Intelligent diagnosis method for vehicle and related device

Abstract

The application discloses a kind of intelligent diagnosis method of vehicle and related device, the method includes: receiving the dashboard image of vehicle uploaded by user;Vehicle feature of the dashboard image is extracted;Obtain the state information of the vehicle in first time period;It is detected that the vehicle feature includes the first feature of characterizing vehicle attribute, the second feature of characterizing vehicle fault state and the third feature of characterizing vehicle operating state, then according to the state information and the vehicle feature, determine target fault type;According to the target fault type, determine target analysis agent;According to target analysis agent and the vehicle feature, determine the fault diagnosis report of the vehicle;According to the fault diagnosis report, generate fault handling strategy.The application can reduce diagnostic cost, improve diagnostic efficiency.

Description

Technical Field

[0001] This application relates to the field of fault diagnosis technology, and in particular to an intelligent diagnostic method and related device for vehicles. Background Technology

[0002] The development of automotive fault diagnosis technology reflects a profound shift from traditional manual testing to intelligent diagnosis. Currently, it is mainly conducted through on-board diagnostic (OBD) devices or remote diagnosis by experienced technicians. However, OBD diagnostic devices are expensive and require professional operation, making them difficult for ordinary users to use. Remote diagnosis, on the other hand, relies on human experience, cannot obtain real-time vehicle status information, and the diagnostic process is cumbersome, making it difficult to guarantee the accuracy and efficiency of repair recommendations. Summary of the Invention

[0003] This application provides a vehicle intelligent diagnostic method and related apparatus to reduce diagnostic costs and improve diagnostic efficiency.

[0004] In a first aspect, embodiments of this application provide an intelligent diagnostic method for a vehicle, comprising:

[0005] Extract vehicle features from the dashboard image of the vehicle uploaded by the user;

[0006] Obtain the vehicle's status information within a first time period, which includes a first time period before the fault time and a second time period after the fault time;

[0007] If the vehicle features are detected to include a first feature characterizing vehicle attributes, a second feature characterizing vehicle fault status, and a third feature characterizing vehicle operating status, then the target fault type is determined based on the status information and the vehicle features.

[0008] Determine the target analysis agent based on the target fault type;

[0009] Based on the target analysis agent and the characteristics of the vehicle, a fault diagnosis report for the vehicle is determined;

[0010] A fault handling strategy is generated based on the fault diagnosis report.

[0011] The step of determining the target fault type based on the status information and the vehicle characteristics includes:

[0012] Based on the first feature and the second feature, at least one first fault type is determined;

[0013] Based on the second feature and the third feature, at least one second fault type is determined;

[0014] Based on the status information, the target fault type is determined from the at least one first fault type and the at least one second fault type.

[0015] The step of determining the target fault type among the at least one first fault type and the at least one second fault type based on the status information includes:

[0016] Determine the user's driving characteristics;

[0017] Determine the degree of matching between the driving feature and each of the at least one first fault types to obtain at least one first degree of matching; and determine the degree of matching between the driving feature and each of the at least one second fault types to obtain at least one second degree of matching.

[0018] Determine the matching degree between the status information and each of the first fault types to obtain at least one third matching degree; and determine the matching degree between the status information and each of the second fault types to obtain at least one fourth matching degree;

[0019] The target fault type is determined based on the at least one first matching degree, the at least one second matching degree, the at least one third matching degree, and the at least one fourth matching degree.

[0020] The step of determining the target fault type based on the at least one first matching degree, the at least one second matching degree, the at least one third matching degree, and the at least one fourth matching degree includes:

[0021] Obtain the maintenance data of the vehicle;

[0022] Based on the maintenance data, determine the first fault frequency of each first fault type and the second fault frequency of each second fault type within a preset historical period;

[0023] Based on the first fault frequency, the at least one third matching degree, and the at least one first matching degree, a probability value for each first fault type is determined; and based on the second fault frequency, the at least one second matching degree, and the at least one fourth matching degree, a probability value for each second fault type is determined.

[0024] The target fault type is determined based on the probability values ​​of each first fault type and each second fault type.

[0025] The method further includes:

[0026] If the vehicle features are detected to include the second feature and the third feature, then the vehicle identification image uploaded by the user is received;

[0027] Extract the features from the vehicle identification image to obtain the fourth feature;

[0028] The target fault type is determined based on the fourth feature, the second feature, the third feature, and the status information.

[0029] Before receiving the vehicle identification image uploaded by the user, the method further includes:

[0030] Determine at least one shooting position of the vehicle;

[0031] Determine the shooting strategy for each of the at least one shooting positions;

[0032] Based on each shooting location and the shooting strategy corresponding to each shooting location, guidance information is generated, which is used to guide the user to upload the vehicle identification image.

[0033] The step of determining the vehicle's fault diagnosis report based on the target analysis agent and the vehicle characteristics includes:

[0034] Obtain the vehicle's driving data and maintenance data;

[0035] The target analysis agent analyzes and processes the vehicle features, driving data, and maintenance data to obtain analysis results.

[0036] If the parsing result is used to indicate successful parsing, then the fault diagnosis report is generated;

[0037] If the parsing result is used to indicate parsing failure, then multiple reasons for the failure will be output.

[0038] Secondly, embodiments of this application provide an intelligent diagnostic device for a vehicle, comprising:

[0039] The extraction unit is used to extract vehicle features from the dashboard image of the vehicle uploaded by the user.

[0040] The acquisition unit is used to acquire the status information of the vehicle within a first time period, the first time period including a first time period before the fault time and a second time period after the fault time;

[0041] The first determining unit is configured to detect that the vehicle features include a first feature characterizing vehicle attributes, a second feature characterizing vehicle fault status, and a third feature characterizing vehicle operating status, and then determine the target fault type based on the status information and the vehicle features.

[0042] The second determining unit is used to determine the target analysis agent based on the target fault type;

[0043] The third determining unit is used to determine the vehicle's fault diagnosis report based on the target analysis agent and the vehicle's characteristics.

[0044] The generation unit is used to generate a fault handling strategy based on the fault diagnosis report.

[0045] Thirdly, embodiments of this application provide an electronic device including a memory, a processor, and executable program code stored in the memory and executable on the processor, wherein the processor executes the executable program code and performs the steps of the method as described in the first aspect.

[0046] Fourthly, embodiments of this application provide a computer-readable storage medium storing executable program code, the executable program code including execution instructions for performing the steps of the method as described in the first aspect.

[0047] Fifthly, embodiments of this application provide a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps described in the first aspect of embodiments of this application. The computer program product may be a software installation package.

[0048] As can be seen, in this embodiment, the system first receives a vehicle dashboard image uploaded by a user; then extracts vehicle features from the dashboard image; subsequently, it obtains the vehicle's status information within a first time period, which includes a first time period before the fault and a second time period after the fault; then, it detects that the vehicle features include a first feature representing vehicle attributes, a second feature representing vehicle fault status, and a third feature representing vehicle operating status; based on the status information and the vehicle features, it determines the target fault type; next, it determines the target analysis agent based on the target fault type; then, based on the target analysis agent and the vehicle features, it determines the vehicle's fault diagnosis report; finally, it generates a fault handling strategy based on the fault diagnosis report.

[0049] As can be seen, in the application, users only need to provide an image of the vehicle's dashboard. By extracting vehicle features from multiple dimensions through image recognition, fault diagnosis results can be obtained quickly without relying on professional equipment or technician experience, thereby reducing diagnostic costs and improving diagnostic efficiency and user experience. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a system architecture diagram of an intelligent diagnostic system provided in an embodiment of this application;

[0052] Figure 2 This is a schematic flowchart of an intelligent diagnostic method for a vehicle provided in an embodiment of this application;

[0053] Figure 3 This is a schematic diagram of a diagnostic operation interface provided in an embodiment of this application;

[0054] Figure 4 This is a schematic diagram of a fault diagnosis interface provided in an embodiment of this application;

[0055] Figure 5 This is a schematic diagram of a prompt window provided in an embodiment of this application;

[0056] Figure 6 This is a functional unit block diagram of an intelligent diagnostic device for a vehicle provided in an embodiment of this application;

[0057] Figure 7 This is a functional unit block diagram of another intelligent diagnostic device for vehicles provided in this application embodiment;

[0058] Figure 8 This is a schematic diagram of the structure of an electronic device proposed in an embodiment of this application. Detailed Implementation

[0059] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present application.

[0060] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0061] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0062] Current vehicle diagnostic methods rely excessively on OBD diagnostic equipment or other specialized diagnostic equipment, or remote diagnostics by experienced technicians. However, diagnostic equipment is expensive and requires professional operation, making it difficult for ordinary users to use independently. Furthermore, remote diagnostics, which depends on human experience and real-time data transmission from the equipment, cannot obtain the vehicle's dynamic operating status in real time, resulting in delays in the diagnostic process.

[0063] Meanwhile, existing diagnostic methods are mainly based on preset fault codes or human experience, making it difficult to fully utilize multi-source vehicle data, such as images, sensor data, and historical operating data, for comprehensive analysis. This results in limited accuracy of diagnostic results and a high risk of misdiagnosis or missed diagnosis.

[0064] Whether based on OBD diagnostics or remote diagnostics, both rely on experienced technicians to operate and interpret the diagnostic results. Ordinary car owners, lacking the professional knowledge and experience, find it difficult to independently diagnose vehicle faults.

[0065] To address the aforementioned problems, this application provides an intelligent diagnostic method and related apparatus for vehicles. The embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0066] Please see Figure 1 , Figure 1 This is a system architecture diagram of an intelligent diagnostic system provided in an embodiment of this application. Figure 1 As shown, the intelligent diagnostic system 100 includes a data acquisition layer 101, an image processing and recognition layer 102, a diagnostic analysis layer 103, and a user interaction layer 104. The data acquisition layer 101, image processing and recognition layer 102, diagnostic analysis layer 103, and user interaction layer 104 are interconnected.

[0067] The data acquisition layer 101 is responsible for acquiring relevant image data of the vehicle, such as dashboard images. Users can take photos of relevant parts of the vehicle using their mobile phones or other terminals; this image data will serve as the basis for diagnosis. It supports various shooting devices, such as mobile phones and tablets. It also provides shooting guidelines to help users obtain high-quality image data. Local storage and uploading of image data are supported.

[0068] The image processing and recognition layer 102 receives relevant vehicle image data acquired by the data acquisition layer 101 and performs operations such as denoising, enhancement, and cropping on the images to ensure that the image quality meets the recognition requirements. Then, a deep learning model is used to extract key features from the images, such as the shape, color, and position of vehicle components; and to identify and record fault codes and warning light statuses on the dashboard, as well as abnormal conditions of the vehicle's exterior and interior.

[0069] The diagnostic analysis layer 103 is used to receive the image recognition results output by the image processing and recognition layer 102. Based on the image recognition results, combined with the vehicle's historical data and operating status, and the fault information appearing on the dashboard, the intelligent agent accurately locates and analyzes the fault, thereby accurately locating the faulty component and fault type, generating a fault diagnosis report and fault handling strategy, and transmitting them to the user.

[0070] The user interaction layer 104 provides a user-friendly interface, allowing users to upload images and receive diagnostic results via mobile software or a web browser. It also supports real-time feedback and personalized suggestions to help users better understand the vehicle's status.

[0071] Based on this, this application provides an intelligent diagnostic method and related device for vehicles, which will be described in detail below with reference to the accompanying drawings.

[0072] Please see Figure 2 , Figure 2 This is a flowchart illustrating a vehicle intelligent diagnostic method provided in an embodiment of this application, as shown below. Figure 2 As shown, the method includes the following steps:

[0073] S210, extract vehicle features from the dashboard image of the vehicle uploaded by the user.

[0074] Users can access the vehicle diagnostic system via mobile app or web browser. If accessing via mobile app, users first need to search for the corresponding vehicle diagnostic software in their phone's app store, then download and install it. After installation, open the software, register an account, fill in relevant personal information, and set a password. After successful registration, log in to the software using the registered account and password. Third-party account login is also supported for convenient and quick access. Once logged in, the phone is connected to the associated vehicle using a specific method to obtain various sensor data, location data, and maintenance data.

[0075] These sensors can include engine run time sensors, door open / close sensors, key insertion / start sensors, and more. These sensors can record information such as the number of vehicle starts, engine runtime, and door opening / closing frequency to infer the vehicle's usage frequency.

[0076] After successful binding, the user will enter the main interface of the vehicle diagnostic system. On this interface, users can see various function options, such as fault diagnosis, real-time data monitoring, and historical diagnostic information, all presented in a simple and intuitive manner.

[0077] If accessing via the web, users can open any browser on a computer or mobile device and enter the vehicle diagnostic system's web address in the address bar. After entering the login page, users can log in using their registered account and password. For first-time users, registration is required; users must fill in the relevant information to complete the registration before logging in. After successful login, first-time users may need to bind associated vehicles to accurately identify them and provide corresponding diagnostic services. For subsequent users, the web interface leads to the vehicle diagnostic system's main interface, which may also display a list of the user's associated vehicles. Clicking on a vehicle's icon or name allows users to view its historical fault information, historical operating data, etc.

[0078] In one possible embodiment, please refer to Figure 3 , Figure 3 This is a schematic diagram of a diagnostic operation interface provided in an embodiment of this application, such as... Figure 3 As shown, the diagnostic operation interface may include personal center controls, fault diagnosis controls, data monitoring controls, etc.

[0079] Clicking the personal center control leads to the personal information interface, which includes a personal information display area showcasing the user's basic information, arranged in a list or card format. The personal information interface also includes a vehicle information management area, where users can view information about their linked vehicles. For example, multiple vehicle controls can be included, each corresponding to a linked vehicle. Each vehicle control can be named after a vehicle brand or icon. Clicking on a single vehicle control allows users to view detailed vehicle information, such as purchase date, mileage, and historical maintenance data.

[0080] Clicking the fault diagnosis control will take you to the fault diagnosis interface. Please refer to [link / reference]. Figure 4 , Figure 4 This is a schematic diagram of a fault diagnosis interface provided in an embodiment of this application, such as... Figure 4 As shown, the fault diagnosis interface includes a prompt message generation control and multiple upload controls. Clicking the prompt message generation control generates a prompt message indicating which vehicle parts need to be photographed, such as the dashboard, exterior, and interior. It also provides shooting guidelines to ensure image quality. Clicking the control with the "+" icon uploads the corresponding image.

[0081] The system supports local storage and instant upload functions.

[0082] Different types of images can correspond to different shooting guidelines. These guidelines can include information such as the shooting angle, shooting direction, and shooting details. For example, if the image is of a dashboard, the guide might be a prompt reminding the user to start the vehicle before shooting, ensuring the dashboard indicator lights and displays are illuminated to fully represent their operational status. It should also indicate that the entire dashboard, including the speedometer, tachometer, various indicator lights, and displays, must be captured in the image.

[0083] In one possible embodiment, the user takes photos of relevant parts of the vehicle using a mobile phone or other terminal and uploads them to the system via software or a web page.

[0084] After receiving the image uploaded by the user, the system will perform a quality assessment. If the quality assessment fails, a prompt window will pop up on the fault diagnosis interface to indicate that the image quality is unqualified and to point out the reason for the failure, such as that the image quality is unqualified or the relevant data of the tachometer is unclear.

[0085] In one possible embodiment, the uploaded image can be analyzed for sharpness, contrast, and completeness to determine whether the user-uploaded image is qualified.

[0086] Sharpness analysis can be performed by calculating the information entropy of an image, which measures the richness of information within the image. Images with high sharpness have higher information entropy because they contain more detail and variation. By calculating the information entropy, the overall sharpness of an image can be evaluated. For example, in a sharp dashboard image, elements such as the speedometer, tachometer, and various indicator lights are clearly displayed, resulting in a relatively high information entropy. After calculating the information entropy, it is compared with a first preset threshold to determine whether the image quality is acceptable.

[0087] Contrast analysis can be performed by combining local and global contrast. Specifically, for local contrast, for each pixel, the average or standard deviation of its grayscale difference with neighboring pixels is calculated as its local contrast value. For example, using a 3×3 neighborhood window, the grayscale difference between the center pixel and its eight surrounding pixels is calculated, and then the average is taken. If this value is large, it indicates that the local contrast of the region containing that pixel is high. For global contrast, the average grayscale value and the standard deviation of the grayscale values ​​of the entire image can be calculated. The average grayscale value reflects the overall brightness of the image, while the standard deviation reflects the dispersion of the grayscale values ​​relative to the average value. The larger the standard deviation, the higher the global contrast of the image. For example, for a bright dashboard image, its average grayscale value is high, but if the standard deviation is also large, it indicates that there are darker and brighter areas in the image, resulting in good overall contrast. Finally, the weights corresponding to each local and global contrast are determined, and a weighted sum is performed to obtain the contrast result of the image. After obtaining the contrast result, it is compared with a second preset threshold to determine whether the quality is acceptable.

[0088] This process involves checking whether the image completely contains the expected key feature elements for a completeness analysis. Specifically, based on task requirements, all necessary key feature elements are listed in detail. For example, in an instrument panel image, key feature elements might include the speedometer, tachometer, various indicator lights, and displays. The image is then divided into different regions, each corresponding to a set of key feature elements. Multi-source image comparisons are performed to determine whether the key feature elements within each region are complete. If any key feature element is incomplete, the image quality is deemed unacceptable.

[0089] In one possible implementation, based on the results of sharpness analysis, contrast analysis, and completeness analysis, it is determined whether the user needs to re-upload the image. If the sharpness is unsatisfactory, or the contrast is unsatisfactory, or a key feature-related element is incomplete, the shooting guidelines are updated according to the analysis results, guiding the user to re-upload a high-quality image to meet the image recognition requirements.

[0090] Clicking the data monitoring control leads to the data display interface, which includes real-time data and historical data controls. Clicking the real-time data control displays real-time data from various vehicle sensors; clicking the historical data control allows users to query data records from a past period. A calendar control at the top of the interface allows users to specify the query period, accurate to the day or month, by selecting a date. After selecting the query period, the data display area below presents the data records for that period in a list format, including the timestamp of each record, the values ​​of various data parameters, and any anomaly markers. Users can further click on a record to view more detailed data information, such as data change curves at specific times and correlation analysis with other data. When a user clicks on a record to view detailed information, a new window pops up or displays detailed data charts and analysis results at the bottom of the original interface.

[0091] In one possible embodiment, the user-uploaded image undergoes preprocessing and feature processing. Specifically, the image quality assessment result can be obtained, and the image can be preprocessed based on the quality assessment result. For example, this may include operations such as noise reduction, enhancement, and cropping to ensure that the image quality meets the recognition requirements.

[0092] In one possible embodiment, if the quality assessment result indicates that the sharpness is acceptable but does not meet the image recognition requirements, image enhancement processing can be performed. For example, the Laplacian operator can be used to calculate the difference between each pixel in the image and its surrounding pixels to highlight the edges and details of the image and improve its recognizability. Noise reduction processing can also be performed, such as mean filtering, which replaces the value of the current pixel with the average value of its neighboring pixels, thereby reducing random noise in the image. Image scaling and cropping can also be performed.

[0093] Key features, such as the shape, color, and location of vehicle components, can be extracted from preprocessed images using deep learning models. This allows for the identification and recording of fault codes and warning light statuses on the dashboard, as well as abnormal conditions of the vehicle's exterior and interior.

[0094] S220, Obtain the status information of the vehicle during the first time period.

[0095] The first time period includes a first time period before the fault time and a second time period after the fault time.

[0096] The status information may include driving information, operation information, and environmental information before the time of the fault, as well as information on changes in vehicle performance and driving performance-related information after the time of the fault, such as whether the vehicle's speed is affected after the fault occurs, whether there is a failure to accelerate, unstable speed, or automatic deceleration.

[0097] S230, if the vehicle features are detected to include a first feature characterizing vehicle attributes, a second feature characterizing vehicle fault status, and a third feature characterizing vehicle operating status, then the target fault type is determined based on the status information and the vehicle features.

[0098] When extracting features from the dashboard image, if the first feature, the second feature, and the third feature are extracted, then the user does not need to upload additional images for recognition.

[0099] The first feature is used to characterize information such as the vehicle's brand, model, and year. The second feature is used to characterize information such as fault codes and warning light status. The third feature is used to characterize the vehicle's operating status, such as engine speed and coolant temperature.

[0100] In one possible embodiment, determining the target fault type based on the status information and the vehicle characteristics includes: determining at least one first fault type based on the first characteristic and the second characteristic; determining at least one second fault type based on the second characteristic and the third characteristic; and determining the target fault type among the at least one first fault type and the at least one second fault type based on the status information.

[0101] In one possible embodiment, the first feature and the second feature are fused to obtain a first comprehensive feature, and then the first comprehensive feature is analyzed to output all possible fault types.

[0102] In one possible embodiment, the third feature and the second feature are fused to obtain a second comprehensive feature, and then the second comprehensive feature is analyzed to output all possible fault types.

[0103] In one possible embodiment, fault types are initially screened based on a first characteristic. Brand-specific issues may exist; different brands of vehicles, due to differences in design philosophy, technological level, and quality control, will have their own common fault types. Faults are then associated with vehicle model characteristics, and fault types are screened based on the vehicle's purpose and structural design. The model year is also considered; as vehicle technology is continuously updated and improved over time, earlier models may have design flaws or immature technological applications. For example, the fuel injection system design of some older models may not be precise enough, potentially leading to severe carbon buildup and excessive fuel consumption.

[0104] The second feature is used to locate the fault type. Specifically, the fault code is interpreted, as each fault code has a specific meaning, corresponding to a specific problem in a vehicle system or component. For example, fault code P0301 indicates a fault in the engine's first cylinder ignition system, which may be due to a problem with the spark plug, ignition coil, or related circuitry; fault code P0420 indicates that the catalytic converter efficiency is below the threshold, which may be due to a blockage in the catalytic converter itself or a faulty oxygen sensor.

[0105] After reading the fault code, the possible fault type is determined by combining it with the status of the warning light.

[0106] The various warning lights on the vehicle's dashboard can visually reflect the vehicle's health status and the types of malfunctions present. For example, an illuminated engine warning light may indicate a serious engine malfunction, such as misfire, overheating, or low oil pressure; a illuminated brake warning light indicates a problem with the braking system, such as insufficient brake fluid, excessively worn brake pads, or electrical leakage in the braking system. Different colors and shapes of warning lights also have different meanings: red warning lights usually indicate a serious emergency requiring immediate stopping and inspection; yellow warning lights indicate certain malfunctions or abnormalities that require attention but do not necessarily require immediate stopping.

[0107] In one possible embodiment, the fault type can be identified by a third feature, which can be identified from multiple aspects, including power performance, braking performance, driving stability, etc.

[0108] In terms of power performance, this involves determining if there is a power malfunction, such as insufficient acceleration or lack of power, and then identifying the possible types of malfunctions. For example, if a vehicle suddenly loses power or becomes unable to move while driving, it could be due to an engine mechanical malfunction or an electrical system malfunction.

[0109] Regarding braking performance, the process involves determining if there are any brake malfunctions, such as whether the braking distance has increased, and then identifying the possible types of malfunctions. For example, if the braking force feels insufficient and the braking distance is significantly longer, it could be due to insufficient brake fluid or poor brake fluid quality; it could also be due to severely worn brake pads; or uneven wear or deformation of the brake discs.

[0110] Regarding driving stability, it involves determining whether the vehicle exhibits noticeable vibrations during idling or driving to identify potential fault types. For example, idling vibrations might be due to damaged engine mounts, failing to effectively cushion engine vibrations; it could also be caused by carbon buildup on the throttle body, leading to abnormal throttle opening and affecting intake air volume and mixture formation. Vibrations during driving could be due to tire dynamic imbalance issues; or driveshaft imbalance, etc.

[0111] The process involves cross-referencing the fault types selected based on the first feature with the fault types determined based on the second feature. If there is overlap, these overlapping fault types are considered to have a higher probability of occurrence.

[0112] The process involves cross-referencing the fault types selected based on the third feature with those determined based on the second feature. If there is overlap, these overlapping fault types are considered to have a higher probability of occurrence.

[0113] In one possible embodiment, determining the target fault type based on the status information among the at least one first fault type and the at least one second fault type includes: determining the user's driving characteristics; determining the matching degree between the driving characteristics and each of the at least one first fault type to obtain at least one first matching degree; and determining the matching degree between the driving characteristics and each of the at least one second fault type to obtain at least one second matching degree; determining the matching degree between the status information and each of the first fault types to obtain at least one third matching degree; and determining the matching degree between the status information and each of the second fault types to obtain at least one fourth matching degree; and determining the target fault type based on the at least one first matching degree, the at least one second matching degree, the at least one third matching degree, and the at least one fourth matching degree.

[0114] This process involves acquiring the user's historical driving data, extracting features from the historical driving data, and outputting the user's driving characteristics, which may include vehicle operation-related features, such as the frequency and average acceleration of rapid acceleration behavior, and the frequency and average deceleration of sudden braking behavior.

[0115] Rapid acceleration typically refers to the operation of quickly pressing the accelerator pedal within a short period of time to rapidly increase the vehicle's speed. Rapid acceleration frequency is the number of rapid accelerations recorded per unit time. Average acceleration amplitude is the average acceleration value during each acceleration operation, reflecting the driver's degree of control over the accelerator pedal during acceleration. Emergency braking refers to the operation of suddenly and forcefully pressing the brake pedal to rapidly decelerate or stop the vehicle. Emergency braking frequency is used to record the number of emergency brakings per unit time. Average deceleration amplitude is the average deceleration value during each deceleration operation, reflecting the driver's operational force during deceleration.

[0116] Driving characteristics can also include trajectory-related features, such as average driving speed and speed change frequency. Average driving speed is the average speed of the user on different road sections or throughout the entire journey, reflecting the overall driving rhythm. Speed ​​change frequency is the number of times the speed changes per unit time, reflecting the frequency with which the driver adjusts the speed during driving.

[0117] Driving characteristics can also include time-related characteristics, such as travel time patterns and vehicle usage frequency.

[0118] Driving characteristics can also include safety-related characteristics, such as the number of speeding violations recorded, the number of minor accidents, minor accident damage to vehicle parts, the number of major accidents, and major accident damage to vehicle parts in the accident history. Specifically, the number of speeding violations refers to the number of times the user exceeded the speed limit while driving; the number of minor accidents refers to the number of minor traffic accidents the user was involved in, such as scrapes and collisions, and records the parts of the vehicle damaged in those minor accidents; the number of major accidents refers to the number of major traffic accidents the user was involved in, such as rollovers and serious collisions, and records the parts of the vehicle damaged in those major accidents.

[0119] In one possible embodiment, the time-related features, safety-related features, driving trajectory-related features, and vehicle operation-related features described above are fused to obtain driving features.

[0120] In one possible embodiment, feature fusion can be a simple concatenation that directly connects feature vectors from different sources or of different types to form a new, higher-dimensional feature vector.

[0121] In one possible embodiment, feature fusion can be weighted fusion, assigning different weights to different features to obtain driving features.

[0122] In one possible implementation, before feature fusion, the original features are first screened to remove redundant or irrelevant features, and then the screened features are fused. This can reduce the dimensionality of the data and improve the efficiency and accuracy of the model. For example, there may be a certain correlation between the average acceleration magnitude and the frequency of rapid acceleration. These redundant features are removed using a feature selection algorithm, and then the remaining features are fused.

[0123] In one possible embodiment, each first fault type and each second fault type can be encoded, mapping different types of fault types to corresponding fault identifiers or vector representations to obtain fault features. Then, the distance between the driving features and the fault features is calculated to measure their differences, for example, using methods such as Euclidean distance, Manhattan distance, or cosine similarity. A smaller distance indicates a higher degree of matching.

[0124] In one possible implementation, the degree of matching is determined by calculating the similarity between driving features and fault features. For example, the Jaccard similarity coefficient or the Pearson correlation coefficient can be used to measure the similarity between the two sets; the higher the similarity, the higher the degree of matching.

[0125] In one possible implementation, a machine learning model can be used to learn the mapping relationship between driving characteristics and fault types, and the matching degree can be determined based on the model's prediction results. Examples include random forests and neural networks.

[0126] The process involves data preprocessing of the status information. First, data cleaning is performed to remove noisy, outlier, and erroneous data. Then, the status information collected before and after the fault is synchronized over time to ensure temporal consistency for subsequent analysis. Finally, status information from different sources and with different dimensions is standardized to make them comparable.

[0127] This process includes feature extraction of the pre-fault state information, calculating statistical characteristics such as mean, variance, standard deviation, maximum, and minimum values ​​over a period of time before the fault, reflecting the characteristics of the equipment or system under normal operating conditions. It also analyzes the changing trends of the pre-fault state information, such as increases, decreases, and fluctuations, extracting trend features through methods like calculating trend coefficients and fitting curves. Furthermore, it performs frequency domain analysis on the state information to extract frequency features such as dominant frequency, harmonics, and overtones, which is beneficial for detecting periodic and vibration faults.

[0128] Among them, feature extraction is performed on the state information after the fault, and the change features are extracted, that is, the change of state information relative to the state before the fault is calculated over a period of time after the fault, such as temperature change, pressure change, speed change, etc., to reflect the degree of impact of the fault on the equipment or system.

[0129] The extracted pre-fault and post-fault state information features are combined to form a comprehensive state feature vector. Features from different sources and types can be combined using methods such as concatenation and fusion for subsequent matching calculations.

[0130] In one possible embodiment, the difference between state features and fault features can be measured by calculating the distance between them, or the matching degree can be determined by calculating the similarity between state features and fault features. Alternatively, a machine learning model can be used to learn the mapping relationship between state features and fault types, and the matching degree can be determined based on the model's prediction results.

[0131] In one possible embodiment, determining the target fault type based on the at least one first matching degree, the at least one second matching degree, the at least one third matching degree, and the at least one fourth matching degree includes: acquiring vehicle maintenance data; determining, based on the maintenance data, a first fault frequency of each first fault type and a second fault frequency of each second fault type in a preset historical period; determining a probability value for each first fault type based on the first fault frequency, the at least one third matching degree, and the at least one first matching degree; determining a probability value for each second fault type based on the second fault frequency, the at least one second matching degree, and the at least one fourth matching degree; and determining the target fault type based on the probability values ​​of each first fault type and each second fault type.

[0132] The maintenance data includes detailed information such as maintenance time, maintenance items, and replaced parts. The frequency of occurrence for each first fault type and each second fault type is determined from the maintenance data. The third weight of the third matching degree and the first weight of the first matching degree are determined based on the first fault frequency. The second weight of the second matching degree and the fourth weight of the fourth matching degree are determined based on the second fault frequency.

[0133] Specifically, the first matching degree, third matching degree, first weight, and third weight are weighted and summed to obtain the probability value of each first fault type; the second matching degree, fourth matching degree, second weight, and fourth weight are weighted and summed to obtain the probability value of each second fault type. The fault type with the highest probability value is determined as the target fault type.

[0134] In one possible embodiment, if the vehicle feature is detected to include the second feature and the third feature, then the vehicle identification image uploaded by the user is received; features of the vehicle identification image are extracted to obtain a fourth feature; and the target fault type is determined based on the fourth feature, the second feature, the third feature, and the status information.

[0135] If the vehicle brand, model, year, or other information cannot be identified from the dashboard image, a prompt window will pop up, prompting the user to upload images of the exterior, interior, and other identifying elements. For example, the prompt window can refer to... Figure 5 , Figure 5 This is a schematic diagram of a prompt window provided in an embodiment of this application, such as... Figure 5 As shown, the prompt window includes a message to the user indicating that there are not enough images and asking whether to continue uploading. By clicking the "Yes" button, the user returns to the fault diagnosis interface and can then continue uploading images.

[0136] The system receives an image uploaded by the user, performs image recognition, and obtains a fourth feature that represents information such as the vehicle brand, model, and year. Combined with the second and third features and status information, the system determines the target fault type.

[0137] In one possible implementation, if the user clicks the "No" control in the prompt window, the agent can collect basic vehicle information through multiple rounds of questioning. Simultaneously, the collected information is fed back into the agent for training.

[0138] For example, a multi-turn dialogue can be:

[0139] The first round focuses on brands and car models.

[0140] Intelligent Agent: "Hello, I would like to know about your vehicle information. First, could you tell me the brand of your vehicle?"

[0141] User: "My car is a Volkswagen."

[0142] Agent: "Okay, Volkswagen is a good brand. So, which Volkswagen model do you own?"

[0143] User: "It's a Volkswagen Lavida."

[0144] Second round: Vehicle identification number (VIN).

[0145] Agent: "Thank you very much for the information. Next, I need your vehicle identification number. Do you know where it is?"

[0146] User: "I'm not sure."

[0147] The agent said, “No problem, let me tell you. You can see a combination of numbers and letters on the front left of the vehicle’s dashboard; that’s the vehicle identification number (VIN). Or, if you have a vehicle registration certificate or insurance policy, it will also have the VIN. Please find it and let me know.”

[0148] User: "Found it, it's this number: 3VW5D7AXE2S4521456."

[0149] Third round: Purchase date and mileage.

[0150] Agent: "Thank you very much for providing the detailed information. Finally, I would like to know your vehicle's purchase date and current mileage, as this will help us to more accurately assess your vehicle's condition. Could you please tell me?"

[0151] User: "I bought my car in 2019, and it has driven about 20,000 kilometers so far."

[0152] The AI ​​agent says: "Thank you very much for providing this information. According to your description, your vehicle is a Volkswagen Lavida purchased in 2019, with the vehicle identification number 3VW5D7AXE2S123456, and the current mileage is approximately 20,000 kilometers. We will provide you with more accurate services based on this information. If you have any further questions or need further assistance, please feel free to let me know."

[0153] Therefore, basic vehicle information was collected based on multiple rounds of questions.

[0154] In one possible embodiment, before receiving the vehicle identification image uploaded by the user, the method further includes: determining at least one shooting location of the vehicle; determining a shooting strategy for each of the at least one shooting location; and generating guidance information based on each shooting location and the shooting strategy corresponding to each shooting location, the guidance information being used to guide the user to upload the vehicle identification image.

[0155] If the user clicks on a control in the prompt window, they need to continue uploading images, which can be interior or exterior images, etc.

[0156] The shooting locations can include the front of the vehicle, showcasing its overall shape, grille design, headlight shape, brand logo, and other important features. It can also include the rear of the vehicle, displaying brand logos, model name markings, and taillight design. Furthermore, it can include the sides of the vehicle, showcasing body lines, window outlines, wheel designs, and door handles. Interior details, such as the steering wheel, center console, and seats, can also be captured.

[0157] Each shooting location corresponds to a shooting strategy. For example, the shooting strategy includes shooting distance, shooting angle, and shooting precautions, etc., to facilitate the uploading of high-quality images.

[0158] S240, determine the target analysis agent based on the target fault type.

[0159] Once the fault type is diagnosed, the system will invoke the corresponding analytical agent module to provide detailed fault analysis and repair suggestions. For example, the analytical agent could be an AI Copilot.

[0160] The analytical agents are artificial intelligence programs trained on a vast amount of vehicle knowledge, maintenance experience, and fault case data. Each analytical agent focuses on a specific fault type or vehicle system, such as engine failure or braking system failure, and includes detailed fault analysis logic and a rich database of maintenance suggestions.

[0161] Once the fault type is determined, the intelligent diagnostic system will automatically invoke the matching analytical agent based on the fault type classification. For example, when an engine fault is determined, the system will invoke the analytical agent related to engine faults, thereby providing the user with the most accurate and relevant fault analysis and repair suggestions.

[0162] S250, based on the target analysis agent and the vehicle characteristics, determine the vehicle's fault diagnosis report.

[0163] In one possible embodiment, determining the vehicle's fault diagnosis report based on the target analysis agent and the vehicle characteristics includes: acquiring the vehicle's driving data and maintenance data; parsing the vehicle characteristics, driving data, and maintenance data through the target analysis agent to obtain a parsing result; if the parsing result indicates successful parsing, generating the fault diagnosis report; if the parsing result indicates failed parsing, outputting multiple fault causes.

[0164] The invoked target analysis agent, upon receiving vehicle characteristics, driving data, and maintenance data, conducts an in-depth analysis of the fault, considering various factors such as the meaning of the fault code, vehicle operating data, the frequency of fault occurrence, and environmental conditions. Through comprehensive analysis, the module can provide a detailed explanation of the fault's cause, helping users understand the mechanism behind the fault.

[0165] The fault diagnosis report includes information such as the location of the fault, the cause of the fault, and the severity of the fault.

[0166] If the parsing fails or the knowledge base lacks relevant information, the system will provide the user with several possible causes of the failure and send a tagged image to the back-end maintenance personnel.

[0167] For fault images that cannot be analyzed, experienced maintenance personnel conduct secondary analysis. After determining the fault location, cause, severity, and repair recommendations, the data is fed back to the analytical agent for self-learning. Through continuous learning, the analytical agent can more quickly and accurately locate faults when encountering similar images in the future.

[0168] For faults that the analytical agent can identify, the system provides real-time diagnostic reports to the user via software or a web interface. These reports are presented in a combination of text and graphics for easy understanding. If a fault cannot be identified, the system will display several potential faults and their corresponding reports to the user.

[0169] In one possible embodiment, the system continuously optimizes the diagnostic model and algorithm based on user feedback and new diagnostic data, thereby improving the system's accuracy and user experience.

[0170] In one possible implementation, a user feedback module can be developed within the system to collect user opinions and optimize system functionality. The intelligent agent diagnostic model can be updated regularly, incorporating the latest data to enhance diagnostic capabilities.

[0171] S260, Generate a fault handling strategy based on the fault diagnosis report.

[0172] After completing the fault analysis, specific repair recommendations can be provided based on the analysis results. These recommendations include repair steps, a list of parts that need to be replaced, the tools and technical requirements that may be needed during the repair process, and repair costs.

[0173] Users can interact with the analytical agent to ask further questions about faults and repairs. The analytical agent will provide more detailed answers and guidance based on the user's questions. After the repair is completed, users can also provide feedback to the system, helping it to continuously optimize the accuracy and practicality of fault analysis and repair suggestions.

[0174] As can be seen, this application fills the data gaps, ensuring the comprehensiveness, reliability, and accuracy of diagnostic and maintenance recommendations. Dynamic learning capabilities and multi-source data fusion enable the system to more accurately match fault codes and maintenance recommendations, solving the accuracy problems caused by limited data and fixed rules in existing technologies. By continuously learning new fault modes and maintenance experience, the system can dynamically optimize diagnostic logic and adapt to constantly changing fault phenomena, solving the problem of lacking dynamic learning capabilities in existing technologies. Personalized, simple, intuitive, and content-rich maintenance recommendations are output based on different operator habits and location coordinates, avoiding user fatigue from monotonous content and improving the user's intuitive experience.

[0175] As can be seen, in this application, users only need to take a picture of the dashboard with their mobile phone to identify the fault. It is low-cost, efficient, and easy to operate. Moreover, the more image data there is, the more accurate the diagnostic results become.

[0176] For examples consistent with the above embodiments, please refer to... Figure 6 , Figure 6 This is a functional unit block diagram of an intelligent diagnostic device for vehicles provided in an embodiment of this application, such as... Figure 6As shown, the intelligent diagnostic device 60 for vehicles includes: an extraction unit 61, used to extract vehicle features from a user-uploaded dashboard image; an acquisition unit 62, used to acquire the vehicle's status information within a first time period, the first time period including a first time period before the fault time and a second time period after the fault time; a first determination unit 63, used to determine a target fault type based on the status information and the vehicle features if the vehicle features include a first feature representing vehicle attributes, a second feature representing vehicle fault status, and a third feature representing vehicle operating status; a second determination unit 64, used to determine a target analysis agent based on the target fault type; a third determination unit 65, used to determine a fault diagnosis report for the vehicle based on the target analysis agent and the vehicle features; and a generation unit 66, used to generate a fault handling strategy based on the fault diagnosis report.

[0177] In one possible embodiment, in determining the target fault type based on the status information and the vehicle characteristics, the first determining unit 63 is specifically configured to: determine at least one first fault type based on the first characteristic and the second characteristic; determine at least one second fault type based on the second characteristic and the third characteristic; and determine the target fault type among the at least one first fault type and the at least one second fault type based on the status information.

[0178] In one possible embodiment, in determining the target fault type among the at least one first fault type and the at least one second fault type based on the status information, the first determining unit 63 is further configured to: determine the user's driving characteristics; determine the matching degree between the driving characteristics and each of the at least one first fault type to obtain at least one first matching degree; and determine the matching degree between the driving characteristics and each of the at least one second fault type to obtain at least one second matching degree; determine the matching degree between the status information and each of the first fault types to obtain at least one third matching degree; and determine the matching degree between the status information and each of the second fault types to obtain at least one fourth matching degree; and determine the target fault type based on the at least one first matching degree, the at least one second matching degree, the at least one third matching degree, and the at least one fourth matching degree.

[0179] In one possible embodiment, in determining the target fault type based on the at least one first matching degree, the at least one second matching degree, the at least one third matching degree, and the at least one fourth matching degree, the first determining unit 63 is further configured to: acquire maintenance data of the vehicle; determine, based on the maintenance data, a first fault frequency of each first fault type and a second fault frequency of each second fault type in a preset historical period; determine a probability value of each first fault type based on the first fault frequency, the at least one third matching degree, and the at least one first matching degree; and determine a probability value of each second fault type based on the second fault frequency, the at least one second matching degree, and the at least one fourth matching degree; and determine the target fault type based on the probability values ​​of each first fault type and each second fault type.

[0180] In one possible embodiment, the intelligent diagnostic device 60 for the vehicle is further configured to: detect that the vehicle features include the second feature and the third feature, then receive the vehicle identification image uploaded by the user; extract features from the vehicle identification image to obtain a fourth feature; and determine the target fault type based on the fourth feature, the second feature, the third feature, and the status information.

[0181] In one possible embodiment, before receiving the vehicle identification image uploaded by the user, the vehicle's intelligent diagnostic device 60 is further configured to: determine at least one shooting location of the vehicle; determine a shooting strategy for each of the at least one shooting location; and generate guidance information based on each shooting location and the shooting strategy corresponding to each shooting location, the guidance information being used to guide the user to upload the vehicle identification image.

[0182] In one possible embodiment, in determining the vehicle's fault diagnosis report based on the target analysis agent and the vehicle characteristics, the third determining unit 65 is specifically configured to: acquire the vehicle's driving data and maintenance data; parse the vehicle characteristics, driving data, and maintenance data through the target analysis agent to obtain a parsing result; if the parsing result indicates successful parsing, generate the fault diagnosis report; if the parsing result indicates failed parsing, output multiple fault causes.

[0183] It is understood that since the method embodiments and the device embodiments are different presentations of the same technical concept, the content of the method embodiment section in this application should be adapted to the device embodiment section, and will not be repeated here.

[0184] In the case of using integrated units, please refer to Figure 7 , Figure 7 This is a functional unit block diagram of another intelligent diagnostic device for vehicles provided in this application embodiment, such as... Figure 7 As shown, the intelligent diagnostic device 60 for a vehicle includes a processing module 602 and a communication module 601. The processing module 602 controls and manages the actions of the intelligent diagnostic device 60, for example, executing the steps of the extraction unit 61, the acquisition unit 62, the first determination unit 63, the second determination unit 64, the third determination unit 65, and the generation unit 66, and / or performing other processes of the technology described herein. The communication module 601 is used for interaction between the intelligent diagnostic device 60 and other devices. Figure 7 As shown, the vehicle's intelligent diagnostic device 60 may also include a storage module 603, which is used to store the program code and data of the vehicle's intelligent diagnostic device 60.

[0185] The processing module 602 can be a processor or controller, such as a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc. The communication module 601 can be a transceiver, RF circuitry, or a communication interface, etc. The storage module 603 can be a memory.

[0186] All relevant content in each scenario involved in the above method embodiments can be referenced from the functional descriptions of the corresponding functional modules, and will not be repeated here. The intelligent diagnostic device 60 of the vehicle described above can perform the above... Figure 2 The intelligent diagnostic method for the vehicle shown.

[0187] Please see Figure 8 , Figure 8 This is a schematic diagram of the structure of an electronic device proposed in an embodiment of this application, such as... Figure 8As shown, the electronic device 800 includes a processor 810, a memory 820, a communication interface 830, and one or more programs 821. The one or more programs 821 are stored in the memory and configured to be executed by the processor. When the program is executed, it includes some or all of the steps of any of the intelligent diagnostic methods for vehicles described in the above method embodiments. The processor, memory, and communication interface are interconnected and complete communication between them.

[0188] The memory can be volatile memory such as Dynamic Random Access Memory (DRAM) or non-volatile memory such as a hard disk drive. The memory stores a set of executable program code, and the processor calls the executable program code stored in the memory to execute some or all of the steps of any vehicle intelligent diagnostic method described in the above embodiments of the vehicle intelligent diagnostic method.

[0189] As can be seen, the electronic device 800 described in this application embodiment first receives a vehicle dashboard image uploaded by a user; then extracts vehicle features from the dashboard image; subsequently obtains the vehicle's status information within a first time period, which includes a first time period before the fault and a second time period after the fault; then detects that the vehicle features include a first feature representing vehicle attributes, a second feature representing vehicle fault status, and a third feature representing vehicle operating status; then determines the target fault type based on the status information and the vehicle features; next, determines the target analysis agent based on the target fault type; then determines the vehicle's fault diagnosis report based on the target analysis agent and the vehicle features; finally, generates a fault handling strategy based on the fault diagnosis report.

[0190] As can be seen, in the application, users only need to provide an image of the vehicle's dashboard. By extracting vehicle features from multiple dimensions through image recognition, fault diagnosis results can be obtained quickly without relying on professional equipment or technician experience, thereby reducing diagnostic costs and improving diagnostic efficiency and user experience.

[0191] This application also provides a computer storage medium storing a computer program for electronic data interchange, which causes a computer to perform some or all of the steps of any of the methods described in the above method embodiments, wherein the computer includes an electronic device.

[0192] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of any of the methods described in the above method embodiments. The computer program product may be a software installation package, and the computer may include an electronic device.

[0193] It should be noted that, for the sake of simplicity, the aforementioned methods are described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are optional, and the actions and modules involved are not necessarily essential to this application.

[0194] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0195] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0196] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0197] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software program module.

[0198] If the integrated unit is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0199] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage device, which may include: a flash drive, a read-only memory, a random access memory, a magnetic disk, or an optical disk, etc.

[0200] The embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The above description of the embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for intelligent diagnosis of a vehicle, characterized in that, include: Extract vehicle features from the dashboard image of the vehicle uploaded by the user; The status information of the vehicle is obtained within a first time period, which includes a first time period before the time of the fault and a second time period after the time of the fault; the status information includes driving information, operation information, environmental information and vehicle performance change information. If the vehicle features are detected to include a first feature characterizing vehicle attributes, a second feature characterizing vehicle fault status, and a third feature characterizing vehicle operating status, then the target fault type is determined based on the status information and the vehicle features. Determine the target analysis agent based on the target fault type; Based on the target analysis agent and the characteristics of the vehicle, a fault diagnosis report for the vehicle is determined; A fault handling strategy is generated based on the fault diagnosis report; The step of determining the target fault type based on the status information and the vehicle characteristics includes: Based on the first feature and the second feature, at least one first fault type is determined; based on the second feature and the third feature, at least one second fault type is determined; the user's driving characteristics are determined; the matching degree between the driving characteristics and each of the at least one first fault type is determined to obtain at least one first matching degree; and the matching degree between the driving characteristics and each of the at least one second fault type is determined to obtain at least one second matching degree; the matching degree between the status information and each of the first fault types is determined to obtain at least one third matching degree; and the matching degree between the status information and each of the second fault types is determined to obtain at least one fourth matching degree; based on the at least one first matching degree, the at least one second matching degree, the at least one third matching degree, and the at least one fourth matching degree, the target fault type is determined; The step of determining the target fault type based on the at least one first matching degree, the at least one second matching degree, the at least one third matching degree, and the at least one fourth matching degree includes: Obtain the maintenance data of the vehicle; Based on the maintenance data, determine the first fault frequency of each first fault type and the second fault frequency of each second fault type within a preset historical period; Based on the first fault frequency, the at least one third matching degree, and the at least one first matching degree, a probability value for each first fault type is determined; and based on the second fault frequency, the at least one second matching degree, and the at least one fourth matching degree, a probability value for each second fault type is determined. The target fault type is determined based on the probability values ​​of each first fault type and each second fault type.

2. The method according to claim 1, characterized in that, The method further includes: If the vehicle features are detected to include the second feature and the third feature, then the vehicle identification image uploaded by the user is received; Extract the features from the vehicle identification image to obtain the fourth feature; The target fault type is determined based on the fourth feature, the second feature, the third feature, and the status information.

3. The method according to claim 2, characterized in that, Before receiving the vehicle identification image uploaded by the user, the method further includes: Determine at least one shooting position of the vehicle; Determine the shooting strategy for each of the at least one shooting positions; Based on each shooting location and the shooting strategy corresponding to each shooting location, guidance information is generated, which is used to guide the user to upload the vehicle identification image.

4. The method according to claim 1, characterized in that, The step of determining the vehicle's fault diagnosis report based on the target analysis agent and the vehicle's characteristics includes: Obtain the vehicle's driving data and maintenance data; The target analysis agent analyzes and processes the vehicle features, driving data, and maintenance data to obtain analysis results. If the parsing result is used to indicate successful parsing, then the fault diagnosis report is generated; If the parsing result indicates parsing failure, multiple fault reasons are output to the user and the back-end maintenance personnel so that the back-end maintenance personnel can analyze the multiple fault reasons.

5. An intelligent diagnostic device for vehicles, characterized in that, include: The extraction unit is used to extract vehicle features from the dashboard image of the vehicle uploaded by the user. The acquisition unit is used to acquire the status information of the vehicle within a first time period, the first time period including a first time period before the fault time and a second time period after the fault time; The first determining unit is configured to detect that the vehicle features include a first feature characterizing vehicle attributes, a second feature characterizing vehicle fault status, and a third feature characterizing vehicle operating status, and then determine the target fault type based on the status information and the vehicle features. The second determining unit is used to determine the target analysis agent based on the target fault type; The third determining unit is used to determine the vehicle's fault diagnosis report based on the target analysis agent and the vehicle's characteristics. A generation unit is used to generate a fault handling strategy based on the fault diagnosis report; The step of determining the target fault type based on the status information and the vehicle characteristics includes: Based on the first feature and the second feature, at least one first fault type is determined; based on the second feature and the third feature, at least one second fault type is determined; the user's driving characteristics are determined; the matching degree between the driving characteristics and each of the at least one first fault type is determined to obtain at least one first matching degree; and the matching degree between the driving characteristics and each of the at least one second fault type is determined to obtain at least one second matching degree; the matching degree between the status information and each of the first fault types is determined to obtain at least one third matching degree; and the matching degree between the status information and each of the second fault types is determined to obtain at least one fourth matching degree; based on the at least one first matching degree, the at least one second matching degree, the at least one third matching degree, and the at least one fourth matching degree, the target fault type is determined; The step of determining the target fault type based on the at least one first matching degree, the at least one second matching degree, the at least one third matching degree, and the at least one fourth matching degree includes: Obtain the maintenance data of the vehicle; Based on the maintenance data, determine the first fault frequency of each first fault type and the second fault frequency of each second fault type within a preset historical period; Based on the first fault frequency, the at least one third matching degree, and the at least one first matching degree, a probability value for each first fault type is determined; and based on the second fault frequency, the at least one second matching degree, and the at least one fourth matching degree, a probability value for each second fault type is determined. The target fault type is determined based on the probability values ​​of each first fault type and each second fault type.

6. An electronic device, characterized in that, The device includes: The system includes a memory, a processor, and executable program code stored in the memory and executable on the processor, wherein the processor executes the executable program code to perform the steps of the intelligent diagnostic method for a vehicle as described in any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores executable program code, the executable program code including execution instructions for performing the steps of the intelligent diagnostic method for the vehicle as described in any one of claims 1-4.

Images