Neural network-based automobile testing fixture design method and system

Abstract

The invention relates to the technical field of testing fixture design, and provides an automobile testing fixture design method and system based on a neural network, and the method comprises the steps: extracting function part data in a testing fixture sample model and part region structure features in an automobile part model, and learning the incidence relation between the function part data and the part region structure features in the neural network; according to the method, the detection tool design model is obtained through training, the current automobile part model is partitioned and then input into the detection tool design model, then the function part setting positions of all the areas can be obtained, the automobile detection tool corresponding to the automobile part is constructed, design of the detection tool of the automobile part can be completed rapidly and efficiently, the detection efficiency is improved, and the automobile production period is shortened.

Description

Technical Field

[0001] The present invention relates to the technical field of inspection tool design, and in particular to a method and system for designing automobile inspection tools based on a neural network. Background Art

[0002] Automotive inspection fixtures can be used to verify whether the design of parts is accurate and ensure that the parts can achieve the expected results during the assembly process. Each automotive part has a special automotive inspection fixture. With the rapid development of the automotive industry, there are more and more various automotive parts, so the corresponding automotive inspection fixtures also need to be customized according to the style of automotive parts. However, the current automotive inspection fixtures are generally designed by engineers according to the part model, which is inefficient and affects the overall production and quality inspection efficiency. Summary of the invention

[0003] The invention provides a method for designing automobile inspection tools based on a neural network, which is used to solve the problem of low efficiency of manually designing automobile inspection tools in the prior art.

[0004] The first aspect of the present invention provides a method for designing automobile inspection tools based on a neural network, comprising: Obtain a sample model of a test fixture and a corresponding automobile component model; divide the area of ​​the automobile component model according to the component characteristics to obtain a plurality of first component areas, and identify the structural features of each first component area; identify the position of the first functional part in the sample model of the test fixture, and associate the first functional part position with the first component area; Inputting the structural features of the first component area and the corresponding first functional part position into a preset neural network model, constructing the correlation between the structural features of the automobile component area and the functional part position, and training to obtain a fixture design model; The current automobile component model to be designed is obtained, and the region is divided according to the component characteristics to obtain multiple second component regions, the structural characteristics of each region are identified and input into the inspection fixture design model to obtain the structural characteristics of each second component region; the structural characteristics of each second component region are respectively input into the inspection fixture design model to obtain the corresponding second functional component position, and the automobile inspection fixture of the current automobile component is constructed according to each second functional component position.

[0005] Optionally, the obtaining of the corresponding second functional component positions and constructing an automobile inspection fixture of the current automobile component according to each second functional component position specifically further includes: Traverse the second functional part location scheme corresponding to the second component area, identify the number of second functional parts of various types in the second functional part location scheme, calculate the detection efficiency of the second functional parts according to the structural characteristics of the second component area, select the second functional part location scheme with the highest detection efficiency of various types of second functional parts, and build an automobile inspection fixture for the current automobile component.

[0006] Optionally, the region of the automobile component model is divided according to the component characteristics to obtain a plurality of first component regions, specifically: The connection positions in the automobile component model are identified, the automobile component model is preliminarily divided based on the connection positions, and then the component characteristics in the preliminarily divided areas are identified to obtain a plurality of first component areas.

[0007] The second aspect of the present application provides a neural network-based automobile inspection fixture design system, comprising: A model data processing module is used to obtain a sample model of a test fixture and a corresponding automobile component model; divide the area of ​​the automobile component model according to the component characteristics to obtain a plurality of first component areas, and identify the structural characteristics of each first component area; identify the position of the first functional part in the sample model of the test fixture, and associate the first functional part position with the first component area; A neural network training module is used to input the structural features of the first component area and the corresponding first functional part position into a preset neural network model, build an association relationship between the structural features of the automobile component area and the functional part position, and train to obtain a fixture design model; The fixture design module is used to obtain the current automobile component model to be designed, and divide the area according to the component characteristics to obtain multiple second component areas, identify the structural characteristics of each area and input them into the fixture design model to obtain the structural characteristics of each second component area; input the structural characteristics of each second component area into the fixture design model respectively to obtain the corresponding second functional part position, and construct the automobile fixture of the current automobile component according to each second functional part position.

[0008] Optionally, in the fixture design module, the corresponding second functional component positions are obtained, and the automobile fixture of the current automobile component is constructed according to the second functional component positions, which specifically also includes: Traverse the second functional part location scheme corresponding to the second component area, identify the number of second functional parts of various types in the second functional part location scheme, calculate the detection efficiency of the second functional parts according to the structural characteristics of the second component area, select the second functional part location scheme with the highest detection efficiency of various types of second functional parts, and build an automobile inspection fixture for the current automobile component.

[0009] Optionally, in the model data processing module, the region of the automobile component model is divided according to the component characteristics to obtain a plurality of first component regions, specifically: The connection positions in the automobile component model are identified, the automobile component model is preliminarily divided based on the connection positions, and then the component characteristics in the preliminarily divided areas are identified to obtain a plurality of first component areas.

[0010] The third aspect of the present application provides a method and device for designing automobile inspection tools based on a neural network, the device comprising a processor and a memory: The memory is used to store program codes and transmit the program codes to the processor; The processor is used to execute the automobile inspection fixture design method based on neural network according to any one of the first aspects of the present invention according to the instructions in the program code.

[0011] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store program codes, and the program codes are used to execute the neural network-based automobile inspection fixture design method according to any one of the first aspects of the present invention.

[0012] It can be seen from the above technical scheme that the present invention has the following advantages: by extracting the functional parts data in the inspection fixture sample model and the component regional structural features in the automobile component model, learning the correlation between the two in the neural network, training to obtain the inspection fixture design model, partitioning the current automobile component model and inputting it into the inspection fixture design model, the functional parts setting positions of each zone can be obtained respectively, and then the automobile inspection fixture corresponding to the automobile component is constructed, which can quickly and efficiently complete the design of the inspection fixture for the automobile component, improve the detection efficiency, and shorten the automobile production cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0014] Figure 1 This is a flowchart of a method for designing automobile inspection fixtures based on neural networks; Figure 2 Design system structure diagram for an automobile inspection fixture based on neural network. DETAILED DESCRIPTION

[0015] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] The invention provides a method for designing automobile inspection tools based on a neural network, which is used to solve the problem of low efficiency of manually designing automobile inspection tools in the prior art.

[0017] See also Figure 1 , Figure 1 The present invention provides a first flow chart of a method for designing automobile inspection tools based on a neural network according to an embodiment of the present invention.

[0018] S100, obtaining a sample model of a test fixture and a corresponding automobile component model; dividing the area of ​​the automobile component model according to the component characteristics to obtain a plurality of first component areas, and identifying the structural features of each first component area; identifying the position of a first functional component in the sample model of the test fixture, and associating the first functional component position with the first component area; It should be noted that the inspection fixture sample model is an automobile inspection fixture designed by engineers based on automobile parts in the past. After the automobile parts are produced, they can be installed in the inspection fixture sample model to inspect the surface shape, peripheral dimensions and connection positions required by the parts. The automobile inspection fixture is specially designed for the corresponding automobile parts, so there is a corresponding relationship between the inspection fixture sample model and the automobile part model. In this embodiment, the inspection fixture sample model and the corresponding automobile part model are both three-dimensional models, and each point of the model can obtain corresponding coordinates in the spatial coordinate system of the same size. Automobile parts are made by welding, casting and other processes, and multiple structural areas with different functions are combined to form an integral automobile part. The first component area has different detection requirements because of the different functions it plays. For example, the assembly areas at both ends of the bumper need to have size and spacing tolerance requirements so that the bumper can be fixedly connected to the frame. The outer plate area of ​​the bumper needs to ensure that the shape curvature, material elasticity and plastic deformation resistance meet the requirements so that the bumper has an anti-collision effect. Therefore, in a complete automobile component model, it can be divided into multiple first component areas according to different characteristics, and each first component area plays a different role in the automobile component model; and due to different functions, the parameters related to the first component area and the corresponding function can be used as the characteristics of the structure, such as the curvature and size of the bumper surface, the bolt position of the connecting structure and other characteristics; A number of functional parts will be arranged on the sample model of automobile inspection fixture to detect the characteristic conditions of automobile parts. The types of functional parts include positioning pins, detection pins, marking pins, gaskets, profile clamps, quick chucks, and detection instruments. The detection of the first functional part should correspond to the first component area. In the sample model of the inspection fixture, the coordinate position of each first functional part can be obtained according to the spatial coordinate system, and the divided first component area also contains coordinate information. Although the sample model of the inspection fixture and the corresponding automobile component model are two independent models in different spatial coordinate systems, the two models need to be installed and combined in actual detection. Therefore, they can be associated according to the coordinate position of the first functional part corresponding to the coordinate position of the first component area. The position of the first functional part specifically includes the coordinate positions of different types of functional parts set in the inspection fixture. In addition to the coordinate data, there is also the type data of the functional part.

[0019] S200, inputting the structural features of the first component area and the corresponding first functional part position into a preset neural network model, constructing a correlation between the structural features of the automobile component area and the functional part position, and training to obtain a fixture design model; It should be noted that the setting position of the functional part should be related to the component area that needs to be inspected, and the larger the size of the component area or the more items that need to be inspected, the more functional parts should be in the same component area; the preset neural network model can adopt a graph neural network (GNN), which is particularly suitable for processing data with complex topological structures, such as points and edges in three-dimensional mechanical models, and can better capture local and global features in the model, improve the accuracy and robustness of the model, and there is a corresponding correlation between the position of the functional part and the structural characteristics of the corresponding component area. The properties of the structural characteristics of the component area are different, and the types of functional parts used are also different. When the data under the structural characteristics of the component area changes, such as the size and thickness change, the number of functional parts will also change accordingly. After the specific shape of the structural characteristics of the component area changes, the setting coordinate position of the functional part will also change. These change relationships constitute an association relationship. After inputting multiple sets of inspection fixture sample models and multiple functional parts positions and component area structural characteristics under the automobile component model into the neural network model, the model can be trained to obtain the inspection fixture design model.

[0020] S300, obtaining a model of a current automobile component to be designed, and dividing the model into regions according to the component characteristics to obtain a plurality of second component regions, identifying the structural features of each region and inputting them into a fixture design model to obtain the structural features of each second component region; inputting the structural features of each second component region into the fixture design model to obtain the corresponding second functional component position, and constructing an automobile fixture for the current automobile component according to each second functional component position.

[0021] It should be noted that, for the current automobile component model for which the jig is to be designed, it is necessary to refer to the area division method in the aforementioned step S100 to obtain multiple second component areas, and identify the structural features of each second component area; the jig design model trained in the aforementioned step S200 can output the position where the jig functional parts need to be set according to the three-dimensional features of the automobile component, and the functional part setting position can also be obtained by inputting the complete automobile component model into the jig design model, but because the shapes and characteristics of the components in different areas will affect each other, the jig design model is extremely inefficient in identifying the position setting of the functional parts; the basic shape of the automobile jig needs to correspond to the current automobile component to be tested. After obtaining the second functional part positions corresponding to each second component area, the second functional parts can be set according to the positions on the automobile jig of the current automobile component.

[0022] In this embodiment, by extracting the functional parts data in the inspection fixture sample model and the component regional structural features in the automobile component model, the correlation between the two is learned in the neural network, and the inspection fixture design model is trained. After the current automobile component model is partitioned and input into the inspection fixture design model, the functional parts setting positions of each zone can be obtained respectively, and then the automobile inspection fixture corresponding to the automobile component is constructed, which can quickly and efficiently complete the design of the inspection fixture for the automobile component, improve the detection efficiency, and shorten the automobile production cycle.

[0023] The above is a detailed description of the first embodiment of a method for designing an automobile inspection tool based on a neural network provided by the present application. The following is a detailed description of the second embodiment of a method for designing an automobile inspection tool based on a neural network provided by the present application.

[0024] In this embodiment, a method for designing automobile inspection fixtures based on a neural network is further provided. In the aforementioned step S300, the corresponding second functional component positions are obtained, and an automobile inspection fixture of the current automobile component is constructed according to each second functional component position, which specifically also includes: Traverse the second functional part location scheme corresponding to the second component area, identify the number of second functional parts of various types in the second functional part location scheme, calculate the detection efficiency of the second functional parts according to the structural characteristics of the second component area, select the second functional part location scheme with the highest detection efficiency of various types of second functional parts, and build an automobile inspection fixture for the current automobile component.

[0025] It should be noted that for the same component area, there may be multiple groups of second functional parts positions obtained by the inspection fixture design model. For example, for a car component surface, densely arranged positioning pins can be used to detect the surface, but several key pins on the surface can also be set to achieve detection. When multiple functional parts of the same type are required to cover the component area, multiple position schemes will appear. These schemes differ in the number and distribution of functional parts. There are many structural features of the second component area input into the inspection fixture design model, including the functional properties, size, thickness and appearance of the component area. Inputting them into the inspection fixture design model in different orders will result in different second functional parts position schemes. When designing the inspection fixture, it is generally followed as much as possible. The principle of realizing detection by using a small number of functional parts can not only reduce the cost of inspection fixture production design but also reduce the inspection process of inspection personnel; therefore, it is necessary to identify the number of second functional parts of various types in the second functional part position scheme. For the same type of second functional parts, the average component area or volume of the functional parts detected is calculated according to the size of the component area, so as to judge the detection efficiency of the second functional parts, that is, the larger the average component area or volume detected by the same type of second functional parts, the greater the detection efficiency. The second functional part position scheme with the highest detection efficiency for each functional part type is adopted, and it is also necessary to consider that the positions of the second functional parts do not overlap with each other to avoid setting conflicts after the scheme is selected, so as to construct the automobile inspection fixture of the current automobile parts.

[0026] Furthermore, in the aforementioned step S100, the region of the automobile component model is divided according to the component characteristics to obtain a plurality of first component regions, specifically: a connection position in the automobile component model is identified, the automobile component model is preliminarily divided based on the connection position, and the component characteristics in the preliminarily divided regions are then identified to obtain a plurality of first component regions; It should be noted that the automobile parts tested in this embodiment generally need to be welded or other connection processes to form multiple castings into a complete automobile part, such as a frame, etc., and when such automobile parts are tested with a gauge, it is necessary to first fix the automobile parts on the gauge, so a preliminary division will be made based on the connection conditions of the automobile parts, and each divided area needs to be set with a functional part for fixation. The feature considered for the functional part setting in the area divided in this step is the shape of the part; and there will be other component characteristics in the area after the preliminary division, such as a positioning hole for connecting the automobile part to other parts or a strength requirement, so in the area after the preliminary division, the component characteristics are used to identify the role of each component and the area that plays the role, and multiple first component areas are divided, and if there are the same functional component characteristics and they are related, they should be divided into the same component area, for example, in a certain automobile part, bolt holes in symmetrical positions are required, and not only the quality of the bolt holes themselves but also the relative positions between the bolt holes need to be tested.

[0027] The above is a detailed description of a neural network-based automobile inspection fixture design method according to the first aspect of the present application. The following is a detailed description of an embodiment of a neural network-based automobile inspection fixture design system according to the second aspect of the present application.

[0028] See also Figure 2 , Figure 2 This is a structural diagram of a vehicle inspection fixture design system based on a neural network. This embodiment provides a vehicle inspection fixture design system based on a neural network, including: The model data processing module 10 is used to obtain a sample model of the inspection fixture and a corresponding automobile component model; divide the area of ​​the automobile component model according to the component characteristics to obtain a plurality of first component areas, and identify the structural characteristics of each first component area; identify the position of the first functional part in the sample model of the inspection fixture, and associate the first functional part position with the first component area; A neural network training module 20 is used to input the structural features of the first component area and the corresponding first functional part position into a preset neural network model, build an association relationship between the structural features of the automobile component area and the functional part position, and train to obtain a fixture design model; The fixture design module 30 is used to obtain the current automobile component model to be designed, and divide the area according to the component characteristics to obtain multiple second component areas, identify the structural characteristics of each area and input them into the fixture design model to obtain the structural characteristics of each second component area; input the structural characteristics of each second component area into the fixture design model respectively to obtain the corresponding second functional component position, and construct the automobile fixture of the current automobile component according to each second functional component position.

[0029] Furthermore, in the fixture design module 30, the corresponding second functional component positions are obtained, and the automobile fixture of the current automobile component is constructed according to the second functional component positions, which specifically includes: Traverse the second functional part location scheme corresponding to the second component area, identify the number of second functional parts of various types in the second functional part location scheme, calculate the detection efficiency of the second functional parts according to the structural characteristics of the second component area, select the second functional part location scheme with the highest detection efficiency of various types of second functional parts, and build an automobile inspection fixture for the current automobile component.

[0030] Furthermore, in the model data processing module 10, the region of the automobile component model is divided according to the component characteristics to obtain a plurality of first component regions, specifically: The connection positions in the automobile component model are identified, the automobile component model is preliminarily divided based on the connection positions, and then the component characteristics in the preliminarily divided areas are identified to obtain a plurality of first component areas.

[0031] The third aspect of the present application also provides a neural network-based automobile inspection fixture design method device, including a processor and a memory: wherein the memory is used to store program code and transmit the program code to the processor; the processor is used to execute the above-mentioned neural network-based automobile inspection fixture design method according to the instructions in the program code.

[0032] A fourth aspect of the present application provides a computer-readable storage medium, characterized in that the computer-readable storage medium is used to store program codes, and the program codes are used to execute the above-mentioned automobile inspection fixture design method based on neural network.

[0033] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and equipment can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0034] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, 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 integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0035] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, 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 solution of this embodiment.

[0036] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0037] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0038] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for designing automobile inspection tools based on neural network, characterized in that include: Obtaining a sample model of a test fixture and a corresponding automobile component model; dividing the area of ​​the automobile component model according to the component characteristics to obtain a plurality of first component areas, and identifying the structural features of each first component area; Identify a first functional part position in the sample model of the inspection fixture, and associate the first functional part position with a first component region; Inputting the structural features of the first component area and the corresponding first functional part position into a preset neural network model, constructing the correlation between the structural features of the automobile component area and the functional part position, and training to obtain a fixture design model; Obtain the current automobile component model to be designed, and divide the region according to the component characteristics to obtain multiple second component regions, identify the structural features of each region and input them into the inspection fixture design model to obtain the structural features of each second component region; The structural features of each second component area are respectively input into the inspection fixture design model to obtain the corresponding second functional component position, and the automobile inspection fixture of the current automobile component is constructed according to each second functional component position.

2. The method for designing automobile inspection tools based on neural network according to claim 1, characterized in that: The obtaining of the corresponding second functional component positions and constructing the automobile inspection fixture of the current automobile component according to the positions of the second functional components specifically also includes: Traverse the second functional part location scheme corresponding to the second component area, identify the number of second functional parts of various types in the second functional part location scheme, calculate the detection efficiency of the second functional parts according to the structural characteristics of the second component area, select the second functional part location scheme with the highest detection efficiency of various types of second functional parts, and build an automobile inspection fixture for the current automobile component.

3. The method for designing automobile inspection tools based on neural network according to claim 1, characterized in that: The region of the automobile component model is divided according to the component characteristics to obtain a plurality of first component regions, specifically: The connection positions in the automobile component model are identified, the automobile component model is preliminarily divided based on the connection positions, and then the component characteristics in the preliminarily divided areas are identified to obtain a plurality of first component areas.

4. A neural network-based automobile inspection fixture design system, characterized in that: include: A model data processing module is used to obtain a sample model of a test fixture and a corresponding automobile component model; divide the area of ​​the automobile component model according to the component characteristics to obtain a plurality of first component areas, and identify the structural features of each first component area; Identify a first functional part position in the sample model of the inspection fixture, and associate the first functional part position with a first component region; A neural network training module is used to input the structural features of the first component area and the corresponding first functional part position into a preset neural network model, build an association relationship between the structural features of the automobile component area and the functional part position, and train to obtain a fixture design model; A fixture design module is used to obtain a current automobile component model to be designed, and divide the region according to the component characteristics to obtain multiple second component regions, identify the structural features of each region and input them into the fixture design model to obtain the structural features of each second component region; The structural features of each second component area are respectively input into the inspection fixture design model to obtain the corresponding second functional component position, and the automobile inspection fixture of the current automobile component is constructed according to each second functional component position.

5. The neural network-based automobile inspection fixture design system according to claim 4, characterized in that: In the fixture design module, the corresponding second functional part positions are obtained, and the automobile fixture of the current automobile component is constructed according to the second functional part positions, which specifically includes: Traverse the second functional part location scheme corresponding to the second component area, identify the number of second functional parts of various types in the second functional part location scheme, calculate the detection efficiency of the second functional parts according to the structural characteristics of the second component area, select the second functional part location scheme with the highest detection efficiency of various types of second functional parts, and build an automobile inspection fixture for the current automobile component.

6. The neural network-based automobile inspection fixture design system according to claim 4 is characterized in that: In the model data processing module, the region of the automobile component model is divided according to the component characteristics to obtain a plurality of first component regions, specifically: The connection positions in the automobile component model are identified, the automobile component model is preliminarily divided based on the connection positions, and then the component characteristics in the preliminarily divided areas are identified to obtain a plurality of first component areas.

7. A neural network-based automobile inspection fixture design device, characterized in that: The device comprises a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the automobile inspection fixture design method based on neural network according to any one of claims 1 to 3 according to the instructions in the program code.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program codes, and the program codes are used to execute the automobile inspection fixture design method based on neural network according to any one of claims 1 to 3.

Images