembly according to the assembly scheme, and performing an assembly quality assessment on the fastening functional component after the assembly is completed to obtain an assembly quality assessment result; According to the assembly quality evaluation result, if the assembly quality of the fastening functional component is lower than the expected quality, calibrating the assembly parameters of the fastening functional component; The assembly process data and assembly process data of the fastening functional parts are obtained, components are extracted according to the assembly process data and the component identification system, and the assembly position and assembly direction of the components are adjusted, and the assembly force of each component is controlled according to the assembly process data to obtain the assembly scheme of the fastening functional parts, which is specifically: Acquiring assembly process data and assembly process data of the fastening functional parts; Acquire laser scanning data of the top view surface of each component in the component conveyor belt in real time, and determine the surface features of the top view surface of each component according to the laser scanning data; Importing the surface features of the top view of each component in the conveyor belt into the component recognition system for feature matching, and determining the degree of matching between the top view features of each component in the conveyor belt and the surface features of the component standard production sample; The component standard production samples with a matching degree higher than a preset value are used as the matching result of each component in the conveyor belt to obtain the component recognition result in the conveyor belt; Determine the assembly component to be assembled that needs to be assembled in the current assembly process according to the assembly process data of the fastening functional part, and randomly grab one of the assembly components in the conveyor belt according to the component identification result; Acquire current gripping angle information of the grasped component to be assembled, the current gripping angle information including vertical angle information and horizontal angle information, adjust the assembly position and assembly direction of the component to be assembled according to the assembly process data and the current gripping angle information, and control the assembly force of each component to obtain an assembly plan for the fastening functional parts; Among them, it also includes: According to the component recognition result in the conveyor belt, marking a plurality of components to be assembled identified in the conveyor belt to obtain marked components; According to the matching degree between the top view feature of each component in the conveyor belt and the surface feature of the component standard production sample, analyzing whether there are multiple components whose surface features of the standard production sample and the top view surface feature of the marked component have a matching degree higher than a preset value; If so, the marked components whose surface feature matching degree with multiple component standard production samples is higher than the preset value are marked as pending components for which the component identification results are to be further confirmed, and the other marked components are marked as non-pending components for which the component identification results do not need to be further confirmed; When grabbing the components to be assembled, the non-pending components are grabbed and assembled first. When all the components to be assembled identified in the conveyor belt are pending components, the multiple components to which the pending components are identified are determined according to the component identification results, and the top view surface of the pending components is determined in the belonging components. According to the surface feature data of each component and the surface features of the top view surface of each component, the feature points of the top view surface of the component to be determined and the surface of each component are connected in pairs to form a feature connection graph; The characteristic connection diagram of the undetermined component is compared with the characteristic connection diagram of each of its belonging components, and the belonging components of the undetermined component are secondary identified to obtain a secondary identification result of the undetermined component, and the component identification result in the conveyor belt is corrected according to the secondary identification result.

2. A fastening functional parts assembly method based on intelligent identification according to claim 1, characterized in that: The step of acquiring laser scanning data of each component of the fastening functional part to construct a component three-dimensional model, and constructing a component identification system according to the component three-dimensional model, specifically includes: Obtaining a standard production sample of each component of the fastening functional part, and obtaining point cloud data of the standard production sample of each component based on a laser scanner; The outliers and noise points of the point cloud data are removed according to the statistical filtering method to obtain the preprocessed point cloud data; Projecting the preprocessed point cloud data into a three-dimensional grid, gridding the preprocessed point cloud data using an octree space partitioning structure, selecting the centroid of all points in each grid cell as a representative point in each grid cell, integrating the representative points of each grid cell to obtain compressed point cloud data; Constructing a three-dimensional model of the surface of each component based on a triangular mesh reconstruction algorithm and the compressed point cloud data, and performing a surface fitting operation on the three-dimensional model; Calculating the Gaussian curvature of the surface of the three-dimensional model of each component, extracting the surface features of each component according to the Gaussian curvature, and obtaining surface feature data of each component; A component recognition system is constructed based on a convolutional neural network, and the surface feature data of each component is imported into the component recognition system for learning and training to obtain a component recognition system with component recognition capability.

3. A fastening functional parts assembly method based on intelligent identification according to claim 2, characterized in that: The Gaussian curvature of the surface of the three-dimensional model of each component is calculated, and the surface features of each component are extracted according to the Gaussian curvature to obtain the surface feature data of each component, specifically: For each component's three-dimensional model, extract a triangular mesh of the three-dimensional model surface, wherein the triangular mesh is composed of a vertex set V, an edge set E, and a triangular facet F; Performing Gaussian curvature calculation on each triangular mesh vertex in the three-dimensional model, and marking the triangular mesh vertex entering the Gaussian curvature calculation as a triangular mesh vertex in a calculation state; Determine adjacent triangle patches to a vertex of a triangular mesh in a computing state, wherein the adjacent triangle patches share the vertex of the triangular mesh in a computing state, determine adjacent vertices to the vertex of the triangular mesh in a computing state according to the adjacent triangle patches, and obtain a vertex neighborhood of the vertex of the triangular mesh in a computing state, wherein the adjacent vertices are directly connected to the vertex of the triangular mesh in a computing state through a triangle edge; Determine the coordinates of each adjacent vertex in the vertex neighborhood according to the point cloud data, calculate each internal angle of the adjacent triangular facets based on the cosine theorem, calculate the sum of the internal angles of all triangular facets, and calculate the angle defect of the triangular mesh vertex in the calculated state according to the sum of the internal angles; Calculate the area of ​​the adjacent triangular facets, take a preset equal-division area from each adjacent triangular facet and add them up to obtain the associated local area area of ​​the triangular mesh vertex in the calculated state, divide the angle defect by the associated local area area to obtain the Gaussian curvature of the triangular mesh vertex in the calculated state, integrate and visualize the Gaussian curvatures of all triangular mesh vertices to obtain a Gaussian curvature distribution map of the three-dimensional model; The surface features of each component are determined according to the Gaussian curvature distribution diagram, wherein the surface features include feature points and geometric feature edges. The surface features are depicted to obtain surface feature data of each component.

4. The method for assembling fastening functional parts based on intelligent identification according to claim 1, characterized in that: The assembly operation of the fastening functional component assembly is performed according to the assembly scheme, and the assembly quality evaluation of the fastening functional component after the assembly is completed is performed to obtain the assembly quality evaluation result, which is specifically: Assembling the fastening functional component assembly according to the assembly scheme, performing mechanical testing on the assembled fastening functional component, evaluating the connection tightness of the component connection parts, determining the structural strength of the component connection parts according to the connection tightness, and obtaining a structural quality evaluation result of the fastening functional component; Performing a functional test on the movable parts of the fastening functional parts to obtain the friction between the movable parts, analyzing the degree of jamming between the movable parts according to the friction, and obtaining the functional test result of the fastening functional parts; An assembly quality assessment of the fastening functional part is performed according to the structural quality assessment result and the functional test result of the fastening functional part to obtain an assembly quality assessment result.

5. The method for assembling fastening functional parts based on intelligent identification according to claim 1, characterized in that: According to the assembly quality evaluation result, if the assembly quality of the fastening functional part is lower than the expected quality, the assembly parameters of the fastening functional part are calibrated, specifically: Acquire preset quality standard data of the fastening functional part, compare the assembly quality evaluation result with the preset quality standard data, and if the assembly quality of the fastening functional part is lower than the preset quality standard, mark the item corresponding to the assembly quality lower than the preset quality standard as the assembly item to be optimized; Conducting source tracing analysis on the assembly items to be optimized, determining assembly control parameters that cause quality problems on the assembly items to be optimized, and determining calibration coefficients of the assembly control parameters according to the quality severity of the assembly items to be optimized; The assembly parameters of the fastening functional parts are calibrated according to the calibration coefficients, and the calibration operation includes calibration of the position of the robot arm and calibration of the torque parameters of the press-fitting equipment.

6. A fastening functional parts assembly system based on intelligent identification, characterized in that: The fastening functional component assembly system based on intelligent identification includes a storage device and a processor, wherein the storage device includes a fastening functional component assembly method program based on intelligent identification, and when the fastening functional component assembly method program based on intelligent identification is executed by the processor, the following steps are implemented: Acquire laser scanning data of each component of the fastening functional part to build a three-dimensional model of the component, and build a component recognition system based on the three-dimensional model of the component; Acquire assembly process data and assembly process data of the fastening functional part, extract components according to the assembly process data and the component identification system, adjust the component assembly position and assembly direction, control the assembly force of each component according to the assembly process data, and obtain the assembly plan of the fastening functional part; Performing an assembly operation on the fastening functional component assembly according to the assembly scheme, and performing an assembly quality assessment on the fastening functional component after the assembly is completed to obtain an assembly quality assessment result; According to the assembly quality evaluation result, if the assembly quality of the fastening functional component is lower than the expected quality, calibrating the assembly parameters of the fastening functional component; The assembly process data and assembly process data of the fastening functional parts are obtained, components are extracted according to the assembly process data and the component identification system, and the assembly position and assembly direction of the components are adjusted, and the assembly force of each component is controlled according to the assembly process data to obtain the assembly scheme of the fastening functional parts, which is specifically: Acquiring assembly process data and assembly process data of the fastening functional parts; Acquire laser scanning data of the top view surface of each component in the component conveyor belt in real time, and determine the surface features of the top view surface of each component according to the laser scanning data; Importing the surface features of the top view of each component in the conveyor belt into the component recognition system for feature matching, and determining the degree of matching between the top view features of each component in the conveyor belt and the surface features of the component standard production sample; The component standard production samples with a matching degree higher than a preset value are used as the matching result of each component in the conveyor belt to obtain the component recognition result in the conveyor belt; Determine the assembly component to be assembled that needs to be assembled in the current assembly process according to the assembly process data of the fastening functional part, and randomly grab one of the assembly components in the conveyor belt according to the component identification result; Acquire current gripping angle information of the grasped component to be assembled, the current gripping angle information including vertical angle information and horizontal angle information, adjust the assembly position and assembly direction of the component to be assembled according to the assembly process data and the current gripping angle information, and control the assembly force of each component to obtain an assembly plan for the fastening functional parts; Among them, it also includes: According to the component recognition result in the conveyor belt, marking a plurality of components to be assembled identified in the conveyor belt to obtain marked components; According to the matching degree between the top view feature of each component in the conveyor belt and the surface feature of the component standard production sample, analyzing whether there are multiple components whose surface features of the standard production sample and the top view surface feature of the marked component have a matching degree higher than a preset value; If so, the marked components whose surface feature matching degree with multiple component standard production samples is higher than the preset value are marked as pending components for which the component identification results are to be further confirmed, and the other marked components are marked as non-pending components for which the component identification results do not need to be further confirmed; When grabbing the components to be assembled, the non-pending components are grabbed and assembled first. When all the components to be assembled identified in the conveyor belt are pending components, the multiple components to which the pending components are identified are determined according to the component identification results, and the top view surface of the pending components is determined in the belonging components. According to the surface feature data of each component and the surface features of the top view surface of each component, the feature points of the top view surface of the component to be determined and the surface of each component are connected in pairs to form a feature connection graph; The characteristic connection diagram of the undetermined component is compared with the characteristic connection diagram of each of its belonging components, and the belonging components of the undetermined component are secondary identified to obtain a secondary identification result of the undetermined component, and the component identification result in the conveyor belt is corrected according to the secondary identification result.

Images