Method, device, system and computer storage medium for optimizing gap and flushness of vehicle body appearance

Abstract

The present application relates to a method, device, system and computer storage medium for optimizing the gap flushness of the vehicle body appearance. The method includes: constructing an initial gap flushness threshold model using the vehicle body gap flushness data accumulated during the vehicle body production stage; training the initial gap flushness threshold model using the final assembly gap flushness data accumulated during the final assembly production stage through machine learning; evaluating the trained initial gap flushness threshold model through machine learning to obtain a gap flushness threshold prediction model that meets preset conditions; and setting the vehicle body gap flushness threshold in the vehicle body production stage using the gap flushness threshold prediction model that meets preset conditions. The gap flushness threshold prediction model constructed according to the present disclosure can predict in advance, adjust in real time, and provide rapid feedback, thereby improving the accuracy of setting the gap flushness threshold, reducing the probability of unqualified products in the final assembly production stage, and improving production efficiency and product quality.

Description

Technical Field

[0001] The present application generally relates to the measurement of the gap and flushness of the vehicle body appearance during the automobile production process. Specifically, the present application relates to a method, device, system and computer storage medium for optimizing the gap and flushness of the vehicle body appearance, and a vehicle production line. Background Art

[0002] With the rapid development of the automobile industry, people have put forward higher and higher requirements for the appearance of automobiles. Therefore, the gap and offset of the body appearance, as an important indicator for evaluating the beauty of the whole vehicle, has been increasingly valued by automobile manufacturers. Many automobile manufacturers have realized that solving the gap and offset problem of the body appearance is one of the key technologies to improve the competitiveness of the automobile market.

[0003] On the other hand, driven by sustainable production and Industry 4.0 technologies, manufacturing is gradually developing towards intelligent flexible manufacturing. With the characteristics of intelligent manufacturing, the output and quality of vehicle manufacturing have been greatly improved. In particular, during the stamping, body, painting, and assembly of the vehicle production line, tedious and repetitive work has gradually been automated through robots, mechanical arms, sensors, etc., and a large amount of data of various types has been accumulated.

[0004] Specifically, in the main production process of the automobile production line (i.e. stamping, body, painting, and final assembly), a large amount of data is involved in measurement, storage, analysis, and processing. For example, at the end of the installation in the body shop and at the end of the installation in the final assembly shop, the gap and flushness of the body appearance will be measured and evaluated according to the production standards. The gap and flushness of the body appearance usually refers to the design gap value (Gap) and deviation (Offset) reserved between two adjacent parts for easy movement and assembly.

[0005] At present, the measurement methods for the gap and flushness of the vehicle body appearance mainly include:

[0006] 1) Traditional manual inspection methods, such as using feeler gauges and flatness gauges for measurement, have limitations such as strong subjectivity and long time consumption. Manual inspection often requires a large number of workers.

[0007] 2) Automated inspection methods that combine laser or robot processing with computer processing mainly use laser sensors and robots to measure and collect data, and use computers for data processing. This inspection method requires the construction of safety fences on the finishing line with limited space, which is also not ideal.

[0008] At present, the measurement of gap flushness is mainly limited to specific production environments and production stages. For example, in the body production stage, the gap flushness of the body appearance will be measured at the end of the body production stage, and then the measured body gap flushness data will be compared with the body gap flushness threshold specified in the body production stage. For vehicles that meet the body gap flushness threshold, they will enter the next production process stage; otherwise, vehicles that do not meet the body gap flushness threshold will be reworked or even eliminated. Similarly, in the final assembly production stage, the final assembly gap flushness of the body appearance will also be measured at the end of the final assembly production stage, and then the measured final assembly gap flushness data will be compared with the final assembly gap flushness threshold specified in the final assembly production stage. For vehicles that meet the final assembly gap flushness threshold, they will be shipped; otherwise, vehicles that do not meet the final assembly gap flushness threshold will be reworked or eliminated.

[0009] The inventors of the present application have noted that:

[0010] 1) With the development of computer technology and the use of various precise measuring tools, the measurement accuracy of gap and flushness data has been improved. At the same time, a large amount of gap and flushness measurement data has been accumulated in the body production stage and the final assembly production stage, but these data have not been fully analyzed and utilized;

[0011] 2) When the body gap face difference in the body production stage fully meets the body gap face difference threshold value specified in the body production stage, the assembly gap face difference in the final assembly production stage still does not meet the assembly gap face difference threshold value specified in the final assembly production stage;

[0012] 3) There is currently no technology that correlates the body gap flushness at the body production stage with the assembly gap flushness at the assembly production stage;

[0013] 4) The analysis of the body gap flushness data in the body production stage is only used to adjust the body gap flushness threshold in the body production stage, and the analysis of the final assembly gap flushness data in the final assembly production stage is also only used to adjust the final assembly gap flushness threshold in the final assembly production stage. The analysis results of the final assembly gap flushness data in the final assembly production stage are not used to adjust the body gap flushness in the body production stage, such as optimizing and adjusting the vehicle or presetting the body gap flushness threshold in the body production stage.

[0014] In view of the problems existing in the above-mentioned prior art, it is hoped to provide a method that can comprehensively consider the body gap flushness in the body production stage and the assembly gap flushness in the final assembly production stage. It is further hoped to optimize and adjust the body gap flushness threshold in the body production stage by analyzing the final assembly gap flushness data in the final assembly production stage and considering the influence of the body gap flushness data on the final assembly gap flushness data. Summary of the invention

[0015] In order to solve the technical problems existing in the prior art, according to the present disclosure, a method for optimizing the gap and flushness of a vehicle body appearance is provided, comprising:

[0016] constructing an initial gap flushness threshold model using body gap flushness data accumulated during the body production phase;

[0017] Training the initial gap and flushness threshold model by machine learning using the final assembly gap and flushness data accumulated during the final assembly production phase;

[0018] By machine learning, the trained initial gap and flushness threshold model is evaluated to obtain a gap and flushness threshold prediction model that meets preset conditions; and

[0019] Use the gap and flushness threshold prediction model that meets the preset conditions to set the body gap and flushness threshold in the body production stage.

[0020] According to the present disclosure, a computer-readable storage medium is also provided, which has executable instructions. When the instructions are executed, the instructions prompt a computer to execute the method for optimizing the gap and flushness of the vehicle body appearance according to the above aspects of the present disclosure.

[0021] According to the present disclosure, there is also provided a device for optimizing the gap and flushness of a vehicle body appearance, the device comprising:

[0022] - a memory storing computer executable instructions; and

[0023] - a processor configured to execute computer executable instructions, wherein when the computer executable instructions are executed by the processor, the processor executes the method for optimizing the gap and flushness of the vehicle body appearance according to the above aspect of the present disclosure.

[0024] According to the present disclosure, a system for optimizing the gap and flushness of a vehicle body appearance is also provided, the system comprising:

[0025] - an industrial robot, taking an image of the vehicle body appearance and acquiring gap and flushness data of the vehicle body appearance;

[0026] - an apparatus for optimizing the gap flushness of a vehicle body appearance according to the above aspect of the present disclosure; and

[0027] -An alarm device is used to output an alarm signal when it is identified that the vehicle body gap difference threshold is exceeded.

[0028] According to the present disclosure, a vehicle production line is also provided, comprising the system for optimizing the gap and flushness of the vehicle body appearance according to the above aspects of the present disclosure.

[0029] According to the present disclosure, data mining and machine learning technologies are fully utilized, the body gap face difference in the body production stage and the assembly gap face difference in the assembly production stage are comprehensively considered, and a gap face difference threshold (prediction) model suitable for actual production use is constructed.

[0030] Moreover, when constructing the gap flushness threshold (prediction) model, the body gap flushness threshold in the body production stage is optimized and adjusted by analyzing the final assembly gap flushness data in the final assembly production stage and considering the influence of the body gap flushness data on the final assembly gap flushness data.

[0031] The gap and flushness threshold (prediction) model constructed according to the present invention can predict in advance, adjust in real time, and provide rapid feedback, thereby improving the accuracy of setting the gap and flushness threshold, reducing the probability of unqualified products in the final assembly production stage, and improving production efficiency and product quality.

[0032] It should be noted that the features, functions, effects and advantages of one aspect of the present invention can also refer to the above description of other aspects of the present invention. In addition, the various aspects described in this article can be combined with each other in a variety of ways.

[0033] These and other features and advantages of the present disclosure will be set forth in the following description, and in part will be apparent from the description, or can be learned by practicing the principles disclosed herein. The features and advantages of the present disclosure can be realized and obtained by means of the means and combinations specifically pointed out in the specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The above-recited advantages and features and other advantages and features of the present disclosure will become apparent by referring to the specific embodiments of the present disclosure illustrated in the accompanying drawings. Understanding that these drawings depict only example embodiments of the present disclosure and are therefore not to be considered limiting of the scope of the present disclosure, the principles herein are more particularly and in more detail described and explained through the use of the accompanying drawings, in which:

[0035] Figure 1 A flow chart schematically shows a method for optimizing the gap and flushness of a vehicle body appearance according to an embodiment of the present application;

[0036] Figure 2 A flowchart of a method for constructing an initial gap flushness threshold model according to an embodiment of the present application is schematically shown;

[0037] Figure 3 A flowchart of a method for training an initial gap flushness threshold model according to an embodiment of the present application is schematically shown;

[0038] Figure 4 A flow chart of a method for evaluating a trained initial gap flushness threshold model according to an embodiment of the present application is schematically shown;

[0039] Figure 5 A diagram schematically showing a correlation analysis between vehicle body gap flushness and assembly gap flushness according to an embodiment of the present application;

[0040] Figure 6 A flowchart schematically shows a method for updating a gap flushness threshold prediction model according to an embodiment of the present application; and

[0041] Figure 7 An exemplary configuration of a computing device capable of implementing an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0042] In the following description of the examples, reference will be made to the accompanying drawings, in which specific examples that may be implemented are shown by way of illustration. It should be understood that other examples may be used and structural changes may be made without departing from the scope of the various examples.

[0043] Various techniques and process flow steps will now be described in detail with reference to the examples shown in the accompanying drawings. In the following description, numerous specific details are set forth to provide a thorough understanding of one or more aspects and / or features described or mentioned therein. However, it is apparent to those skilled in the art that one or more aspects and / or features described or mentioned may be implemented without some or all of these specific details. In other cases, known process steps and / or structures are not described in detail so as not to obscure some of the aspects and / or features described or mentioned herein.

[0044] With the development of Internet technology, data analysis, data mining, data warehouse, online analysis, deep learning, machine learning, artificial intelligence, etc. are increasingly being applied to all walks of life. The relationship between data mining and machine learning is getting closer and closer. Data mining is to mine hidden, useful, and correct knowledge from a large amount of business data to promote the execution of decisions. Data mining is usually related to computer science, and it achieves the above goals through many methods such as statistics, online analytical processing, intelligence retrieval, machine learning, expert systems (relying on past empirical rules) and pattern recognition. Data mining is a technology that analyzes each data and finds its rules from a large amount of data. It mainly includes three steps: data preparation, rule finding and rule expression. Data preparation is to select the required data from relevant data sources and integrate them into a data set for data mining; rule finding is to find the rules contained in the data set by some method; rule expression is to express the found rules in a way that users can understand as much as possible (such as visualization). The tasks of data mining include association analysis, cluster analysis, classification analysis, anomaly analysis, specific group analysis and evolution analysis.

[0045] The inventors noticed the technical problems existing in the prior art and creatively thought of applying data mining technology and machine learning to the measurement of body gap flushness and the setting of thresholds. In particular, the general inventive concept of the present application is to apply big data mining technology to a large amount of accumulated gap flushness measurement data (preferably, body gap flushness measurement data in the body production stage and assembly gap flushness measurement data in the assembly production stage), use machine learning methods to construct a gap flushness threshold model, and train and optimize the constructed gap flushness threshold model, thereby using the optimized gap flushness threshold model to set the body gap flushness threshold in the body production stage in advance or adjust the body gap flushness threshold in advance.

[0046] The following describes in detail the method, device, system, computer storage medium, etc. for optimizing the gap and flushness of the vehicle body appearance according to various embodiments of the present application in conjunction with the accompanying drawings.

[0047] Figure 1 A flow chart of a method for optimizing the gap and flushness of a vehicle body appearance according to an embodiment of the present application is schematically shown.

[0048] Refer to the attached Figure 1 First, in step 102, an initial gap flushness threshold model is constructed using the body gap flushness data accumulated during the body production stage.

[0049] In the main production processes of automobile production lines (i.e., stamping, body, painting, and assembly), a large amount of data is measured, saved, analyzed, and processed. In particular, in the current automobile production line environment, a large amount of various measurement data about the body is accumulated. Specifically, the vehicle that has completed the body production process needs to be quality inspected, which includes measuring the body gap and flushness, and thus a large amount of body gap and flushness data is accumulated during the body production stage.

[0050] The body gap flushness data accumulated during the body production stage includes: body gap flushness data measured for a plurality of vehicles at the end of the body production stage during a preset time period, wherein a plurality of sample points are set on the body of each vehicle, and the body gap flushness data of each sample point is identified by the body identification number I-GEF and the sample point number. The preset time period may be a recent period in the current production environment, such as 3 months, 6 months, 1 year or a longer period.

[0051] Generally, multiple sample points on the vehicle body are selected according to the quality specifications of the vehicle body production stage. Preferably, the selection of sample points is mainly concentrated on the positions between the components on the vehicle body.

[0052] In addition, the vehicle body gap flushness measurement for the selected sample point includes performing multiple gap flushness measurements on the same sample point, and selecting an average value of the multiple vehicle body gap flushness measurement values ​​as the vehicle body gap flushness data for the sample point.

[0053] Preferably, the vehicle body gap flushness data of the plurality of sample points may be stored in a local and / or remote server to facilitate viewing and retrieval via a network.

[0054] The following will be combined with the attached Figure 2 The method steps for constructing an initial gap flushness threshold model using vehicle body gap flushness data accumulated during a vehicle body production stage according to an embodiment of the present application are described in detail.

[0055] Next, at step 104, the initial gap and flushness threshold model is trained by machine learning using the final assembly gap and flushness data accumulated during the final assembly production phase.

[0056] Similar to measuring the body gap and flushness during the body production stage, vehicles that have completed the final assembly production process also need to be quality inspected to meet factory requirements. This also includes measuring the final assembly gap and flushness, and thus a large amount of final assembly gap and flushness data is accumulated during the final assembly production stage.

[0057] The final assembly gap flushness data accumulated during the final assembly production stage includes: final assembly gap flushness data measured for a plurality of vehicles at the end of the final assembly production stage during a preset time period, wherein a plurality of sample points are set on the body of each vehicle, and the final assembly gap flushness data of each sample point is identified by a vehicle identification number VIN and a sample point number. The preset time period may be a recent period in the current production environment, such as 3 months, 6 months, 1 year or a longer period.

[0058] Generally, multiple sample points on the vehicle body are selected according to factory specifications or customer requirements. Preferably, the selection of sample points is mainly concentrated on the positions between the components on the vehicle body.

[0059] Similarly, performing final assembly gap and flushness measurement on the selected sample point includes performing multiple gap and flushness measurements on the same sample point, and selecting an average value of the multiple final assembly gap and flushness measurement values ​​as final assembly gap and flushness data of the sample point.

[0060] Preferably, the selection of the vehicle body sample points during final assembly production may correspond to the selection of the vehicle body sample points during vehicle body production.

[0061] Preferably, the assembly gap and flushness data of the plurality of sample points may be stored in a local and / or remote server to facilitate viewing and retrieval via a network.

[0062] Preferably, the body gap face difference data is associated with the final assembly gap face difference data by the body identification number I-GEF, the vehicle identification number VIN and the sample point number. Furthermore, for the same sample point, there are associated body gap face difference data and final assembly gap face difference data.

[0063] Preferably, by machine learning, the final assembly gap flushness data is input into the initial gap flushness threshold model to train the initial gap flushness threshold model.

[0064] The following will be combined with the attached Figure 3 The method steps for training the initial gap and flushness threshold model using the final assembly gap and flushness data accumulated during the final assembly production stage according to an embodiment of the present application are described in detail.

[0065] Next, in step 106, the trained initial gap and flushness threshold model is evaluated through machine learning to obtain a gap and flushness threshold prediction model that meets preset conditions.

[0066] According to one embodiment of the present application, satisfying the preset condition means that when the initial gap flushness threshold model is trained using the final assembly gap flushness data, the output result (prediction result) of the initial gap flushness threshold model indicates that not only the vehicle body gap flushness data satisfies the vehicle body gap flushness threshold range, but also the final assembly gap flushness data satisfies the final assembly gap flushness threshold range. This means that the currently set vehicle body gap flushness threshold is appropriate. The determination result of step 108 is "yes", so the process proceeds to step 110.

[0067] Otherwise, when the determination result in step 108 is "No", proceed to step 109. In step 109, the initial gap flushness threshold model is modified, and further returns to step 104, and the initial gap flushness threshold model is trained again using the final assembly gap flushness data through machine learning until a gap flushness threshold prediction model that meets the preset conditions is obtained.

[0068] The following will be combined with the attached Figure 4 The method steps for evaluating the trained initial gap flushness threshold model and modifying it accordingly according to an embodiment of the present application are described in detail.

[0069] Next, at step 112 , the gap and flushness threshold prediction model that meets preset conditions is used to set or modify the body gap and flushness threshold in the body production stage.

[0070] Preferably, the initial gap flushness threshold output by the initial gap flushness threshold model that meets preset conditions may be set as the vehicle body gap flushness threshold.

[0071] Preferably, the range of the initial gap flushness threshold value can be further narrowed to further improve the accuracy of the gap flushness and improve the gap flushness quality of the vehicle body appearance.

[0072] Next reference Figure 2 , which schematically shows a flow chart of a method for constructing an initial gap flushness threshold model according to an embodiment of the present application. According to an embodiment of the present disclosure, an initial gap flushness threshold model is constructed using body gap flushness data accumulated during a body production stage.

[0073] First, in step 202, preferably, the vehicle body gap flushness data is preprocessed. A variety of known algorithms can be used to preprocess the vehicle body gap flushness data. Preferably, the preprocessing of the vehicle body gap flushness data may include removing outliers, filling missing values, normalizing, and standardizing.

[0074] Optionally, feature analysis and extraction may be performed on the preprocessed vehicle body gap flushness data (step 204) to further improve the efficiency of machine learning.

[0075] Next, in step 206, the preprocessed body gap flushness data is compared with a predetermined body gap flushness threshold value. The predetermined body gap flushness threshold value may be a quality specification requirement of the body production stage. The body gap flushness data that does not meet the body gap flushness threshold range is removed to improve the accuracy of the initial gap flushness threshold model to be constructed.

[0076] Then, at step 208 , the vehicle body gap flushness data satisfying the vehicle body gap flushness threshold range is used to construct an initial gap flushness threshold model.

[0077] In addition, optionally, in step 210, sample points corresponding to the body gap flushness data used to construct the initial gap flushness threshold model may be marked. Such sample points may be referred to as "qualified sample points". A list of all qualified sample points in the body production stage may be saved as a collection file in a local and remote server to facilitate subsequent review and retrieval.

[0078] Next reference Figure 3 , which schematically shows a flow chart of a method for training an initial gap flushness threshold model according to an embodiment of the present application. According to an embodiment of the present disclosure, the initial gap flushness threshold model is trained by machine learning using the final assembly gap flushness data accumulated during the final assembly production stage.

[0079] First, in step 302, preferably, the assembly gap flushness data is preprocessed. A variety of known algorithms can be used to preprocess the assembly gap flushness data. Preferably, the preprocessing of the assembly gap flushness data includes removing outliers, filling missing values, normalizing, and standardizing.

[0080] Optionally, feature analysis and extraction may be performed on the preprocessed assembly clearance and flushness data (step 304) to further improve the efficiency of machine learning.

[0081] Next, in step 306, the pre-processed final assembly gap flushness data is compared with a predetermined final assembly gap flushness threshold, wherein the predetermined final assembly gap flushness threshold may be a quality specification requirement of the final assembly production stage, a factory requirement, or a customer requirement.

[0082] Then, in step 308 , it is determined whether the final assembly gap flushness data is qualified. When the final assembly gap flushness data meets the final assembly gap flushness threshold range and is determined to be “qualified”, the process proceeds to step 310 , otherwise the process proceeds to step 312 .

[0083] In step 310, the assembly gap flushness data that meets the assembly gap flushness threshold range is used to train the initial gap flushness threshold model. Preferably, the sample points corresponding to the assembly gap flushness data that meets the assembly gap flushness threshold range are marked as "qualified sample points".

[0084] On the other hand, in step 312, the final assembly gap flushness data that does not meet the final assembly gap flushness threshold range is analyzed, and the sample points corresponding to the final assembly gap flushness data that does not meet the final assembly gap flushness threshold range are marked as "unqualified sample points".

[0085] Next, for the "unqualified sample point", the body gap face difference data and the final assembly gap face difference data corresponding to the "unqualified sample point" are obtained. It can be understood that the body gap face difference data and the final assembly gap face difference data corresponding to the "unqualified sample point" can be obtained through the association between the body identification number I-GEF, the vehicle identification number VIN and the sample point number.

[0086] In addition, preferably, "unqualified sample points" can be marked and selected, that is, sample points corresponding to the final assembly gap flushness data that do not meet the final assembly gap flushness threshold range. For the "unqualified sample points", the correlation between the vehicle body gap flushness data and the final assembly gap flushness data of the unqualified sample points is analyzed, and based on the correlation analysis results, the initial gap flushness threshold model is further modified or adjusted to obtain a gap flushness threshold prediction model that meets the preset conditions. The following is combined with Figure 4The method steps for evaluating the initial gap flushness threshold model using the final assembly gap flushness data (preferably, using the final assembly gap flushness data of “unqualified sample points”) according to one embodiment of the present application are described in detail.

[0087] In addition, preferably, the lists of all "qualified sample points" and "unqualified sample points" in the final assembly production stage can be saved as separate collection files on local and remote servers to facilitate subsequent review and retrieval.

[0088] According to one embodiment, preferably, for sample points that are "qualified" in both the body production stage and the final assembly production stage, the number of measurements can be reduced to improve measurement efficiency. For sample points that are "qualified" in the body production stage but "unqualified" in the final assembly production stage, more attention can be paid to them, and the number of measurements for such sample points can be increased to improve measurement accuracy.

[0089] Next reference Figure 4 , which schematically shows a flow chart of a method for evaluating a trained initial gap flushness threshold model according to an embodiment of the present application. Specifically, Figure 4 The method steps of evaluating the initial gap and flushness threshold model using the final assembly gap and flushness data (preferably, using the final assembly gap and flushness data of “unqualified sample points”) are schematically shown.

[0090] In step 402, the final assembly gap flushness data is compared with a predetermined final assembly gap flushness threshold.

[0091] In step 404, sample points corresponding to the final assembly gap and flushness data that do not meet the final assembly gap and flushness threshold range are determined as “unqualified sample points”.

[0092] For the determined "unqualified sample point", the body gap face difference data and the final assembly gap face difference data corresponding to the "unqualified sample point" can be obtained. It can be understood that the body gap face difference data and the final assembly gap face difference data corresponding to the "unqualified sample point" can be obtained through the association between the body identification number I-GEF, the vehicle identification number VIN and the sample point number.

[0093] In step 406, for the determined "unqualified sample point", preferably, the probability distribution of the influence of the vehicle body gap flushness data of the unqualified sample point on the final assembly gap flushness data can be calculated.

[0094] Figure 5 A chart schematically shows a correlation analysis between vehicle body gap flushness and assembly gap flushness according to an embodiment of the present application.

[0095] Specifically, the body gap (BS_Gap) and body offset (BS_Offset) in the body (BS) production stage, and the assembly gap (AS_Gap) and assembly offset (AS_Offset) in the assembly (AS) production stage are mainly used in this analysis. 18 sample points (measurement points) on the body are mainly selected to measure the scatter distribution and regression trend of the correlation coefficients of these 18 sample points, thereby respectively obtaining the influence of the body gap (BS_Gap) and body offset (BS_Offset) in the body production stage on the assembly gap (AS_Gap) and assembly offset (AS_Offset) in the assembly production stage. It can be understood that more or fewer sample points can be selected.

[0096] from Figure 5 From the analysis of factors affecting the gap and flushness in Figure 1-6, it can be concluded that:

[0097] Chart 1 is an analysis of the impact of BS_Gap on AS_Gap. It can be seen from the figure that the correlation coefficient is above 0.2, which means the correlation is high. Every vehicle is produced from BS to AS, and after going through a complex production process, the inventor found that the gap value Gap at the time of initial installation, that is, the Gap value at the end of BS workshop installation, has an impact factor of more than 0.2 on the final result value, indicating that the influencing factor is relatively large. Therefore, the inventor believes that if other operations are normal, appropriately adjusting the BS_Gap error range will help improve the final quality.

[0098] Chart 2 is an analysis of the impact of BS_Offset on AS_Gap. It can be found that the correlation coefficient fluctuates around 0, which means that the impact of BS_Offset value on AS_Gap is very small, and the correlation coefficient is low, almost 0.

[0099] Chart 3 shows the impact of BS_Offset on AS_Offset. It can be seen from the figure that there is a correlation, but the distribution of the correlation coefficient is slightly smaller than the impact between the gaps of the two workshops in Chart 1, and the correlation coefficient is low.

[0100] Chart 4 shows the impact of BS_Gap on AS_Offset, which also fluctuates around 0 and has a low correlation coefficient, indicating almost no impact.

[0101] Chart 5 shows the correlation coefficient between the Gap and Offset values ​​in the AS workshop, which is also around 0, indicating no correlation.

[0102] Figure 6 shows the correlation coefficient between the Gap and Offset values ​​in the BS workshop, which is also around 0 and has no correlation.

[0103] The process returns to Figure 4In step 408, based on the probability distribution of the influence of the determined vehicle body gap flushness data of the unqualified sample points on the final assembly gap flushness data, an adjustment amount for adjusting the vehicle body gap flushness threshold is determined.

[0104] After completing the correlation analysis between the body gap flushness and the assembly gap flushness, the BS_Gap threshold range of the sample points with a high correlation coefficient (for example, greater than 0.2) will be appropriately adjusted. Preferably, the adjustment amount of the BS_Gap threshold range is determined based on the probability distribution of the correlation effect (the influence of the body gap flushness data of the unqualified sample points on the assembly gap flushness data) to improve the accuracy of the gap flushness of the final vehicle leaving the factory and the quality of the product.

[0105] Further, in step 410, the determined adjustment amount of the vehicle body gap flushness threshold may be applied to the initial gap flushness threshold model to obtain a gap flushness threshold prediction model that meets preset conditions.

[0106] According to the present invention, by adjusting the BS_Gap threshold range, such as further reducing the predetermined BS_Gap threshold by an adjustment amount, the "unqualified sample points" in the final assembly production stage can be reduced.

[0107] refer to Figure 6 To further illustrate the method for updating the gap flushness threshold prediction model based on the correlation analysis between the vehicle body gap flushness and the assembly gap flushness according to an embodiment of the present application.

[0108] Figure 6 A flowchart of a method for updating a gap and flushness threshold prediction model according to an embodiment of the present application is schematically shown.

[0109] In step 602 , a preset threshold value of a vehicle body gap flushness expected to be used in a vehicle body production stage is input into a gap flushness threshold prediction model.

[0110] It is understandable that in the vehicle body production stage, according to the quality specification requirements, a threshold value is set for the gap flushness of the vehicle body appearance, that is, the vehicle body gap flushness preset threshold value. This vehicle body gap flushness preset threshold value is input as an input into the constructed gap flushness threshold prediction model.

[0111] In step 604, the vehicle body gap flushness preset threshold is adjusted according to the prediction output result of the gap flushness threshold prediction model.

[0112] As described above, sample points on the vehicle body are selected, and the correlation coefficient scatter distribution and regression trend between the vehicle body gap flushness and the assembly gap flushness of each sample point are calculated to determine the correlation between the vehicle body gap flushness and the assembly gap flushness, and find out the influencing probability distribution with a high correlation coefficient, such as BS_Gap and AS_Gap, which is used as the prediction output result of the gap flushness threshold prediction model to adjust the preset threshold of the vehicle body gap flushness.

[0113] Next, at step 606, the gap flushness threshold prediction model is updated based on the adjusted vehicle body gap flushness preset threshold. The updated gap flushness threshold prediction model can be used for prediction or adjustment of a newly input gap flushness.

[0114] The method for optimizing the gap and flushness of the vehicle body appearance according to the present disclosure mainly utilizes data mining and machine learning technology. The machine learning includes supervised machine learning algorithms, and the supervised machine learning algorithms may include: decision trees and random forests, support vector machines (SVM), naive Bayes method, K-nearest neighbors (KNN) algorithm, neural network, linear regression and logistic regression algorithms. The machine learning also includes unsupervised machine learning algorithms, and the unsupervised machine learning algorithms include: clustering, visualization and dimensionality reduction, association rule learning and other algorithms.

[0115] According to the present disclosure, data mining and machine learning technologies are fully utilized, the body gap face difference in the body production stage and the assembly gap face difference in the assembly production stage are comprehensively considered, and a gap face difference model suitable for actual production use is constructed.

[0116] Moreover, when constructing the gap flushness model, the body gap flushness threshold in the body production stage is optimized and adjusted by analyzing the final assembly gap flushness data in the final assembly production stage and considering the influence of the body gap flushness data on the final assembly gap flushness data.

[0117] The gap and flushness threshold prediction model constructed according to the present disclosure can predict in advance, adjust in real time, and provide rapid feedback, thereby improving the accuracy of setting the gap and flushness threshold, reducing the probability of unqualified products in the final assembly production stage, and improving production efficiency and product quality.

[0118] Figure 7 An exemplary configuration of a computing device 700 capable of implementing an embodiment of the present disclosure is schematically shown.

[0119] The computing device 700 is an example of a hardware device to which the above aspects of the present disclosure can be applied. The computing device 700 can be any machine configured to perform processing and / or computing. The computing device 700 can be, but is not limited to, a workstation, a server, a desktop computer, a laptop computer, a tablet computer, a personal data assistant (PDA), a smart phone, a vehicle-mounted computer, or a combination thereof.

[0120] like Figure 7 As shown, the computing device 700 may include one or more elements that can be connected or communicated with the bus 702 via one or more interfaces. The bus 702 may include, but is not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MCA) bus, an Enhanced ISA (EISA) bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus, etc. The computing device 700 may include, for example, one or more processors 704, one or more input devices 707, and one or more output devices 708. The one or more processors 704 may be any kind of processor, and may include, but is not limited to, one or more general-purpose processors or special-purpose processors (such as special-purpose processing chips). The processor 704 may be configured to implement a classification method for user data according to various embodiments of the present disclosure, for example. The input device 706 may be any type of input device capable of inputting information to the computing device, and may include, but is not limited to, a mouse, a keyboard, a touch screen, a microphone, and / or a remote controller. The output device 708 may be any type of device capable of presenting information, and may include, but is not limited to, a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer.

[0121] The computing device 700 may also include or be connected to a non-transitory storage device 714, which may be any non-transitory storage device that can implement data storage, and may include, but is not limited to, a disk drive, an optical storage device, a solid-state memory, a floppy disk, a flexible disk, a hard disk, a tape, or any other magnetic medium, a compressed disk or any other optical medium, a cache memory, and / or any other storage chip or module, and / or any other medium from which a computer can read data, instructions, and / or code. The computing device 700 may also include a random access memory (RAM) 710 and a read-only memory (ROM) 712. The ROM 712 may store programs, utilities, or processes to be executed in a non-volatile manner. The RAM 710 may provide volatile data storage and store instructions related to the operation of the computing device 700. The computing device 700 may also include a network / bus interface 716 coupled to a data link 718. The network / bus interface 716 may be any type of device or system capable of enabling communication with an external device and / or a network, and may include but is not limited to a modem, a network card, an infrared communication device, a wireless communication device, and / or a chipset (such as a Bluetooth TM equipment, 802.11 equipment, WiFi equipment, WiMax equipment, cellular communication facilities, etc.).

[0122] The present disclosure also provides a computer-readable storage medium having executable instructions. When the instructions are executed, the instructions prompt a computer to execute the method for optimizing the gap and flushness of a vehicle body appearance according to the above aspects of the present disclosure.

[0123] The present disclosure also provides a device for optimizing the gap and flushness of a vehicle body appearance, the device comprising:

[0124] - a memory storing computer executable instructions; and

[0125] - a processor configured to execute computer executable instructions, wherein when the computer executable instructions are executed by the processor, the processor executes the method for optimizing the gap and flushness of the vehicle body appearance according to the above aspect of the present disclosure.

[0126] The present disclosure also provides a system for optimizing the gap and flushness of a vehicle body appearance, the system comprising:

[0127] - an industrial robot, taking an image of the vehicle body appearance and acquiring gap and flushness data of the vehicle body appearance;

[0128] - an apparatus for optimizing the gap flushness of a vehicle body appearance according to the above aspect of the present disclosure; and

[0129] -An alarm device is used to output an alarm signal when it is identified that the vehicle body gap difference threshold is exceeded.

[0130] The present disclosure also provides a vehicle production line, including the system for optimizing the gap and flushness of a vehicle body appearance according to the above aspects of the present disclosure.

[0131] For clarity of explanation, in some cases, the present technology may be presented as including individual functional blocks including functional blocks comprising devices, device components, steps or routines in a method implemented in software, or a combination of hardware and software.

[0132] The present invention may be a computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions thereon for causing a processor to execute various aspects of the present invention.

[0133] A computer readable storage medium may be a tangible device that can retain and store instructions for use by an instruction execution device. A computer readable storage medium may be, for example but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing.

[0134] The computer readable program instructions may execute entirely on the terminal's computer, partly on the terminal's computer, as a separate software package, partly on the terminal's computer and partly on a remote computer, or entirely on a remote computer or server.

[0135] Therefore, a computer-readable storage medium may be machine-readable or computer-readable. Therefore, in some embodiments, a computer-readable storage medium includes a data carrier having executable instructions that can cooperate with a programmable computer system or a programmable hardware component to implement one of the methods described herein. Therefore, one embodiment is a data carrier, a digital storage medium, or a computer-readable storage medium, on which a program for implementing one of the methods described herein is recorded.

[0136] Furthermore, another embodiment is a data stream, a signal sequence or a signal sequence, which is a program for implementing one of the methods described herein. The data stream, signal sequence or signal sequence can be configured, for example, for transmission via a data communication connection, for example via the Internet or another network. Thus, an embodiment can also be a signal sequence representing data, which is suitable for transmission via a network or a data communication connection, wherein the data is a program.

[0137] The method according to the above example can be implemented using computer executable instructions stored or otherwise obtainable from a computer readable medium. Such instructions may include, for example, instructions and data that cause a general-purpose computer, a special-purpose computer, or a special-purpose processing device to perform a certain function or a group of functions or otherwise configure a general-purpose computer, a special-purpose computer, or a special-purpose processing device to perform a certain function or a group of functions. The part of the computer resources used can be accessed through a network. Computer executable instructions can, for example, be binary, intermediate format instructions, such as assembly language, firmware, or source code. Examples of computer readable media that can be used to store instructions, information used, and / or information created during the method according to the described example include disks or optical disks, flash memory, USB devices provided with non-volatile memory, networked storage devices, etc.

[0138] The apparatus for implementing the methods according to these disclosures may include hardware, firmware and / or software, and may adopt any of a variety of rules. Typical examples of such specifications include laptop computers, mobile phones, small-format personal computers, personal digital assistants, etc. The functions described herein may also be implemented in peripherals or plug-in cards. By way of further example, such functions may also be implemented on a circuit board between different chips or different processes executed in a single device.

[0139] The instructions, the media for transmitting such instructions, the computing resources for executing them, and other structures for supporting such computing resources are the means for providing the functionality described in these disclosures.

[0140] This paper describes various aspects of the present invention with reference to the flow chart and / or block diagram of the method, system and computer program product according to embodiments of the present invention.It should be understood that each box in the flow chart and / or block diagram and the combination of boxes in the flow chart and / or block diagram can be implemented by computer readable program instructions.

[0141] The present invention is not limited to the embodiments shown, but includes or extends to all technical equivalents that can fall within the scope of validity. The selected position descriptions in the description, such as top, bottom, left, right, etc., refer to the direct description and the drawings shown and can be transferred to the new position according to the meaning when the position changes.

[0142] The features disclosed in the present application may be essential and may be implemented both individually and in any combination for the realization of the exemplary embodiments in different configurations.

[0143] Although some aspects are described in association with devices, it should be understood that these aspects are also descriptions of corresponding methods, so that a device of a module or a device of a system can also be understood as a corresponding method step or a feature of a method step. Similarly, aspects described in association with a method step or as a method step are also descriptions of corresponding modules or details or features of corresponding devices.

[0144] Although various examples and other information are used to illustrate various aspects, no limitation should be implied based on the specific features or arrangements in such examples, as those of ordinary skill will be able to use these examples to derive a variety of implementations. Further, although a subject matter may have been described in language specific to examples of structural features and / or method steps, it is understood that the subject matter defined is not necessarily limited to these described features or actions. For example, such functionality may be distributed differently or performed in components other than those identified herein. Instead, the features and steps described are disclosed as examples of components of systems and methods within the scope.

[0145] Although the disclosed examples have been fully described with reference to the accompanying drawings, it should be noted that various changes and modifications will become apparent to those skilled in the art. It should be understood that such changes and modifications are considered to be included within the scope of the disclosed examples defined by the accompanying drawings.

Claims

1. A method for optimizing the gap and flushness of a vehicle body appearance, comprising: constructing an initial gap flushness threshold model using body gap flushness data accumulated during the body production phase; performing pre-processing on the final assembly gap and flushness data accumulated during the final assembly production phase; Comparing the preprocessed assembly gap flushness data with a predetermined assembly gap flushness threshold; Through machine learning, the final assembly gap and flushness data that meets the final assembly gap and flushness threshold range is used to train the initial gap and flushness threshold model; Performing correlation analysis on assembly gap flushness data that does not meet the assembly gap flushness threshold range and vehicle body gap flushness data of corresponding sample points set on the vehicle body; The sample points corresponding to the assembly gap and flushness data that do not meet the assembly gap and flushness threshold range are determined as unqualified sample points; For the determined unqualified sample points, calculate the probability distribution of the influence of the body gap flushness data corresponding to the unqualified sample points on the assembly gap flushness data; Based on the determined probability distribution, determining an adjustment amount for adjusting a vehicle body gap flushness threshold; Applying the determined adjustment amount of the vehicle body gap flushness threshold to the initial gap flushness threshold model through machine learning to obtain a gap flushness threshold prediction model that meets preset conditions; as well as Use the gap and flushness threshold prediction model that meets the preset conditions to set the body gap and flushness threshold in the body production stage.

2. The method according to claim 1, The body gap and flushness data accumulated during the body production phase include: Body gap flushness data measured for a plurality of vehicles at the end of a body production stage during a preset time period, wherein a plurality of sample points are provided on the body of each vehicle, and the body gap flushness data of each sample point is identified by a body identification number I-GEF and a sample point number; The final assembly gap flushness data accumulated during the final assembly production stage includes: final assembly gap flushness data measured for a plurality of vehicles at the end of the final assembly production stage during a preset time period, wherein a plurality of sample points are set on the body of each vehicle, and the final assembly gap flushness data of each sample point is identified by a vehicle identification number VIN and a sample point number; and The body gap difference data is associated with the final assembly gap difference data through the body identification number I-GEF, the vehicle identification number VIN and the sample point number.

3. The method according to claim 1 or 2, further comprising: Perform pre-processing on body gap and flushness data; comparing the preprocessed body gap flushness data with a predetermined body gap flushness threshold; as well as The vehicle body gap flushness data satisfying the vehicle body gap flushness threshold range is used to construct an initial gap flushness threshold model.

4. The method according to claim 3, The preprocessing of the vehicle body gap difference data includes removing outliers, filling missing values, normalizing, and standardizing; Performing feature analysis and extraction on the pre-processed vehicle body gap flushness data; and Label the sample points corresponding to the vehicle body gap flushness data used to construct the initial gap flushness threshold model.

5. The method according to claim 1, The preprocessing of the assembly gap and flushness data includes removing outliers, filling missing values, normalizing, and standardizing; Perform feature analysis and extraction on the pre-processed assembly gap and flushness data.

6. The method according to claim 1, further comprising: Mark and select sample points corresponding to the assembly gap flushness data that do not meet the assembly gap flushness threshold range; as well as For the marked and selected sample points, based on the correlation analysis results between the body gap flushness data and the assembly gap flushness data of the selected sample points, the body gap flushness threshold prediction model is used to reset the body gap flushness threshold in the body production stage.

7. The method according to claim 1, further comprising: Obtain and save a list of unqualified sample points as a reference for subsequent prediction of the body gap flushness threshold.

8. The method according to claim 1, further comprising: Inputting a preset body gap flushness threshold value expected to be used in the body production stage into the gap flushness threshold prediction model; Adjusting the preset threshold of the vehicle body gap flushness according to the prediction output result of the gap flushness threshold prediction model; as well as Based on the adjusted vehicle body gap flushness preset threshold, the gap flushness threshold prediction model is updated.

9. The method of claim 1 or 2, wherein the machine learning comprises a supervised machine learning algorithm, and wherein the supervised machine learning algorithm comprises: Decision trees and random forests, support vector machines (SVMs), naive Bayes, K-nearest neighbors, neural networks, linear regression, and logistic regression.

10. The method of claim 1 or 2, wherein the machine learning comprises an unsupervised machine learning algorithm, and wherein the unsupervised machine learning algorithm comprises: Clustering, visualization and dimensionality reduction, and association rule learning. 11 . A computer-readable storage medium having executable instructions, which, when executed, cause a computer to execute the method for optimizing the gap and flushness of a vehicle body appearance according to any one of claims 1 to 10.

12. A device for optimizing the gap and flushness of a vehicle body appearance, the device comprising: - a memory storing computer executable instructions; as well as - a processor configured to execute computer executable instructions, wherein when the computer executable instructions are executed by the processor, the processor executes the method for optimizing the gap and flushness of the vehicle body appearance according to any one of claims 1 to 10.

13. A system for optimizing the gap and flushness of a vehicle body appearance, the system comprising: - an industrial robot, taking an image of the vehicle body appearance and acquiring gap and flushness data of the vehicle body appearance; - a device for optimizing the gap flushness of a vehicle body appearance according to claim 12; and -An alarm device is used to output an alarm signal when it is identified that the vehicle body gap difference threshold is exceeded.

14. A vehicle production line comprising the system for optimizing the gap and flushness of a vehicle body appearance according to claim 13.

Images