Control parameter optimization method for automobile polishing and waxing machine

Abstract

The invention provides a control parameter optimization method for an automobile polishing and waxing machine, and relates to the technical field of data processing, and the method comprises the steps: carrying out the material recognition and stain evaluation based on the automobile surface spectrum data and image information, obtaining the material feature information and stain distribution feature information, and carrying out the polishing and waxing scheme matching; performing polishing and waxing monitoring based on the initial polishing and waxing scheme, and performing deviation analysis to determine deviation adjustment pace parameters; and adjusting the initial polishing and waxing parameter scheme to obtain a plurality of polishing and waxing parameter schemes, carrying out optimization analysis, and determining a target polishing and waxing parameter scheme to carry out polishing and waxing control. By means of the automobile polishing and waxing method and device, the technical problem that the automobile polishing and waxing quality is poor due to the fact that the fineness and accuracy of automobile surface data analysis are insufficient and the polishing and waxing control parameters cannot be adjusted and optimized in time according to the actual effect in the polishing and waxing process can be solved, and the high-quality automobile polishing and waxing effect can be achieved.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular to a control parameter optimization method for an automobile polishing and waxing machine. Background Art

[0002] Car polishing and waxing are important parts of car beauty, and their main purpose is to restore and enhance the glossiness of the car surface while protecting the paint surface from damage by the external environment. Existing car polishing and waxing methods usually analyze the characteristics of the car paint surface based on manual experience, and set the polishing and waxing control plan for equipment control based on the analysis results, resulting in insufficient accuracy in the setting of the polishing and waxing control plan, which affects the quality of polishing and waxing.

[0003] At present, the existing automobile polishing and waxing methods have technical problems such as poor quality of automobile polishing and waxing due to insufficient precision and accuracy in analyzing automobile surface data and the inability to timely adjust and optimize polishing and waxing control parameters according to actual results during the polishing and waxing process. Summary of the invention

[0004] The purpose of this application is to provide a control parameter optimization method for a car polishing and waxing machine, so as to solve the technical problem that the existing car polishing and waxing methods have poor quality due to insufficient precision and accuracy in the analysis of car surface data and the inability to timely adjust and optimize the polishing and waxing control parameters according to the actual effects during the polishing and waxing process.

[0005] In view of the above problems, the present application provides a control parameter optimization method for a car polishing and waxing machine, the method comprising: S1: obtaining a car sensing detection module, the car sensing detection module comprising a spectral sensor and a CMOS image device, and collecting and obtaining the car surface spectral data and car surface image information of the target vehicle through the car sensing detection module; S2: performing material identification and stain evaluation based on the car surface spectral data and car surface image information, obtaining car material feature information and car stain distribution feature information, performing polishing and waxing scheme matching analysis based on the car material feature information and car stain distribution feature information, and obtaining an initial polishing and waxing parameter scheme; S3: performing polishing and waxing monitoring on the target vehicle based on the initial polishing and waxing parameter scheme through the car polishing and waxing machine, obtaining a vehicle polishing and waxing control data stream, performing deviation analysis on the vehicle polishing and waxing control data stream, and determining a deviation adjustment step parameter; S4: adaptively adjusting the initial polishing and waxing parameter scheme based on the deviation adjustment step parameter, obtaining multiple polishing and waxing parameter schemes, performing optimization analysis based on the multiple polishing and waxing parameter schemes, and determining a target polishing and waxing parameter scheme for vehicle polishing and waxing optimization control.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: By acquiring an automobile sensing detection module, the automobile sensing detection module includes a spectral sensor and a CMOS image device, and the automobile surface spectral data and automobile surface image information of the target vehicle are collected and acquired through the automobile sensing detection module; material identification and stain evaluation are performed based on the automobile surface spectral data and automobile surface image information, and automobile material characteristic information and automobile stain distribution characteristic information are acquired; polishing and waxing scheme matching analysis is performed based on the automobile material characteristic information and automobile stain distribution characteristic information to obtain an initial polishing and waxing parameter scheme; polishing and waxing monitoring is performed on the target vehicle based on the initial polishing and waxing parameter scheme by an automobile polishing and waxing machine, and a vehicle polishing and waxing control data stream is acquired, and a deviation analysis is performed on the vehicle polishing and waxing control data stream to determine a deviation adjustment pace parameter; the initial polishing and waxing parameter scheme is adaptively adjusted based on the deviation adjustment pace parameter to obtain multiple polishing and waxing parameter schemes, and optimization analysis is performed based on the multiple polishing and waxing parameter schemes to determine a target polishing and waxing parameter scheme for vehicle polishing and waxing optimization control. That is to say, by constructing an automobile sensing detection module to collect data on the automobile surface of the target vehicle, the automobile surface spectral data and automobile surface image information are obtained; further, material recognition and stain evaluation are performed based on the automobile surface spectral data and automobile surface image information to obtain automobile material characteristic information and automobile stain distribution characteristic information, which can improve the precision and accuracy of automobile surface data analysis; then, based on the automobile material characteristic information and automobile stain distribution characteristic information, a polishing and waxing scheme matching analysis is performed to obtain an initial polishing and waxing parameter scheme, which can improve the adaptability of the initial polishing and waxing parameter scheme to the automobile surface characteristics; then, based on the automobile material characteristic information and automobile stain distribution characteristic information, a polishing and waxing scheme matching analysis is performed to obtain an initial polishing and waxing parameter scheme, which can improve the adaptability of the initial polishing and waxing parameter scheme to the automobile surface characteristics. When the target vehicle is polished and waxed using the initial polishing and waxing parameter scheme, the polishing and waxing effect is monitored in real time, and a deviation analysis is performed on the effect monitoring results. Based on the deviation analysis results, the initial polishing and waxing parameter scheme is adaptively adjusted to obtain multiple polishing and waxing parameter schemes. Finally, an optimization analysis is performed based on the multiple polishing and waxing parameter schemes to obtain a target polishing and waxing parameter scheme for optimal control of vehicle polishing and waxing, which can improve the precision and accuracy of automobile surface data analysis. At the same time, the polishing and waxing effect is monitored in real time, and the polishing and waxing control parameters are optimized in time according to the deviation of the waxing effect, thereby achieving high-quality automobile polishing and waxing effects.

[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the present application or 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 in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0009] Figure 1 A schematic flow chart of a control parameter optimization method for a car polishing and waxing machine according to the present application; Figure 2 A schematic diagram of a process for obtaining automobile material characteristic information and automobile stain characteristic information in a control parameter optimization method for an automobile polishing and waxing machine in the present application. DETAILED DESCRIPTION

[0010] The present application provides a control parameter optimization method for a car polishing and waxing machine, which solves the technical problem that the existing car polishing and waxing methods have poor quality of car polishing and waxing due to insufficient precision and accuracy of car surface data analysis and inability to timely adjust and optimize polishing and waxing control parameters according to actual effects during the polishing and waxing process. The precision and accuracy of car surface data analysis can be improved, and at the same time, the polishing and waxing effects can be monitored in real time, and the polishing and waxing control parameters can be optimized in time according to the deviation of the waxing effect, so as to achieve high-quality car polishing and waxing effects.

[0011] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0012] For examples, please see the attached Figure 1The present application provides a control parameter optimization method for a car polishing and waxing machine, the method specifically comprising the following steps: S1: Acquire a vehicle sensing detection module, wherein the vehicle sensing detection module includes a spectral sensor and a CMOS image device, and collects and acquires vehicle surface spectral data and vehicle surface image information of a target vehicle through the vehicle sensing detection module.

[0013] Specifically, first, a car sensing detection module is constructed, wherein the car sensing detection module includes a spectral sensor and a CMOS image device, wherein the spectral sensor is a device for measuring a spectrum, and its principle is based on the difference in wavelength and intensity of light emitted or absorbed by an object or substance, and the composition and properties of the object or substance are obtained through spectrometer sampling and analysis; the CMOS image device is an image acquisition device manufactured using CMOS technology, which is used for image acquisition and has the advantages of small size and low power consumption. Then, the spectral data and image data of the target car surface are collected through the car sensing detection module to obtain the car surface spectral data and car surface image information of the target vehicle. By obtaining the car surface spectral data and car surface image information, support is provided for the next step of analyzing the car surface features and matching the appropriate polishing and waxing scheme.

[0014] S2: Perform material identification and stain assessment based on the automobile surface spectral data and automobile surface image information, obtain automobile material feature information and automobile stain distribution feature information, perform polishing and waxing scheme matching analysis based on the automobile material feature information and automobile stain distribution feature information, and obtain an initial polishing and waxing parameter scheme.

[0015] Specifically, material recognition and stain assessment are performed based on the automobile surface spectral data and automobile surface image information, wherein material recognition refers to the recognition of the automobile paint material and surface structure characteristics, including paint type, paint color, paint composition, automobile surface texture characteristics, etc.; stain assessment refers to determining the stain distribution state on the automobile surface, the size of the stain area, the severity of the stain, etc.; and obtaining automobile material feature information and automobile stain distribution feature information. Then, based on the automobile material feature information and automobile stain distribution feature information, a polishing and waxing scheme matching analysis is performed to determine the initial polishing and waxing parameter scheme. By determining the initial polishing and waxing parameter scheme based on the analysis of the automobile material feature information and automobile stain distribution feature information, the adaptability of the initial polishing and waxing parameter scheme to the automobile surface characteristics can be improved, thereby improving the automobile polishing and waxing quality.

[0016] S3: Performing polishing and waxing monitoring on the target vehicle based on the initial polishing and waxing parameter scheme through a car polishing and waxing machine, obtaining a vehicle polishing and waxing control data stream, performing deviation analysis on the vehicle polishing and waxing control data stream, and determining a deviation adjustment pace parameter.

[0017] Specifically, based on the initial polishing and waxing parameter scheme, the target vehicle is polished and waxed by a car polishing and waxing machine, and the car polishing and waxing effect is monitored in real time during the polishing and waxing process, that is, data collection is performed on the car polishing and waxing by the car sensing detection module, and the polishing and waxing effect is evaluated based on the data collection result, and the vehicle polishing and waxing control data stream is obtained, and the vehicle polishing and waxing control data stream refers to the feedback data stream of the polishing and waxing effect; the expected polishing and waxing effect of the initial polishing and waxing parameter scheme is obtained, for example: multiple effect evaluation indicators are set, including glossiness index, flatness index, uniformity index, etc. Then, based on the expected polishing and waxing effect, the vehicle polishing and waxing control data stream is subjected to deviation analysis, that is, the deviation between the actual polishing effect and the expected polishing effect is calculated, and the deviation adjustment step parameter is determined based on the deviation analysis result, and the deviation adjustment step parameter refers to the specific effect deviation value, including glossiness deviation, flatness deviation, etc. By real-time monitoring of the polishing and waxing process, performing polishing and waxing effect deviation analysis based on the monitoring results, and obtaining the deviation adjustment pace parameters, abnormal polishing and waxing deviations can be discovered in time, thus providing a basis for the subsequent adjustment and optimization of polishing and waxing control parameters.

[0018] S4: Adaptively adjust the initial polishing and waxing parameter scheme based on the deviation adjustment step parameter to obtain multiple polishing and waxing parameter schemes, perform optimization analysis based on the multiple polishing and waxing parameter schemes, and determine the target polishing and waxing parameter scheme to perform vehicle polishing and waxing optimization control.

[0019] Specifically, based on the deviation adjustment step parameter, a correlation analysis is performed on the waxing control parameters in the initial polishing and waxing parameter scheme, wherein the waxing control parameters include polishing wheel speed, waxing amount, working time, polishing pressure, etc., and the waxing control schemes with correlation in the initial polishing and waxing parameter scheme are adaptively adjusted to obtain multiple polishing and waxing parameter schemes. For example, if the deviation adjustment step parameter is insufficient glossiness, and the waxing control parameters related to glossiness are the waxing amount and working time, the waxing amount and working time can be appropriately increased according to the deviation status. The polishing and waxing effects of the multiple polishing and waxing parameter schemes are further evaluated, and the polishing and waxing parameter scheme corresponding to the optimal waxing evaluation effect is selected and set as the target polishing and waxing parameter scheme, wherein the polishing and waxing effect evaluation method can be set according to actual conditions, for example: a scheme quality evaluation model can be constructed based on a BP neural network, wherein the input data of the scheme quality evaluation model is the polishing and waxing parameter scheme, and the output data is the quality evaluation coefficient, wherein the quality evaluation coefficient represents the quality of polishing and waxing, and the larger the quality evaluation coefficient, the better the polishing and waxing quality; then historical polishing and waxing data is called to perform supervised training on the scheme quality evaluation model, and a convergent scheme quality evaluation model that meets the expected requirements is obtained for effect evaluation; finally, vehicle polishing and waxing optimization control is performed according to the target polishing and waxing parameter scheme.

[0020] Further, the initial polishing and waxing parameter scheme is obtained in S2, including: Constructing a polishing and waxing machine solution space, taking the automobile material feature information and the automobile stain distribution feature information as constraint parameters, performing feature matching with the polishing and waxing machine solution space, and obtaining a polishing and waxing machine solution selection interval; Obtain polishing and waxing effect evaluation rules, evaluate and fit the polishing and waxing machine solution space based on the polishing and waxing effect evaluation rules, and generate a polishing and waxing effect evaluation function; Based on the polishing and waxing effect evaluation function, evaluation, comparison and optimization are performed within the polishing and waxing machine scheme selection interval to obtain the initial polishing and waxing parameter scheme.

[0021] Specifically, a polishing and waxing machine solution space is constructed, wherein the polishing and waxing machine solution space includes multiple control parameter thresholds of the polishing and waxing machine, such as: polishing wheel speed threshold, waxing amount threshold, working time threshold, etc.; further, the automobile material feature information and the automobile stain distribution feature information are used as constraint parameters and input into the polishing and waxing machine solution space for feature matching to obtain a polishing and waxing machine solution selection interval, wherein the polishing and waxing machine solution selection interval refers to the control parameter adaptation threshold of the polishing and waxing machine.

[0022] Obtain polishing and waxing effect evaluation rules, which include multiple effect evaluation indicators, such as glossiness indicator, flatness indicator, etc. Then, evaluate and fit the polishing and waxing machine solution space based on the polishing and waxing effect evaluation rules, that is, evaluate and fit the polishing and waxing effect data in the solution space according to multiple effect evaluation indicators, and construct a polishing and waxing effect evaluation function, which includes multiple effect evaluation indicators and corresponding indicator weight ratios. Finally, randomly select a solution within the polishing and waxing machine solution selection interval, and evaluate and compare based on the polishing and waxing effect evaluation function to find the best solution. Based on the optimization result, obtain the initial polishing and waxing parameter solution, which is the polishing and waxing parameter solution with the highest polishing and waxing effect evaluation coefficient.

[0023] The above method can solve the technical problem of poor quality of automobile polishing and waxing caused by the existing automobile polishing and waxing methods due to insufficient precision and accuracy in automobile surface data analysis and the inability to timely adjust and optimize polishing and waxing control parameters according to actual effects during the polishing and waxing process. First, obtain an automobile sensing detection module, which includes a spectral sensor and a CMOS image device, and collect and obtain automobile surface spectral data and automobile surface image information of the target vehicle through the automobile sensing detection module; then, perform material identification and stain evaluation based on the automobile surface spectral data and automobile surface image information, obtain automobile material feature information and automobile stain distribution feature information, perform polishing and waxing scheme matching analysis based on the automobile material feature information and automobile stain distribution feature information, and obtain an initial polishing and waxing parameter scheme; then, perform polishing and waxing monitoring on the target vehicle based on the initial polishing and waxing parameter scheme through an automobile polishing and waxing machine, obtain a vehicle polishing and waxing control data stream, perform deviation analysis on the vehicle polishing and waxing control data stream, and determine a deviation adjustment pace parameter; finally, perform adaptive adjustment on the initial polishing and waxing parameter scheme based on the deviation adjustment pace parameter, obtain multiple polishing and waxing parameter schemes, perform optimization analysis based on the multiple polishing and waxing parameter schemes, and determine a target polishing and waxing parameter scheme for vehicle polishing and waxing optimization control. It can improve the precision and accuracy of automobile surface data analysis, and at the same time achieve high-quality automobile polishing and waxing effects by real-time monitoring of the polishing and waxing effects and timely optimizing the polishing and waxing control parameters according to the deviation of the waxing effects.

[0024] Further, as attached Figure 2 As shown, the step S2 acquires the automobile material feature information and the automobile stain feature information, including: S21: constructing a dual channel for automobile data processing, and performing mapping and association processing on the automobile surface spectral data and the automobile surface image information based on the dual channel for data processing to obtain automobile surface spectral feature information and automobile surface image feature information; Specifically, first, a dual channel for automobile data processing is constructed, and the dual channel for automobile data processing includes a spectral data processing channel and an image data processing channel; then, based on the spectral data processing channel and the image data processing channel, the automobile surface spectral data and the automobile surface image information are mapped and associated to obtain the automobile surface spectral feature information and the automobile surface image feature information.

[0025] Furthermore, the S21 constructs a dual channel for automobile data processing, including: According to the spectral data application standard and the image data application standard, a spectral data preprocessing program and an image data preprocessing program are respectively set; Perform channel preprocessing node configuration based on the spectral data preprocessing program and the image data preprocessing program to obtain spectral data preprocessing channel nodes and image data preprocessing channel nodes; Acquire a spectral feature extraction node, wherein the spectral feature extraction node extracts automobile surface related features through a recursive feature elimination algorithm, and constructs an image feature extraction node through convolutional neural network structure training; Based on the spectral data preprocessing channel node and the spectral feature extraction node, as well as the image data preprocessing channel node and the image feature extraction node, channel processing node configurations are performed respectively to construct the automobile data processing dual channel.

[0026] Specifically, the method for constructing a dual channel for automobile data processing is as follows: first, a spectral data preprocessing program is set based on a spectral data application standard, wherein the spectral data application standard refers to the expected quality standard of spectral data, including data format, data accuracy, etc., and can be set according to actual conditions; the spectral data preprocessing program refers to a preprocessing scheme for spectral data, such as: spectral data needs to be preprocessed by denoising, data cleaning, etc.; an image data preprocessing program is set based on an image data application standard, wherein the image data application standard includes image resolution, color, etc., and the image data preprocessing program includes filtering denoising, image enhancement, etc. Then, channel preprocessing node configuration is performed based on the spectral data preprocessing program and the image data preprocessing program, that is, corresponding processing nodes are set in the data processing dual channels, for example, data preprocessing is first performed when data is input into the channel, and a spectral data preprocessing channel node and an image data preprocessing channel node are obtained.

[0027] The recursive feature elimination algorithm is a feature selection algorithm that selects features by building a model (such as a decision tree or support vector machine). The main idea of ​​recursive feature elimination is to repeatedly build the model, select the best (or worst) feature each time, and remove it in subsequent iterations. This process continues until all features have been traversed. Then, based on the recursive feature elimination algorithm, the associated features of the car surface are extracted, that is, the peak value, valley value, reflectivity of a specific band, etc. of the spectral data are extracted through frequency domain features, and then the recursive feature elimination algorithm is used to screen out the associated features for the identification, classification or quantitative analysis of the car surface, which reflect the material and stain characteristics of the car surface; based on the feature extraction results, a spectral feature extraction node is constructed.

[0028] Then, an image feature extraction node is constructed based on a convolutional neural network. The image feature extraction node is a neural network model that can be iteratively optimized in machine learning and is obtained through supervised training with a sample training set. The input data of the image feature extraction node is a car surface image, and the output data is the associated features of the material and stain characteristics of the car surface. A sample data set is obtained to train the image feature extraction node to a convergence state.

[0029] Finally, a spectral data processing channel is formed based on the spectral data preprocessing channel node and the spectral feature extraction node, an image data processing channel is formed based on the image data preprocessing channel node and the image feature extraction node, and the automobile data processing dual channel is constructed based on the spectral data processing channel and the image data processing channel.

[0030] By building a dual channel for automobile data processing based on a recursive feature elimination algorithm and a convolutional neural network, the efficiency of spectral data and image data analysis and processing and the accuracy of feature extraction can be improved, thereby improving the efficiency and accuracy of automobile surface material recognition and stain assessment.

[0031] S22: Acquire a vehicle material database, and perform feature comparison with the vehicle material database based on the vehicle surface spectral feature information and the vehicle surface image feature information to obtain the vehicle material feature information; S23: performing stain feature analysis and marking on the automobile surface image information, obtaining a set of automobile surface stain regions of interest, and generating an automobile surface stain distribution map according to the set of automobile surface stain regions of interest; S24: verifying and evaluating the stain distribution map on the vehicle surface through the spectral characteristic information on the vehicle surface to determine the stain distribution characteristic information on the vehicle.

[0032] Specifically, a vehicle material database is obtained, wherein the vehicle material database includes multiple vehicle material information and corresponding automobile surface spectral feature information and automobile surface image feature information, wherein the vehicle material information and the automobile surface spectral feature information and the automobile surface image feature information correspond one to one, and the vehicle material database can be constructed based on the decision tree principle; then the automobile surface spectral feature information and the automobile surface image feature information are input into the vehicle material database for matching to obtain the automobile material feature information.

[0033] Then, based on the car surface image information, stain regions are identified and marked to obtain a set of car surface stain regions of interest, where the car surface stain regions of interest refer to stain regions on the car surface; then, a car surface stain distribution map is generated based on the car surface stain region of interest set. Further, the car surface stain distribution map is verified and evaluated based on the car surface spectral feature information, that is, the stain regions in the car surface stain distribution map are cross-verified based on the spectral features, and the car stain distribution feature information is determined based on the cross-validation results. Further, determining the automobile stain distribution characteristic information in S24 includes: Cross-validating the automobile surface stain distribution map through the automobile surface spectral feature information to obtain a stain validation distribution area set; Constructing a stain type label library, and using the stain type label library to label the stain verification distribution area set by type, to obtain a stain labeled distribution area set; Segment and mark the set of stain labeled distribution areas to obtain a set of stain segmentation area positioning features; A stain level assessment is performed based on the stain labeled distribution area set and the stain segmentation area positioning feature set to determine the automobile stain distribution feature information.

[0034] Specifically, the method for determining the characteristic information of automobile stain distribution is as follows: first, the automobile surface stain distribution map is cross-validated by the automobile surface spectral characteristic information to obtain a stain verification distribution area set, that is, a cross-comparison verification is performed in two ways, and the intersection area of ​​the two is taken as the stain area to determine the final verified and compared stain area set. By cross-validating the automobile surface stain distribution map based on the automobile surface spectral characteristic information, misjudgment can be avoided and the accuracy of stain area setting can be improved.

[0035] Construct a stain type label library, wherein the stain type label library stores multiple stain types and corresponding associated stain features, wherein the stain types include dust, oil stains, paint, glue, etc. Then, the stain area in the stain verification distribution area set is labeled with a type label by using the stain type label library to obtain a stain labeled distribution area set. Further, the stain labeled distribution area set area is segmented and marked, i.e., the size of the stain area and the position coordinates of the stain area are marked to obtain a stain segmentation area positioning feature set.

[0036] Finally, a stain level assessment is performed based on the stain labeled distribution area set and the stain segmentation area positioning feature set, that is, the stain degree is assessed based on the stain type, the size of the stain area and the importance of the stain area position, wherein the more difficult the stain type is to clean, the larger the stain area, the higher the importance of the stain area position, and the greater the difficulty of polishing and cleaning, the higher the corresponding stain level, and corresponding weight coefficients can be set for the stain type, the stain area and the importance of the stain area position respectively, wherein the greater the influence of the indicator on the overall stain treatment, the greater the corresponding weight coefficient; then, a weighted calculation is performed on the stain type, the size of the stain area and the importance of the stain area position according to the weight coefficient, and the stain level is set based on the weighted calculation result; and the automobile stain distribution feature information is obtained, wherein the automobile stain distribution feature information includes the size of the stain area, the location of the stain area and the stain level features.

[0037] Further, in S4, a plurality of polishing and waxing parameter schemes are obtained, including: S41: performing correlation process parameter analysis based on the deviation adjustment step parameter to obtain polishing and waxing correlation parameter information to be optimized; S42: Analyze the adjustment direction of the polishing and waxing related parameter information to be optimized, and determine the related parameter fine-tuning selection threshold; S43: randomly selecting a plurality of polishing and waxing process parameters within the associated parameter fine-tuning selection threshold, and evaluating the effects of the plurality of polishing and waxing process parameters using the polishing and waxing effect evaluation function to generate a plurality of polishing and waxing parameter effects; S44: clustering, adjusting and expanding the multiple polishing and waxing process parameters based on the multiple polishing and waxing parameter effects to obtain the multiple polishing and waxing parameter solutions.

[0038] Specifically, firstly, the associated process parameter analysis is performed based on the deviation adjustment step parameter, that is, the polishing and waxing parameter type associated with the failure to achieve the preset polishing and waxing effect is determined, and the polishing and waxing associated parameter information to be optimized is obtained. The adjustment direction of the polishing and waxing associated parameter information to be optimized is further analyzed, and the associated parameter fine-tuning selection threshold is determined in combination with the polishing and waxing machine solution selection space.

[0039] Then, multiple polishing and waxing process parameters are randomly selected within the associated parameter fine-tuning selection threshold, and the effects of the multiple polishing and waxing process parameters are evaluated based on the polishing and waxing effect evaluation function to obtain multiple polishing and waxing parameter effects; further, the multiple polishing and waxing process parameters are clustered, adjusted and expanded according to the multiple polishing and waxing parameter effects to obtain multiple polishing and waxing parameter schemes.

[0040] Further, the multiple polishing and waxing parameter schemes are obtained in S44, including: Determine a plurality of parameter cluster centers based on the plurality of polishing and waxing parameter effects, cluster the plurality of polishing and waxing process parameters according to the plurality of parameter cluster centers, and generate a plurality of polishing and waxing parameter clusters; Setting a parameter variation step length, and based on the parameter variation step length, performing variation update on the remaining polishing and waxing parameters in the multiple polishing and waxing parameter clusters toward the center of the multiple parameter clusters to obtain multiple polishing and waxing variation parameter clusters; The multiple polishing and waxing parameter clusters are randomly cross-expanded to obtain the multiple polishing and waxing parameter solutions.

[0041] Specifically, the method for obtaining multiple polishing and waxing parameter schemes is as follows: first, the multiple polishing and waxing process parameters are arranged from large to small according to the polishing and waxing parameter effects to construct a process parameter sequence; further, a preset number of process parameters in the process parameter sequence are selected as parameter cluster centers, and the remaining process parameters in the process parameter sequence are clustered according to the multiple parameter cluster centers to obtain multiple polishing and waxing parameter clusters.

[0042] The parameter variation step length is set, and the parameter variation step length includes multiple parameter adjustment steps, wherein each parameter corresponds to an adjustment step length, which can be set according to the parameter type; then, the parameter cluster center in the polishing and waxing parameter cluster is used as the optimization direction, and the remaining polishing and waxing parameters in the multiple polishing and waxing parameter clusters are adjusted based on the parameter variation step length to obtain multiple polishing and waxing variation parameter clusters. Then, the multiple polishing and waxing variation parameter clusters are randomly cross-expanded to obtain multiple polishing and waxing parameter solutions.

[0043] Further, the random cross-expansion of the plurality of polishing and waxing variation parameter clusters comprises: Based on the multiple polishing and waxing variation parameter effects of the multiple polishing and waxing variation parameter clusters, perform parameter ratio division within the cluster to obtain multiple optimal polishing and waxing parameters within the cluster; A cross-dimensional analysis is performed on the superior polishing and waxing parameters in the multiple clusters, a parameter dimension cross-rule is constructed, and based on the parameter dimension cross-rule, the multiple polishing and waxing variation parameter clusters are randomly cross-expanded.

[0044] Specifically, first, the polishing and waxing effect evaluation function is used to evaluate the effects of multiple polishing and waxing variation parameters of the multiple polishing and waxing variation parameter clusters; then, the parameter ratios within the cluster are divided based on the effects of multiple polishing and waxing variation parameters of the multiple polishing and waxing variation parameter clusters, and a certain proportion of better variation parameters are selected as the superior polishing and waxing parameters within the cluster to obtain multiple superior polishing and waxing parameters within the cluster. Further, a cross-dimensional analysis is performed on the multiple superior polishing and waxing parameters within the cluster, and a parameter dimension crossover rule is constructed, that is, the polishing and waxing parameters with better effects within the cluster are analyzed in parameter dimensions to be optimized. For example, if the waxing amount dimension in the parameter needs to be cross-optimized, the waxing amount dimension is cross-expanded; finally, the multiple polishing and waxing variation parameter clusters are randomly cross-expanded based on the parameter dimension crossover rule.

[0045] In summary, the control parameter optimization method for a car polishing and waxing machine provided in this application has the following technical effects: 1. By constructing an automobile sensing detection module, data is collected on the automobile surface of the target vehicle to obtain automobile surface spectral data and automobile surface image information; further, material recognition and stain evaluation are performed based on the automobile surface spectral data and automobile surface image information to obtain automobile material feature information and automobile stain distribution feature information, which can improve the precision and accuracy of automobile surface data analysis; then, polishing and waxing scheme matching analysis is performed based on the automobile material feature information and automobile stain distribution feature information to obtain an initial polishing and waxing parameter scheme, which can improve the adaptability of the initial polishing and waxing parameter scheme to the automobile surface characteristics; then, based on the initial When the target vehicle is polished and waxed according to the initial polishing and waxing parameter scheme, the polishing and waxing effect is monitored in real time, and the deviation analysis is performed on the effect monitoring results. Based on the deviation analysis results, the initial polishing and waxing parameter scheme is adaptively adjusted to obtain multiple polishing and waxing parameter schemes; finally, optimization analysis is performed based on multiple polishing and waxing parameter schemes to obtain the target polishing and waxing parameter scheme for vehicle polishing and waxing optimization control, which can improve the precision and accuracy of automobile surface data analysis. At the same time, by real-time monitoring of the polishing and waxing effect, and timely optimization of the polishing and waxing control parameters according to the deviation of the waxing effect, high-quality automobile polishing and waxing effect can be achieved.

[0046] 2. By building a dual-channel automotive data processing system based on a recursive feature elimination algorithm and a convolutional neural network, the efficiency of spectral data and image data analysis and processing and the accuracy of feature extraction can be improved, thereby improving the efficiency and accuracy of automotive surface material recognition and stain assessment.

[0047] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0048] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A control parameter optimization method for a car polishing and waxing machine, characterized in that: The method comprises: S1: Acquire a vehicle sensing detection module, wherein the vehicle sensing detection module includes a spectral sensor and a CMOS image device, and collect and acquire vehicle surface spectral data and vehicle surface image information of a target vehicle through the vehicle sensing detection module; S2: performing material identification and stain evaluation based on the automobile surface spectral data and automobile surface image information, obtaining automobile material feature information and automobile stain distribution feature information, performing polishing and waxing scheme matching analysis based on the automobile material feature information and automobile stain distribution feature information, and obtaining an initial polishing and waxing parameter scheme; S3: Performing polishing and waxing monitoring on the target vehicle based on the initial polishing and waxing parameter scheme by using a car polishing and waxing machine, obtaining a vehicle polishing and waxing control data stream, performing deviation analysis on the vehicle polishing and waxing control data stream, and determining a deviation adjustment step parameter; S4: Adaptively adjusting the initial polishing and waxing parameter scheme based on the deviation adjustment step parameter to obtain multiple polishing and waxing parameter schemes, performing optimization analysis based on the multiple polishing and waxing parameter schemes, and determining a target polishing and waxing parameter scheme to perform vehicle polishing and waxing optimization control; The initial polishing and waxing parameter scheme is obtained in S2, including: Constructing a polishing and waxing machine solution space, taking the automobile material feature information and the automobile stain distribution feature information as constraint parameters, performing feature matching with the polishing and waxing machine solution space, and obtaining a polishing and waxing machine solution selection interval; Obtain polishing and waxing effect evaluation rules, evaluate and fit the polishing and waxing machine solution space based on the polishing and waxing effect evaluation rules, and generate a polishing and waxing effect evaluation function; Based on the polishing and waxing effect evaluation function, evaluation, comparison and optimization are performed within the polishing and waxing machine scheme selection interval to obtain the initial polishing and waxing parameter scheme.

2. The method according to claim 1, characterized in that The step S2 acquires the vehicle material feature information and the vehicle stain feature information, including: S21: constructing a dual channel for automobile data processing, and performing mapping and association processing on the automobile surface spectral data and the automobile surface image information based on the dual channel for data processing to obtain automobile surface spectral feature information and automobile surface image feature information; S22: Acquire a vehicle material database, and perform feature comparison with the vehicle material database based on the vehicle surface spectral feature information and the vehicle surface image feature information to obtain the vehicle material feature information; S23: performing stain feature analysis and marking on the automobile surface image information, obtaining a set of automobile surface stain regions of interest, and generating an automobile surface stain distribution map according to the set of automobile surface stain regions of interest; S24: verifying and evaluating the stain distribution map on the vehicle surface through the spectral characteristic information on the vehicle surface to determine the stain distribution characteristic information on the vehicle.

3. The method according to claim 2, characterized in that The S21 constructs a dual channel for automobile data processing, including: According to the spectral data application standard and the image data application standard, a spectral data preprocessing program and an image data preprocessing program are respectively set; Perform channel preprocessing node configuration based on the spectral data preprocessing program and the image data preprocessing program to obtain spectral data preprocessing channel nodes and image data preprocessing channel nodes; Acquire a spectral feature extraction node, wherein the spectral feature extraction node extracts automobile surface related features through a recursive feature elimination algorithm, and constructs an image feature extraction node through convolutional neural network structure training; Based on the spectral data preprocessing channel node and the spectral feature extraction node, as well as the image data preprocessing channel node and the image feature extraction node, channel processing node configurations are performed respectively to construct the automobile data processing dual channel.

4. The method according to claim 3, characterized in that Determining the automobile stain distribution characteristic information in S24 includes: Cross-validating the automobile surface stain distribution map through the automobile surface spectral feature information to obtain a stain validation distribution area set; Constructing a stain type label library, and using the stain type label library to label the stain verification distribution area set by type, to obtain a stain labeled distribution area set; Segment and mark the set of stain labeled distribution areas to obtain a set of stain segmentation area positioning features; A stain level assessment is performed based on the stain labeled distribution area set and the stain segmentation area positioning feature set to determine the automobile stain distribution feature information.

5. The method according to claim 1, characterized in that In S4, a plurality of polishing and waxing parameter schemes are obtained, including: S41: performing correlation process parameter analysis based on the deviation adjustment step parameter to obtain polishing and waxing correlation parameter information to be optimized; S42: Analyze the adjustment direction of the polishing and waxing related parameter information to be optimized, and determine the related parameter fine-tuning selection threshold; S43: randomly selecting a plurality of polishing and waxing process parameters within the associated parameter fine-tuning selection threshold, and evaluating the effects of the plurality of polishing and waxing process parameters using the polishing and waxing effect evaluation function to generate a plurality of polishing and waxing parameter effects; S44: clustering, adjusting and expanding the multiple polishing and waxing process parameters based on the multiple polishing and waxing parameter effects to obtain the multiple polishing and waxing parameter solutions.

6. The method according to claim 5, characterized in that The multiple polishing and waxing parameter schemes are obtained in S44, including: Determine a plurality of parameter cluster centers based on the plurality of polishing and waxing parameter effects, cluster the plurality of polishing and waxing process parameters according to the plurality of parameter cluster centers, and generate a plurality of polishing and waxing parameter clusters; Setting a parameter variation step length, and based on the parameter variation step length, performing variation update on the remaining polishing and waxing parameters in the multiple polishing and waxing parameter clusters toward the center of the multiple parameter clusters to obtain multiple polishing and waxing variation parameter clusters; The multiple polishing and waxing parameter clusters are randomly cross-expanded to obtain the multiple polishing and waxing parameter solutions.

7. The method according to claim 6, characterized in that The randomly cross-expanding the plurality of polishing and waxing variation parameter clusters comprises: Based on the multiple polishing and waxing variation parameter effects of the multiple polishing and waxing variation parameter clusters, perform parameter ratio division within the cluster to obtain multiple optimal polishing and waxing parameters within the cluster; A cross-dimensional analysis is performed on the superior polishing and waxing parameters in the multiple clusters, a parameter dimension cross-rule is constructed, and based on the parameter dimension cross-rule, the multiple polishing and waxing variation parameter clusters are randomly cross-expanded.

Images