Sample enhancement method, system and storage medium for vehicle fault diagnosis
By constructing a classification model and clustering process for vehicle fault diagnosis, augmented data samples are formed, which solves the problem that normal samples are difficult to define or are relatively small in vehicle fault diagnosis, and improves the accuracy of fault prediction and diagnosis.
Patent Information
- Application Number
- CN202311287000.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-10-07
- Publication Date
- 2025-04-08
AI Technical Summary
There are difficulties in predicting and diagnosing vehicle failures because the failure occurs suddenly and the lack of data to characterize the failure signs, making it difficult to achieve pure manual analysis. The data-driven method relies on a certain amount of abnormal samples, and normal samples are difficult to define or have a small number of them.
By receiving basic data samples of vehicles that do not have faults in the vehicle drive system, a classification model is built to determine whether the vehicle has a risk of failure, classify vehicles through the model, receive actual verification results, cluster normal vehicles, and form enhanced data samples to supplement or replace basic data samples.
The accuracy of vehicle fault prediction and diagnosis is improved, the shortcomings of the original samples are made up, and the accuracy of fault recognition is improved by iteratively optimizing the samples and models.
Smart Images
Figure CN119782850A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of vehicle fault diagnosis, and in particular, to a method and system for sample enhancement and a storage medium for vehicle fault diagnosis. Background Art
[0002] The prediction and diagnosis of vehicle failures is a difficult problem today, because vehicle failures occur suddenly, and the collected data lacks information that characterizes the signs of failure. Therefore, it is difficult to predict and diagnose faults through purely manual analysis methods. On the other hand, by using a data-driven approach to compare and learn the differences in operating data under fault conditions and normal conditions, effective feature information before the vehicle fails can be obtained in some scenarios. However, the above method is effective only on the premise of having a certain amount of abnormal samples (for fault detection, samples under abnormal conditions are positive samples, and samples under normal conditions are negative samples). Most vehicles experience failures gradually after running for a period of time, so there may be a problem that normal samples are difficult to define or are too few.
[0003] In view of this, it is necessary to propose a sample enhancement method for vehicle fault diagnosis. Summary of the invention
[0004] Embodiments of the present application provide a method and system for sample enhancement for vehicle fault diagnosis, and a storage medium, for improving the accuracy of prediction and diagnosis of vehicle faults.
[0005] According to one aspect of the present application, a method for sample enhancement for vehicle fault diagnosis is provided, comprising the following steps: receiving a basic data sample from a vehicle whose vehicle drive system does not have a fault, wherein the basic data sample includes a plurality of sample features associated with the vehicle drive system; constructing a classification model for determining whether the vehicle has a risk of vehicle drive system failure based on the plurality of sample features; classifying the owned vehicles through the classification model to determine the risk vehicles with a risk of vehicle drive system failure among the owned vehicles; receiving the result of actual verification of the vehicle drive system of the risk vehicle, the result indicating a normal vehicle without the vehicle drive system failure among the risk vehicles; clustering the normal vehicles with respect to the plurality of sample features, and taking the normal vehicles in the set with the highest degree of similarity in the cluster as candidate vehicles; and forming an enhanced data sample for supplementing or replacing the basic data sample based on the plurality of sample features of the candidate vehicles.
[0006] In some embodiments of the present application, optionally, the vehicle whose vehicle drive system has no faults includes a vehicle whose time from the production date is less than a threshold.
[0007] In some embodiments of the present application, optionally, the vehicle drive system includes: a drive motor, a speed change mechanism, an inverter, and a DC / DC converter.
[0008] In some embodiments of the present application, optionally, constructing a classification model for determining whether a vehicle has a risk of vehicle drive system failure based on the multiple sample features includes: performing dimensionality reduction processing on the multiple sample features to obtain multiple model features; and using the multiple model features as negative sample inputs of the model to train to generate the classification model.
[0009] In some embodiments of the present application, optionally, dimensionality reduction processing is performed on the multiple sample features using an ensemble tree algorithm.
[0010] In some embodiments of the present application, optionally, clustering the normal vehicles with respect to the multiple sample features includes: performing nonlinear dimensionality reduction on the multiple sample features to obtain multiple reduced dimensionality features; and clustering the normal vehicles with respect to the multiple reduced dimensionality features through a density estimation algorithm, and determining the degree of similarity of each cluster.
[0011] In some embodiments of the present application, optionally, the method further includes the step of: using the enhanced data sample as a negative sample input of the classification model, and training to update the classification model.
[0012] According to another aspect of the present application, a system for sample enhancement for vehicle fault diagnosis is provided, comprising: a memory configured to store instructions; and a processor configured to execute the instructions so as to perform any one of the methods described above.
[0013] According to another aspect of the present application, a system for sample enhancement for vehicle fault diagnosis is provided, comprising: a receiving unit, configured to receive a basic data sample from a vehicle whose vehicle drive system does not have a fault, wherein the basic data sample includes a plurality of sample features associated with the vehicle drive system; a model unit, configured to construct a classification model for determining whether the vehicle has a risk of vehicle drive system failure based on the plurality of sample features; a classification unit, configured to classify the owned vehicles through the classification model to determine a risk vehicle with a risk of vehicle drive system failure among the owned vehicles; a verification unit, configured to receive a result of actual verification of the risk vehicle with respect to the vehicle drive system, the result indicating that there is no normal vehicle with the vehicle drive system failure among the risk vehicles; a clustering unit, configured to cluster the normal vehicles with respect to the plurality of sample features, and use the normal vehicles in the set with the highest degree of similarity in the cluster as candidate vehicles; and an enhancement unit, configured to form an enhanced data sample for supplementing or replacing the basic data sample based on the plurality of sample features of the candidate vehicle.
[0014] According to another aspect of the present application, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium, and wherein when the instructions are executed by a processor, the processor is caused to execute any one of the methods described above.
[0015] Some examples of this application aim at the problem of missing or insufficient negative samples in electric drive fault warning. Based on the model trained with existing samples, vehicles with higher similarity scores can be retrieved from vehicles of the same model as supplementary negative samples. In this way, more available samples can be obtained without additional cost investment when actual samples are missing or insufficient, and the algorithm's fault recognition accuracy can be iteratively improved based on the supplementary samples. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] The above and other objects and advantages of the present application will become more fully apparent from the following detailed description in conjunction with the accompanying drawings, wherein the same or similar elements are represented by the same reference numerals.
[0017] Figure 1 A method for sample enhancement for vehicle fault diagnosis according to an embodiment of the present application is shown;
[0018] Figure 2 A system for sample enhancement for vehicle fault diagnosis according to an embodiment of the present application is shown. DETAILED DESCRIPTION
[0019] For the purpose of brevity and illustration, the principles of the present application are described herein mainly with reference to its exemplary embodiments. However, those skilled in the art will readily recognize that the same principles are equally applicable to all types of sample enhancement methods and systems, storage media for vehicle fault diagnosis, and that these same or similar principles may be implemented therein, and any such changes do not depart from the true spirit and scope of the present application.
[0020] According to one aspect of the present application, a method for sample enhancement for vehicle fault diagnosis is provided. Figure 1 As shown, a method 10 for sample enhancement of vehicle fault diagnosis (hereinafter referred to as method 10) includes the following steps: receiving a basic data sample (including a plurality of sample features associated with the vehicle drive system) from a vehicle without a fault in the vehicle drive system in step S102; constructing a classification model for determining whether the vehicle has a risk of a vehicle drive system fault according to the plurality of sample features in step S104; classifying the owned vehicles by the classification model in step S106 to determine a risk vehicle with a risk of a vehicle drive system fault among the owned vehicles; receiving the result of an actual verification of the vehicle drive system of the risk vehicle in step S108, the result indicating that there is no normal vehicle with the vehicle drive system fault among the risk vehicles; clustering the normal vehicles with respect to the plurality of sample features in step S110 and taking the normal vehicles in the set with the highest similarity in the cluster as candidate vehicles; and forming an enhanced data sample for supplementing or replacing the basic data sample according to the plurality of sample features of the candidate vehicles in step S112. Method 10 can make up for the deficiency of the original sample through the above steps, so that the negative sample can be enhanced, and can also be iterated to achieve continuous optimization of the sample and the model. The working principle of each step in method 10 will be described in detail below.
[0021] In step S102, method 10 receives basic data samples from vehicles whose vehicle drive systems do not have faults. The basic data samples include multiple sample features associated with the vehicle drive system. The basic data samples here are originally collected negative samples that can be used for processing (for example, training a neural network model) to determine whether a vehicle exists. As mentioned above, due to the limitation of the data source, the basic data samples are generally small in scale, and therefore are not suitable for relying solely on the basic data samples to build a final model to predict or diagnose vehicle drive system faults. To this end, the present application also enhances the samples through other steps of method 10, which will be described in detail below.
[0022] As understood, the basic data samples are negative samples, and their source is a vehicle with no faults in the vehicle drive system. In some embodiments of the present application, the vehicle with no faults in the vehicle drive system includes a vehicle with a time threshold less than the production date, because new vehicles are generally not prone to drive system problems. In some other examples, the vehicle with no faults in the vehicle drive system may also be a vehicle that has been verified to be free of faults, for example, a vehicle that has been repaired after a recall, a vehicle that has completed maintenance, etc.
[0023] Method 10 can monitor one or more components in a vehicle drive system. In some embodiments of the present application, the vehicle drive system includes components such as a drive motor, a speed change mechanism, an inverter, and a DC / DC transformer. The basic data sample is associated with the vehicle drive system. Specifically, the basic data sample can be a physical quantity collected by sensors that are directly or indirectly coupled to these components. For example, it can be a physical quantity such as current, voltage, and temperature. The present application does not limit the form of the basic data sample. Generally speaking, the physical quantity that can be directly or indirectly affected by the device in the vehicle drive system can be used as a basic data sample. In some examples, some physical quantities can be added or subtracted from the basic data sample based on experience.
[0024] The basic data sample can be processed into multiple sample features according to the data source, etc. As understood, these sample features included in the basic data sample are also associated with the vehicle drive system. In other words, these sample features include / contain information about the working state of the vehicle drive system. The association relationship between these sample features and faults can be established through computer analysis methods such as neural networks, and this relationship can be expressed through various models as described below.
[0025] In step S104, method 10 constructs a classification model for determining whether a vehicle has a risk of a vehicle drive system failure based on multiple sample features. The present application does not limit the specific structure or working principle of the model, as long as it can be trained based on the sample features and the trained model can distinguish whether a vehicle has a risk of a drive system failure. For example, the model can be a neural network model, a support vector machine model, a decision tree model, etc.
[0026] In some embodiments of the present application, the dimension of the sample features can be reduced in step S104. In this process, noise features and invalid features can be eliminated, and the feature dimension can be reduced to enhance generalization. Specifically, in step S104, multiple sample features can first be subjected to dimensionality reduction processing to obtain multiple model features, and then the multiple model features can be used as negative sample input training of the model to generate a classification model. For example, multiple model features can be input into a neural network model, and the neural network model can be trained accordingly. The model trained according to the basic data sample will be used as the initial classification model for subsequent processing steps. In addition, as described in detail below, the classification model can also be further trained according to the enhanced data sample, so as to further improve the fault diagnosis accuracy of the model.
[0027] In some embodiments of the present application, in step S104, multiple sample features can be subjected to dimensionality reduction processing by using an ensemble tree algorithm. The ensemble tree algorithm is a commonly used machine learning algorithm, which is composed of a plurality of decision tree models. These decision tree models can be homogeneous or heterogeneous. Their prediction results are combined as the final output. The ensemble tree algorithm usually uses techniques such as bagging, boosting and random forest to construct the model, which can effectively improve the prediction performance and robustness of the model. Compared with a single decision tree, the ensemble tree algorithm has a stronger generalization ability.
[0028] In step S106, method 10 classifies the owned vehicles through the classification model to determine the risk vehicles with vehicle drive system failure risks among the owned vehicles. These "owned vehicles" can be vehicles of the same model and the same batch as the "vehicles without vehicle drive system failures". In some other examples, the vehicle drive systems of the "owned vehicles" and the "vehicles without vehicle drive system failures" are the same. In some other examples, the vehicle drive systems of the "owned vehicles" and the "vehicles without vehicle drive system failures" have some common components. Therefore, based on the feature association between the two, the basic data samples generated according to the "vehicles without vehicle drive system failures" will also be applicable to the "owned vehicles"; on the other hand, the data samples generated according to the "owned vehicles" will also be applicable to the "vehicles without vehicle drive system failures".
[0029] The introduction of "owned vehicles" can greatly expand the range of data available for sample collection, thus alleviating the problem of insufficient sample size. It is also important to screen the newly introduced samples, and for this purpose, method 10 will further select the usable part from "owned vehicles" through the following steps.
[0030] In step S108, the method 10 receives the result of the actual verification of the vehicle drive system of the risk vehicle, and the result indicates that there is no normal vehicle with a vehicle drive system failure among the risk vehicles. The risk vehicle determined in step S106 can be sent to the maintenance site by recall or the like, and the recall information can be sent to the vehicle owner by, for example, remote push, announcement, telephone notification, etc. Subsequently, the maintenance personnel can perform the actual verification process to determine whether the risk vehicle does have a drive system failure. The maintenance personnel can label the part of the risk vehicles that are determined to have no drive system failure as "normal vehicles" and export data samples of the "normal vehicles". In step S108, the label as the verification result and the data sample of the "normal vehicle" can be received and used in the subsequent steps of the method 10.
[0031] In step S110, method 10 clusters normal vehicles with respect to multiple sample features, and selects normal vehicles in the set with the highest similarity in the cluster as candidate vehicles. Although vehicles labeled as "normal vehicles" are confirmed to not include drive system failures, the performance of the data collected from these vehicles in terms of sample features is not necessarily suitable for training and updating models. For this reason, in step S110, the "normal vehicles" will be further screened in order to select a portion of vehicles with the highest sample feature similarity for training and updating models.
[0032] Specifically, clustering normal vehicles with respect to multiple sample features in step S110 may include the following steps: (1) First, nonlinear dimensionality reduction is performed on multiple sample features to obtain multiple reduced dimensionality features. The nonlinear dimensionality reduction may be different from the dimensionality reduction algorithm of step S104 above, so the features obtained may be different from the features before the nonlinear dimensionality reduction in form and physical sense. (2) Subsequently, the normal vehicles may be clustered with respect to multiple reduced dimensionality features by means of a density estimation algorithm, and the degree of similarity of each cluster may be determined. The working principles of the nonlinear dimensionality reduction algorithm and the density estimation algorithm may refer to the prior art, and will not be elaborated in this article due to space limitations.
[0033] In step S112, the method 10 forms an enhanced data sample for supplementing or replacing the basic data sample based on multiple sample features of the candidate vehicle. From this point on, the method 10 can start from the original "basic data sample" and use the "basic data sample" generated based on the candidate vehicle as the enhanced data sample. According to this process, the sample size will be expanded several times, dozens of times, hundreds of times, or even thousands of times.
[0034] Based on the enhanced negative samples, it is possible to iteratively upgrade the classification model, thereby improving the accuracy of the classification model. Figure 1The method 10 further includes step S114, in which the enhanced data sample can be used as a negative sample input of the classification model to train and update the classification model. The classification model can be deployed on the server side of the vehicle factory to achieve accurate early warning of vehicles that have failed or are about to fail, to prevent the situation from escalating due to the failure and affecting driving safety.
[0035] According to another aspect of the present application, a system for sample enhancement for vehicle fault diagnosis is provided, which includes: a memory configured to store instructions; and a processor configured to execute the instructions so as to perform any one of the methods described above.
[0036] According to another aspect of the present application, a system for sample enhancement for vehicle fault diagnosis is provided. Figure 2 As shown, a system 20 for sample enhancement for vehicle fault diagnosis (hereinafter referred to as system 20) includes a receiving unit 201, a model unit 202, a classification unit 203, a verification unit 204, a clustering unit 205 and an enhancement unit 206. The system 20 can make up for the deficiency of the original sample, so that the negative sample can be enhanced, and it can also be iterated to achieve continuous optimization of the sample and the model. The working principle of each part of the system 20 will be briefly described below. The more detailed working details can refer to the various steps of the method 10 mentioned above. Due to the limited space, this article will not repeat them here, and the relevant contents in the above text are quoted here.
[0037] The receiving unit 201 of the system 20 is configured to receive basic data samples from vehicles whose vehicle drive systems do not have faults, wherein the basic data samples include multiple sample features associated with the vehicle drive systems. As understood, the basic data samples are negative samples, and their source is a vehicle whose vehicle drive systems do not have faults. In some embodiments of the present application, vehicles whose vehicle drive systems do not have faults include vehicles whose time from the production date is less than a threshold. In some other examples, vehicles whose vehicle drive systems do not have faults may also be vehicles that have been verified to be free of faults, for example, vehicles that have been repaired after a recall, vehicles that have completed maintenance, etc.
[0038] The model unit 202 of the system 20 is configured to construct a classification model for determining whether a vehicle has a risk of a vehicle drive system failure based on a plurality of sample features. The present application does not limit the specific structure or working principle of the model, as long as it can be trained based on the sample features and the trained model can distinguish whether a vehicle has a risk of a drive system failure. For example, the model can be a neural network model, a support vector machine model, a decision tree model, etc.
[0039] The classification unit 203 of the system 20 is configured to classify the owned vehicles through the classification model to determine the risk vehicles with vehicle drive system failure risks among the owned vehicles. The introduction of "owned vehicles" can greatly expand the data range that can be used to collect samples, thereby alleviating the problem of insufficient sample size. It is also important to screen the newly introduced samples. For this purpose, the method 10 will further select the usable part from the "owned vehicles" through the following steps.
[0040] The verification unit 204 of the system 20 is configured to receive the result of the actual verification of the vehicle drive system of the risk vehicle, and the result indicates that there is no normal vehicle with a vehicle drive system failure in the risk vehicle. The risk vehicle determined by the classification unit 203 can be sent to the maintenance site by recall or the like, and the recall information can be sent to the vehicle owner by, for example, remote push, announcement, telephone notification, etc. Subsequently, the maintenance personnel can perform the actual verification process to determine whether the risk vehicle does have a drive system failure. The maintenance personnel can label the part of the risk vehicle that is determined to have no drive system failure as a "normal vehicle" and export the data sample of the "normal vehicle". The verification unit 204 can receive the label and the data sample of the "normal vehicle" as the verification result, and will be used in the subsequent steps of the method 10.
[0041] The clustering unit 205 of the system 20 is configured to cluster normal vehicles with respect to multiple sample features, and to select normal vehicles in the set with the highest similarity in the cluster as candidate vehicles. Specifically, the clustering unit 205 can cluster normal vehicles with respect to multiple sample features through the following steps: (1) First, nonlinear dimensionality reduction is performed on multiple sample features to obtain multiple reduced dimensionality features. (2) Subsequently, the normal vehicles can be clustered with respect to multiple reduced dimensionality features through a density estimation algorithm, and the similarity of each cluster can be determined. The working principles of the nonlinear dimensionality reduction algorithm and the density estimation algorithm can refer to the prior art, and will not be elaborated in this article due to space limitations.
[0042] The enhancement unit 206 of the system 20 is configured to form an enhanced data sample for supplementing or replacing the basic data sample based on multiple sample features of the candidate vehicle. The system 20 can start from the original "basic data sample" and use the "basic data sample" generated based on the candidate vehicle as the enhanced data sample. According to this process, the sample size will be expanded several times, dozens of times, hundreds of times, or even thousands of times.
[0043] Based on the enhanced negative samples, it is possible to iteratively upgrade the classification model, thereby improving the accuracy of the classification model. Figure 2The system 20 further includes an updating unit 207, which is configured to use the enhanced data sample as a negative sample input of the classification model for training to update the classification model. The classification model can be deployed on the server side of the vehicle factory to achieve accurate early warning of vehicles that have or are about to have a fault, to prevent the situation from escalating due to the fault and affecting driving safety.
[0044] According to another aspect of the present application, a computer-readable storage medium is provided, in which instructions are stored. When the instructions are executed by a processor, the processor executes any one of the sample enhancement methods for vehicle fault diagnosis as described above. The computer-readable medium referred to in the present application includes various types of computer storage media, which can be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, the computer-readable medium can include RAM, ROM, EPROM, EEPROM, etc. 2 PROM, register, hard disk, removable disk, CD-ROM or other optical disk storage, disk storage or other magnetic storage device, or can be used to carry or store the desired program code unit in the form of instruction or data structure and can be accessed by general or special computer, or general or special processor. As used herein, the disk usually copies data magnetically, while the dish uses laser to optically copy data. The above combination should also be included in the protection scope of computer-readable medium. Exemplary storage medium is coupled to the processor so that the processor can read and write information from / to the storage medium. In an alternative solution, the storage medium can be integrated into the processor. The processor and the storage medium can reside in the ASIC. The ASIC can reside in the user terminal. In an alternative solution, the processor and the storage medium can reside in the user terminal as discrete components.
[0045] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Those skilled in the art can think of other feasible changes or substitutions based on the technical scope disclosed in the present application, and such changes or substitutions are all included in the protection scope of the present application. In the absence of conflict, the implementation modes of the present application and the features in the implementation modes can also be combined with each other. The protection scope of the present application shall be subject to the description of the claims.
Claims
1. A sample enhancement method for vehicle fault diagnosis, characterized in that: The method comprises: Receiving a basic data sample from a vehicle having no fault in the vehicle drive system, wherein the basic data sample includes a plurality of sample features associated with the vehicle drive system; constructing a classification model for determining whether the vehicle has a risk of vehicle drive system failure based on the plurality of sample features; Classifying the owned vehicles by using the classification model to determine risky vehicles among the owned vehicles that have a risk of vehicle drive system failure; receiving a result of actual verification of the vehicle drive system on the risk vehicle, wherein the result indicates that there is no normal vehicle with a fault in the vehicle drive system among the risk vehicles; Clustering the normal vehicles with respect to the plurality of sample features, and taking the normal vehicles in the set with the highest similarity in the cluster as candidate vehicles; and An enhanced data sample for supplementing or replacing the basic data sample is formed according to the multiple sample features of the candidate vehicle.
2. The method according to claim 1, wherein: The vehicles whose vehicle drive systems do not have faults include vehicles whose time from production date is less than a threshold.
3. The method according to claim 1, wherein: The vehicle driving system includes: a driving motor, a speed changing mechanism, an inverter, and a DC / DC transformer.
4. The method according to claim 1, wherein: Constructing a classification model for determining whether a vehicle has a vehicle drive system failure risk according to the plurality of sample features includes: Performing dimensionality reduction processing on the multiple sample features to obtain multiple model features; and The multiple model features are used as negative sample inputs of the model for training to generate the classification model.
5. The method according to claim 4, wherein: The multiple sample features are subjected to dimensionality reduction processing by using an ensemble tree algorithm.
6. The method according to claim 1, wherein: Clustering the normal vehicles with respect to the plurality of sample features comprises: Performing nonlinear dimensionality reduction on the multiple sample features to obtain multiple dimensionality reduction features; and The normal vehicles are clustered with respect to the multiple dimensionality reduction features by using a density estimation algorithm, and the similarity of each cluster is determined.
7. The method according to claim 1, further comprising: The enhanced data samples are used as negative sample inputs of the classification model for training to update the classification model.
8. A sample enhancement system for vehicle fault diagnosis, characterized in that: The system comprises: a memory configured to store instructions; and A processor configured to execute the instructions so as to perform the method according to any one of claims 1 to 7.
9. A sample enhancement system for vehicle fault diagnosis, characterized in that: The system comprises: a receiving unit configured to receive a basic data sample from a vehicle having no fault in a vehicle drive system, wherein the basic data sample includes a plurality of sample features associated with the vehicle drive system; a model unit configured to construct a classification model for determining whether the vehicle has a risk of vehicle drive system failure based on the plurality of sample features; a classification unit configured to classify the owned vehicles by using the classification model to determine risky vehicles having a risk of vehicle drive system failure among the owned vehicles; a verification unit configured to receive a result of actual verification of the risk vehicle with respect to the vehicle drive system, the result indicating that there is no normal vehicle with a fault in the vehicle drive system among the risk vehicles; a clustering unit configured to cluster the normal vehicles with respect to the plurality of sample features, and to select the normal vehicles in a set with the highest similarity among the clusters as candidate vehicles; and An enhancement unit is configured to form an enhanced data sample for supplementing or replacing the basic data sample according to the plurality of sample features of the candidate vehicle.
10. A computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium, characterized in that: When the instructions are executed by a processor, the processor is caused to perform the method according to any one of claims 1 to 7.