Vehicle diagnosis method based on vehicle-mounted automatic diagnosis system and related device

Abstract

The embodiment of the invention discloses a vehicle diagnosis method based on a vehicle-mounted automatic diagnosis system and a related device. The method comprises the following steps: acquiring voice data, personnel data and historical interaction data of a target vehicle; the personnel data is used for reflecting at least one of the following personnel information: personnel quantity, personnel identity, personnel position, personnel behavior and personnel acoustic characteristics; the historical interaction data is used for reflecting at least one of the following interaction information: historical driving habits and historical voice instruction records; analyzing the voice data according to the personnel data and the historical interaction data to obtain a target statement; searching a preset instruction database according to the target statement to obtain a first target instruction corresponding to the target statement; acquiring vehicle data of the target vehicle according to the first target instruction; the target vehicle is diagnosed according to the vehicle data, and diagnosis information is obtained; and pushing the diagnosis information. By adopting the embodiment of the invention, the manual operation steps of the vehicle owner can be simplified, and the vehicle diagnosis efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicles, and in particular to a vehicle diagnosis method and related devices based on an on-vehicle automatic diagnosis system. Background Art

[0002] With the rapid development of vehicle intelligence, vehicle owners have higher and higher requirements for vehicle management, especially in terms of fault diagnosis, real-time monitoring, and remote control.

[0003] However, the disadvantage of traditional on-vehicle automatic diagnosis systems is that they rely on equipment. Vehicle owners or maintenance technicians must use on-vehicle automatic diagnosis system reading tools, such as on-vehicle automatic diagnosis system code readers or mobile applications, to query fault codes and diagnostic information. This may be relatively complex for non-technical vehicle owners and lacks sufficient interactivity and intelligent functions. This operation method requires vehicle owners to manually operate the equipment, which may cause distraction while driving and affect driving safety. Vehicle owners cannot conveniently obtain diagnostic information or perform system control during driving. Manually entering fault codes or querying vehicle data may lead to operation errors or misunderstandings, reducing the diagnostic efficiency.

[0004] Therefore, there is an urgent need for a vehicle diagnosis method based on an on-vehicle automatic diagnosis system that can simplify the manual operation steps of vehicle owners and improve the vehicle diagnosis efficiency when vehicle owners need to diagnose the vehicle. Summary of the Invention

[0005] To solve the above problems, embodiments of the present invention provide a vehicle diagnosis method and related devices based on an on-vehicle automatic diagnosis system, which can simplify the manual operation steps of vehicle owners and improve the vehicle diagnosis efficiency when vehicle owners need to diagnose the vehicle.

[0006] In a first aspect, embodiments of the present invention provide a vehicle diagnosis method based on an on-vehicle automatic diagnosis system, including:

[0007] Obtain voice data, personnel data, and historical interaction data of a target vehicle; the personnel data is used to reflect at least one of the following personnel information: the number of personnel, personnel identity, personnel location, personnel behavior, and personnel acoustic characteristics; the historical interaction data is used to reflect at least one of the following interaction information: historical driving habits and historical voice command records;

[0008] Parse the voice data according to the personnel data and the historical interaction data to obtain a target statement;

[0009] Search a preset instruction database according to the target statement to obtain a first target instruction corresponding to the target statement;

[0010] Obtain vehicle data of the target vehicle according to the first target instruction;

[0011] Diagnose the target vehicle based on the vehicle data to obtain diagnostic information;

[0012] Push the diagnostic information.

[0013] In a second aspect, an embodiment of the present invention provides a vehicle diagnostic device based on an on-board diagnostic system. The device includes an acquisition unit and a processing unit;

[0014] The acquisition unit is configured to acquire voice data, personnel data, and historical interaction data of a target vehicle; the personnel data is used to reflect at least one of the following personnel information: the number of personnel, personnel identity, personnel location, personnel behavior, and personnel acoustic characteristics; the historical interaction data is used to reflect at least one of the following interaction information: historical driving habits and historical voice command records;

[0015] The processing unit is configured to parse the voice data according to the personnel data and the historical interaction data to obtain a target statement;

[0016] Search a preset instruction database according to the target statement to obtain a first target instruction corresponding to the target statement;

[0017] Obtain the vehicle data of the target vehicle according to the first target instruction;

[0018] Diagnose the target vehicle based on the vehicle data to obtain diagnostic information;

[0019] Push the diagnostic information.

[0020] In a third aspect, an embodiment of the present invention provides an electronic device. The electronic device includes a processor and a memory. The processor is connected to the memory. The memory is used to store a computer program. The processor is used to execute the computer program stored in the memory so that the electronic device executes the method described in the first aspect.

[0021] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium. The computer-readable storage medium stores a computer program. The computer program is executed by a processor to implement the method described in the first aspect.

[0022] In a fifth aspect, an embodiment of the present application provides a computer program product. The computer program product includes a non-transitory computer-readable storage medium storing a computer program. The computer program is operable to cause a computer to execute the method described in the first aspect.

[0023] Implementing the embodiments of the present application has the following beneficial effects:

[0024] In the embodiment of the present application, first, voice data, personnel data, and historical interaction data of the target vehicle are obtained. The personnel data is used to reflect at least one of the following personnel information: the number of personnel, the identity of personnel, the location of personnel, the behavior of personnel, and the acoustic characteristics of personnel. The historical interaction data is used to reflect at least one of the following interaction information: historical driving habits and historical voice command records. Then, the voice data is parsed based on the personnel data and the historical interaction data to obtain a target statement, and the preset instruction database is searched according to the target statement to obtain a first target instruction corresponding to the target statement. Next, vehicle data of the target vehicle is obtained according to the first target instruction, and the target vehicle is diagnosed based on the vehicle data to obtain diagnostic information. Finally, the diagnostic information is pushed. Thus, by obtaining voice data, determining the corresponding first target instruction for the voice data, diagnosing according to the first target instruction to obtain diagnostic information, and pushing the diagnostic information, when the vehicle owner needs to diagnose the vehicle, the manual operation steps of the vehicle owner can be simplified, and the vehicle diagnosis efficiency can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the background art, the following will describe the drawings required to be used in the embodiments of the present invention or the background art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0026] Figure 1 is a schematic diagram of the architecture of an on-vehicle automatic diagnostic system provided by an embodiment of the present application;

[0027] Figure 2 is a flowchart of a vehicle diagnosis method based on an on-vehicle automatic diagnostic system provided by an embodiment of the present application;

[0028] Figure 3 is a schematic diagram of a data interaction method based on intelligent control of the vehicle owner provided by an embodiment of the present application;

[0029] Figure 4 is a schematic diagram of an application scenario of a vehicle diagnosis method based on intelligent control of the vehicle owner provided by an embodiment of the present application;

[0030] Figure 5 is a flowchart of a vehicle diagnosis method based on a voice assistant and intelligent control of the vehicle owner provided by an embodiment of the present application;

[0031] Figure 6 is a timing diagram of a vehicle diagnosis method based on a voice assistant and intelligent control of the vehicle owner provided by an embodiment of the present application;

[0032] Figure 7It is a schematic structural diagram of a vehicle diagnosis device based on an on-vehicle automatic diagnosis system provided by an embodiment of the present application;

[0033] Figure 8 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Specific embodiments

[0034] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0035] The terms "first", "second", "third", "fourth", etc. in the specification and claims of the present application and the accompanying drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or modules is not limited to the listed steps or modules, but optionally further includes steps or modules not listed, or optionally further includes other steps or modules inherent to these processes, methods, products or devices.

[0036] Referring to "embodiments" herein means that specific features, results or characteristics described in connection with the embodiments may be included in at least one embodiment of the present application. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein may be combined with other embodiments.

[0037] The traditional On-Board Diagnostics (OBD) system is complex and inconvenient to operate. The vehicle owner needs to use a mobile application or a physical interface to query fault codes, monitor vehicle health, etc. This operation method requires the owner to manually operate the device, which may cause distraction while driving, affecting driving safety, and is not convenient. The vehicle owner cannot conveniently obtain diagnostic information or perform system control during driving. Especially for vehicle owners who are not familiar with technology, manually entering fault codes or querying vehicle data may lead to operation errors or misunderstandings. Existing OBD systems usually provide basic functions such as fault code query and real-time data monitoring, but lack intelligent and personalized interaction and reminders. Vehicle owners usually need to actively query data or check the vehicle status, lacking proactive health reminders and maintenance suggestions. The interaction method of existing systems is single, and they fail to make full use of intelligent assistants and voice technology to enhance the diagnostic experience. When obtaining vehicle health information, vehicle owners cannot enjoy personalized voice interaction or intelligent suggestions. Most traditional OBD applications do not support functions such as remotely controlling the vehicle, such as remotely starting self-check, adjusting air conditioner temperature, opening and closing windows, etc. Even if some applications have remote functions, they lack the convenience of voice control. When the vehicle breaks down or requires routine inspection, if the vehicle owner is not near the vehicle, they often cannot react or take corresponding actions in time. During driving, vehicle owners usually cannot conveniently check the vehicle status or adjust settings through manual operation. For example, querying vehicle fault information, adjusting air conditioner temperature, etc. all require the vehicle owner to be distracted, affecting safety. Voice control enables vehicle owners to conveniently obtain vehicle information or perform routine operations without leaving the driving state, enhancing driving safety and convenience. Currently, most OBD systems rely on vehicle owners to actively query, lacking intelligent proactive services. Vehicle owners often neglect or delay vehicle health checks, causing potential risks.

[0038] By introducing a voice assistant and intelligent vehicle owner control functions, the present invention enables voice control for fault query and real-time data monitoring. Vehicle owners can query fault codes and obtain real-time vehicle data, such as oil temperature and engine status, through voice commands, avoiding the need for distracted operation while driving and enhancing safety and convenience. It can provide intelligent health reminders and personalized suggestions. The voice assistant can not only respond to vehicle owners' queries but also actively push personalized health reminders, such as brake pad wear and increased fuel consumption, to help vehicle owners detect and solve potential problems in a timely manner. It can perform remote control and vehicle operation management, supporting voice commands to remotely control some vehicle functions, such as starting self-check, adjusting air conditioning temperature, and opening / closing windows. Vehicle owners can conveniently manage vehicle functions without approaching the vehicle. It can enhance driving safety. Through voice control, vehicle owners can complete tasks such as querying and operating without leaving the driving state, thereby reducing distractions during driving and improving safety. By combining the intelligent assistant with OBD intelligent diagnosis, it provides services such as real-time fault diagnosis, active reminder, and personalized suggestions for vehicle owners, realizing more intelligent and comprehensive vehicle management.

[0039] Refer to Figure 1 , Figure 1 FIG. is a schematic structural diagram of an on-vehicle automatic diagnosis system provided by an embodiment of the present application. As Figure 1 shown, the vehicle diagnosis method based on the on-vehicle automatic diagnosis system provided by the embodiment of the present application is applied to the on-vehicle automatic diagnosis system. The system integrates a voice assistant module, an OBD-II interface module, a data verification and analysis module, an intelligent health reminder and push module, a remote control module, and a user interaction interface module. Among them, in combination with the intelligent voice assistant, it provides natural language processing functions. Vehicle owners can query vehicle status, obtain real-time diagnosis reports, adjust vehicle settings, etc. through voice commands. Connect to the vehicle's electronic control unit through the OBD-II port to obtain real-time vehicle diagnosis data, such as fault codes, oil temperature, and engine status. The OBD-II interface module is responsible for data collection and transmission. The data verification and analysis module is used to verify the validity of the data obtained from the OBD-II interface module and perform intelligent analysis based on the vehicle owner's historical data and vehicle status to generate fault reports and health reminders. The intelligent health reminder and push module can actively push health reminders, maintenance suggestions, fault code analysis, etc. to vehicle owners based on the diagnosis results. The remote control module connects to the vehicle owner's intelligent device, such as a mobile phone or in-vehicle device, through a mobile application or cloud service to realize remote control of vehicle functions, such as starting self-check, adjusting the air conditioner, and opening / closing windows. The user interaction interface module enables vehicle owners to interact with the vehicle through a mobile application or in-vehicle information system to query diagnosis information, set reminders, and obtain suggestions.

[0040] Refer to Figure 2 , Figure 2It is a flowchart of a vehicle diagnosis method based on an on-vehicle automatic diagnosis system provided by an embodiment of the present application. As Figure 2 shown, the vehicle diagnosis method based on the on-vehicle automatic diagnosis system provided by the embodiment of the present application includes, but is not limited to, the following steps:

[0041] Step S101: Obtain the voice data, personnel data, and historical interaction data of the target vehicle;

[0042] Among them, the personnel data is used to reflect at least one of the following personnel information: the number of personnel, personnel identity, personnel location, personnel behavior, and personnel acoustic characteristics; the historical interaction data is used to reflect at least one of the following interaction information: historical driving habits and historical voice command records;

[0043] Step S102: Analyze the voice data according to the personnel data and historical interaction data to obtain the target statement;

[0044] Step S103: Search the preset instruction database according to the target statement to obtain the first target instruction corresponding to the target statement;

[0045] Step S104: Obtain the vehicle data of the target vehicle according to the first target instruction;

[0046] Step S105: Diagnose the target vehicle according to the vehicle data to obtain the diagnosis information;

[0047] Step S106: Push the diagnosis information.

[0048] In a possible embodiment, voice data, personnel data, and historical interaction data of the target vehicle are obtained. The personnel data is used to reflect at least one of the following personnel information: the number of personnel, personnel identity, personnel location, personnel behavior, and personnel acoustic characteristics. The historical interaction data is used to reflect at least one of the following interaction information: historical driving habits and historical voice command records. The voice assistant module includes an in-vehicle microphone array. The voice data is obtained through this in-vehicle microphone array, and this array needs to have the characteristics of high sensitivity and low noise interference, and can clearly record the voice information emitted from various positions in the vehicle. Then, the collected analog voice signal is subjected to real-time analog-to-digital conversion to convert it into digital voice data for subsequent processing and analysis. Preliminary noise reduction and gain adjustment and other preprocessing operations are performed on the digital voice data to improve the quality of the voice data. When obtaining the personnel data, the image recognition technology of the in-vehicle camera can be used, combined with deep learning algorithms, to accurately count the personnel in the vehicle. Pressure sensors can also be used to estimate the number of personnel by sensing the pressure distribution on the seat. Face recognition technology is used to compare the face images captured by the camera with the pre-recorded personnel identity database to determine the identity of the personnel in the vehicle. Fingerprint recognition or iris recognition and other technologies can also be combined to improve the accuracy of identity recognition. Through the joint positioning of multiple in-vehicle cameras, combined with image analysis algorithms, the specific positions of the personnel in the vehicle are determined. Ultrasonic sensors or millimeter-wave radars can also be used to detect the position information of the personnel in the vehicle. Behavior recognition algorithms are used to analyze the video data collected by the in-vehicle camera to identify various behaviors of the personnel, such as whether they are operating a mobile phone or wearing a seat belt. Acoustic feature extraction, such as pitch, duration, and timbre, is performed on the collected voice data to establish an acoustic feature model of the personnel. Historical driving habits can be extracted from the vehicle's on-board computer's historical driving data, including vehicle speed, acceleration, braking frequency, gear shifting timing, etc. Data analysis algorithms are used to mine and analyze these data to determine the driver's historical driving habits. The acquisition of historical voice command records can establish a logging function in the vehicle's voice interaction system to record the voice commands issued by the user and the system's response in real time. In the embodiment of the present application, through multi-dimensional data acquisition, the situation of the personnel in the vehicle and the historical interaction information of the vehicle can be comprehensively understood, providing a richer and more accurate basis for subsequent voice data parsing and vehicle diagnosis. For example, the personnel identity information can help the system provide personalized services and diagnoses according to the habits and needs of different drivers. Moreover, combined with the personnel acoustic characteristics and historical voice command records, the voice characteristics and usage habits of the driver can be better adapted, improving the accuracy and efficiency of voice recognition.

[0049] In a possible embodiment, the obtained personnel data and historical interaction data are integrated to form a comprehensive reference data set. The trained acoustic model is used to extract features and match the speech data, convert the speech signal into a text sequence, and combine the personnel data and historical interaction data to perform context analysis on the converted text sequence, eliminate semantic ambiguity, and determine the accurate meaning of the target statement. For example, if the historical interaction data shows that the driver often issues voice commands regarding vehicle fault diagnosis, and the current speech data mentions "There seems to be something wrong with the car", then the intention of this statement can be understood more accurately. In the embodiment of the present application, using the personnel data and historical interaction data for context analysis can more accurately understand the meaning of the voice command, eliminate semantic ambiguity, avoid incorrect operations caused by misunderstandings, and provide personalized voice interaction services for the driver according to the information and historical interaction habits of different personnel, improving the user experience.

[0050] In a possible embodiment, keyword extraction is performed on the target statement to identify the key information therein, such as "fault", "inspection", "engine", etc. The extracted keywords are matched with a preset instruction database, and a combination of fuzzy matching and exact matching is used to find the first target instruction most relevant to the target statement. If multiple possible instructions are matched, they are sorted and filtered according to factors such as the matching degree between the instruction and the target statement and the historical usage frequency to determine the final first target instruction. In the embodiment of the present application, through keyword extraction and database matching, the instruction corresponding to the target statement can be found quickly and accurately, improving the efficiency and accuracy of instruction search. The sorting and filtering mechanism can provide the most relevant and frequently used instructions for the user according to factors such as the matching degree and historical usage frequency, reducing the user's selection time and error rate.

[0051] In a possible embodiment, a communication connection is established with the vehicle's electronic control unit through the vehicle's diagnostic interface to ensure that various vehicle data can be obtained in real time. According to the requirements of the first target instruction, corresponding vehicle data, such as engine speed, water temperature, oil pressure, battery voltage, etc., are collected from various sensors and systems of the vehicle. Preprocessing operations such as cleaning, filtering, and normalization are performed on the collected vehicle data to remove noise and outliers, improving the quality and usability of the data. In the embodiment of the present application, through professional data interfaces and collection methods, various vehicle data can be obtained in real time and accurately, providing a reliable basis for vehicle diagnosis. Preprocessing the collected vehicle data can improve the quality and usability of the data and reduce the impact of noise and outliers on the diagnosis result.

[0052] In a possible embodiment, the acquired vehicle data is compared with a pre-established fault model to determine whether there is a fault in the vehicle, as well as the type and severity of the fault. Using preset diagnostic rules, the vehicle data is inferred and analyzed to further determine the cause of the fault and possible solutions. The diagnostic results are verified and confirmed. Through multiple data acquisitions and analyses, the accuracy and reliability of the diagnostic information are ensured. In the embodiments of the present application, by combining fault model matching and rule reasoning, the fault type and cause of the vehicle can be judged more accurately, more effective solutions can be provided, and the diagnostic results are verified through multiple data acquisitions and analyses, which can ensure the accuracy and reliability of the diagnostic information and avoid misjudgment and missed judgment.

[0053] In a possible embodiment, the diagnostic information is sorted and summarized to form a concise and clear report, including content such as fault description, fault cause, and solution. According to the driver's preferences and the vehicle's configuration, a suitable push method is selected, such as displaying on the in-vehicle display screen, pushing through a mobile phone program, voice broadcast, etc. After the diagnostic information is generated, it is promptly pushed to the driver to ensure that the driver can timely understand the vehicle's condition and take corresponding measures. In the embodiments of the present application, real-time pushing of diagnostic information enables the driver to timely understand the vehicle's condition and take corresponding measures to avoid further deterioration of the fault. Selecting a suitable push method according to the driver's preferences can improve the effect of information transmission and user experience.

[0054] Optionally, step S102, parsing the voice data according to the personnel data and historical interaction data to obtain the target statement may include the following steps:

[0055] Step S201: Decompose the voice data into m voice segments according to the personnel data and historical interaction data; m is a positive integer; the m voice segments correspond to m personnel;

[0056] Step S202: Determine the acoustic state sequence of each of the m voice segments to obtain m acoustic state sequences; the m acoustic state sequences are used to reflect the voice change information of the m voice segments;

[0057] Step S203: Determine the voice recognition result corresponding to each of the m voice segments according to the m acoustic state sequences to obtain m voice recognition results;

[0058] Step S204: Determine the time sequence and logical association relationship of the m voice recognition results;

[0059] Step S205: Determine the instruction authority of each of the m personnel to obtain m instruction authorities;

[0060] Step S206: Determine the instruction probability of each speech recognition result among the m speech recognition results according to the m instruction authorities, chronological order, and logical association relationship, obtaining m instruction probabilities;

[0061] Step S207: Determine the target statement as the speech recognition result corresponding to the maximum value among the m instruction probabilities.

[0062] In a possible embodiment, based on the person acoustic features in the person data, using a feature extraction algorithm, extract the unique acoustic feature vectors of each person from the speech data. Then, use a clustering algorithm to divide the speech data into different clusters according to these acoustic feature vectors, with each cluster corresponding to a person. Among them, the number m of clusters is determined by the number of persons in the person data. For each cluster, combine the person location information and features such as the energy and spectrum of the speech signal to determine the start and end positions of the speech segment. For example, when the energy of the speech signal is lower than a certain threshold, it can be considered that the speech is temporarily interrupted, and this is used as the basis for segment division. Then, decompose the speech data into m speech segments according to the segmentation points, with each speech segment corresponding to a person. In the embodiment of the present application, through person acoustic feature clustering and speech segment division, the speech of different persons can be accurately separated from the mixed speech data, avoiding interference between the speech of different persons, providing a clear input for subsequent speech recognition and instruction judgment, and decomposing the speech data by person enables subsequent speech recognition and analysis to be personalized for each person, taking into account the speech characteristics and instruction authorities of different persons, and improving the accuracy and efficiency of recognition.

[0063] In a possible embodiment, for each speech segment, further extract its acoustic features, such as pitch, duration, intensity, etc., and at the same time perform necessary preprocessing operations, such as noise reduction, normalization, etc., to improve the quality and stability of the features. Convert the extracted acoustic features into a format suitable for model processing, such as vector representation, and use a pre-trained acoustic model, such as a Hidden Markov Model or a Deep Neural Network Acoustic Model, to process the acoustic features of each speech segment. Calculate the acoustic state probability of each time step through the model, and use the Viterbi algorithm, etc. to find the most likely acoustic state sequence, finally obtaining m acoustic state sequences, which record the speech change information of each speech segment over time. In the embodiment of the present application, the acoustic state sequence can record the speech change information in detail, providing richer features for speech recognition, and combined with the language model, the speech can be more accurately converted into text, improving the accuracy of speech recognition.

[0064] In a possible embodiment, an acoustic state sequence is input into a speech recognition system, and the speech is decoded in combination with a language model such as a statistical language model or a neural network language model. Among them, the language model can provide information about vocabulary, grammar, and semantics in the language, helping to improve the accuracy of speech recognition. Through the decoding process, the acoustic state sequence is converted into a corresponding text sequence, and m speech recognition results are obtained. In the embodiments of the present application, by fusing the acoustic model and the language model, making full use of the acoustic features of the speech and the semantic and grammatical information of the language, the ambiguity problem in speech recognition can be effectively solved, and the reliability of the recognition results can be improved.

[0065] In a possible embodiment, during the process of speech data decomposition and processing, time stamps are marked for each speech segment, recording its start and end times. According to the time stamps, the time order of the m speech recognition results is determined to form an ordered sequence. Using natural language processing techniques, semantic analysis and syntactic analysis are performed on the m speech recognition results. For example, dependency syntactic analysis is used to determine the logical relationship between words in a sentence, and semantic role labeling is used to determine the role of each word in the semantics. According to the analysis results, the logical association relationship between the speech recognition results is judged, such as causal relationship, parallel relationship, sequential relationship, etc. In the embodiments of the present application, determining the time order and logical association relationship of the speech recognition results can restore the real speech interaction scenario of the vehicle occupants, help to understand the context and intention of the speech commands, and avoid understanding each speech recognition result in isolation. By analyzing the logical association relationship, the causal, parallel, etc. relationships between the speech commands can be better grasped, so as to more accurately judge the effectiveness and priority of the commands.

[0066] In a possible embodiment, according to the personnel identity information in the personnel data, combined with preset permission rules, instruction permissions are set for each person. For example, the driver of the vehicle may have the highest instruction permission and can execute all vehicle operation instructions, while passengers may only have partial permissions, such as adjusting the in-vehicle music volume, etc. These permission rules are stored in a permission database for subsequent query and use. For m persons, their corresponding instruction permissions are respectively queried from the permission database to obtain m instruction permissions. In the embodiments of the present application, setting instruction permissions according to the personnel identity can ensure that only persons with corresponding permissions can execute specific instructions, ensuring the safety and reasonable use of the vehicle system. For example, preventing passengers from accidentally operating some key vehicle functions. Different persons having different instruction permissions can achieve personalized services and management, meet the needs of different persons, and improve the user experience.

[0067] In a possible embodiment, an instruction probability calculation model is established according to m instruction permissions, chronological order, and logical association relationships. This model can be implemented using machine learning algorithms such as the Naive Bayes classifier, support vector machine, etc., or deep learning models such as long short-term memory networks. The input of the model includes information such as the text content of each speech recognition result, the corresponding instruction permission, chronological order, and logical association relationships, and the output is the probability that the speech recognition result is a valid instruction. Input the m speech recognition results and their related information into the instruction probability calculation model, calculate the instruction probability of each speech recognition result, and obtain m instruction probabilities. Moreover, according to new speech interaction data and actual execution situations, continuously update the parameters of the model to improve the accuracy of instruction probability calculation. In the embodiments of the present application, by comprehensively considering instruction permissions, chronological order, and logical association relationships to calculate instruction probabilities, it is possible to more comprehensively and objectively judge the possibility of each speech recognition result being a valid instruction, and avoid misjudgments caused by single factors.

[0068] In a possible embodiment, compare the magnitudes of the m instruction probabilities, find the maximum value among them, and determine the speech recognition result corresponding to the maximum value as the target statement. In the embodiments of the present application, selecting the speech recognition result with the highest instruction probability as the target statement can maximize the effectiveness and accuracy of the selected instruction and improve the response quality of the vehicle speech interaction system.

[0069] Optionally, step S103, searching the preset instruction database according to the target statement to obtain the first target instruction corresponding to the target statement may include the following steps:

[0070] Step S301: Extract n speech keywords from the target statement; n is a positive integer;

[0071] Step S302: Determine the first score of each speech keyword among the n speech keywords to obtain n first scores; the n first scores are used to reflect the degree of association of each speech keyword among the n speech keywords with the instruction database;

[0072] Step S303: Determine the syntactic relationship between the n speech keywords;

[0073] Step S304: Determine the second score of the target statement according to the n first scores and the syntactic relationship; the second score is used to reflect the accuracy of reference of the target statement;

[0074] Step S305: Search the instruction database according to the second score and the target statement to obtain k instruction search results; k is a natural number;

[0075] Step S306: Determine the instruction levels of each instruction search result among the k instruction search results to obtain k instruction levels; the k instruction levels are used to reflect the instruction execution sequence relationship among the k instruction search results.

[0076] Step S307: Determine the first target instruction from the k instruction search results according to the k instruction levels.

[0077] In a possible embodiment, a natural language processing tool is used to segment the target statement into individual words. For example, for the target statement "Help me check the faults of the engine", it may be segmented into words such as "help", "me", "check", "for", "engine", "of", "faults", etc. According to preset rules or machine learning models, speech keywords related to instructions are filtered out from the segmentation results, such as filtering out some modal particles and words without actual instruction meaning. Finally, n speech keywords such as "check", "engine", "faults" are obtained. In the embodiment of the present application, extracting speech keywords can quickly capture the core instruction elements in the target statement, making subsequent processing more targeted.

[0078] In a possible embodiment, using the correspondence between instructions and words in the instruction database, as well as relevant language statistics such as word frequency and inverse document frequency, a model for calculating the association degree between speech keywords and the instruction database is constructed. For each speech keyword, it is input into the association degree calculation model. The model outputs a value as the association degree between the keyword and the instruction database, that is, the first score, by analyzing factors such as the frequency of the keyword appearing in the instruction database and the co-occurrence relationship with other keywords. For example, "engine" appears frequently in instructions related to vehicle fault diagnosis and may obtain a relatively high first score. In the embodiment of the present application, determining the first score quantifies the association degree between each keyword and the instruction database, provides basic data for evaluating the relevance between the target statement and the instruction, and can narrow the search range when searching the instruction database subsequently, improve the search efficiency, and avoid invalid matching of a large number of irrelevant instructions by pre-evaluating the association degree between the keyword and the instruction database.

[0079] In a possible embodiment, a syntactic analysis tool is used to analyze the target statement. According to the grammar rules, this tool determines the syntactic relationships between various words in the statement, such as subject-predicate relationship, verb-object relationship, modifier-head relationship, etc. For example, in the statement "Check the faults of the engine", the relationship between "Check" and "engine" is a verb-object relationship, and the relationship between "engine" and "faults" is a modifier-head relationship. Then, the obtained syntactic relationships are represented in a structured manner. For example, a tree structure or a relationship matrix can be used to show the syntactic connections between each keyword. In the embodiment of this application, determining the syntactic relationships helps to deeply understand the semantic structure of the target statement, clarify the grammatical connections between each keyword, and thus more accurately grasp the complete meaning of the statement.

[0080] In a possible embodiment, a model that comprehensively considers the first score and syntactic relationships is constructed. This model can be a regression model based on machine learning or a weighted calculation model set according to experience. Taking the n first scores and the syntactic relationships between the speech keywords as inputs, the model performs weighted summation on the first scores and combines the rationality judgment of the syntactic relationships. For example, syntactic relationships that conform to common grammatical structures are given higher weights, and a numerical value is output as the second score of the target statement. A higher second score indicates that the reference of the target statement is more accurate and the semantics is clearer. In the embodiment of this application, the second score obtained by comprehensively considering the first score and syntactic relationships can comprehensively reflect the accuracy of the reference of the target statement. A statement with an accurate reference is more conducive to accurately matching the correct instruction and reducing instruction errors caused by semantic ambiguity.

[0081] In a possible embodiment, matching rules for searching the instruction database are set according to the second score. For example, when the second score is higher than a certain threshold, more precise matching is performed. When the second score is lower, the matching conditions are appropriately relaxed to obtain more possible instruction search results. Then, the target statement and the set matching rules are applied to the instruction database for searching. The database searches for instructions that are semantically similar to the target statement or contain relevant speech keywords according to the rules, and finally obtains k instruction search results. In the embodiment of this application, flexibly adjusting the search rules according to the second score can, while ensuring accuracy, comprehensively obtain the instruction search results related to the target statement and avoid missing important instructions.

[0082] In a possible embodiment, in the instruction database, an instruction level is preset for each instruction, and this level reflects the priority of the instruction in the execution order. For example, instructions related to vehicle safety may be given a higher instruction level, while some comfort adjustment instructions may have a lower level. For k instruction search results, the instruction level corresponding to each result is obtained from the instruction database, resulting in k instruction levels. In the embodiment of the present application, setting the instruction level for each instruction clearly defines the execution order relationship between different instructions. When the vehicle receives multiple instructions simultaneously, it can execute them in a reasonable order to ensure the safety and normal operation of the vehicle.

[0083] In a possible embodiment, the k instruction search results are sorted according to the k instruction levels, with the ones having a higher instruction level ranked in the front. The instruction search result ranked first is selected as the first target instruction to ensure that among multiple possible instructions, the instruction that is most critical to vehicle operation or user needs is preferentially executed.

[0084] Optionally, step S305, searching the instruction database according to the second score and the target statement to obtain k instruction search results may include the following steps:

[0085] Step S401: If the second score is lower than the preset accuracy score threshold, obtain the sensor data of the target vehicle; the sensor data includes at least one of the following: vehicle speed, steering wheel angle, driving time;

[0086] Step S402: Semantically expand the target statement according to the sensor data and the historical interaction data to obtain an expanded statement;

[0087] Step S403: Search the instruction database according to the expanded statement to obtain k instruction search results;

[0088] Step S404: If the second score is higher than or equal to the accuracy score threshold, search the instruction database according to the n first scores and the n voice keywords to obtain k instruction search results.

[0089] In a possible embodiment, the second score of the target statement is calculated, which comprehensively reflects the referential accuracy of the statement, and is compared with a preset accuracy score threshold, which is used to distinguish the clarity and accuracy of the semantics of the target statement. When the second score is lower than the threshold, it indicates that the referential accuracy of the target statement may be insufficient and more information is needed to understand the user's intention. Then, data is obtained from various sensors of the target vehicle. For example, the vehicle speed is obtained through a vehicle speed sensor, the steering wheel angle is measured using a steering wheel angle sensor, and the driving time is recorded with an on-vehicle clock. These sensors are connected to the vehicle's electronic control unit, and the data can be read through a standard interface. At the same time, relevant information is extracted from the historical interaction data record of the vehicle. This data includes the user's past operation habits, common instruction patterns, etc. For example, if the user often queries the vehicle's battery life at a specific time period, then this historical habit may be related to the currently semantically ambiguous statement. Then, natural language processing and data analysis techniques are used to expand the semantics of the target statement by combining the sensor data and the historical interaction data. For example, if the target statement is "There is a problem", and the vehicle speed is abnormal at this time, combined with the user's attention to speed-related problems in the historical interaction data, the expanded statement may be "There is a problem with the vehicle speed". This process supplements the missing or ambiguous semantic part of the target statement through the correlation analysis of various data. The expanded statement is used as a new query condition to search in the instruction database. The database system will search for matching instructions according to the keywords, semantic structures, etc. in the statement, and finally obtain k instruction search results.

[0090] In a possible embodiment, when the second score is higher than or equal to the threshold, it indicates that the semantics of the target statement is relatively clear and accurate. At this time, the instruction database is searched according to n voice keywords and their corresponding first scores. First, the n voice keywords in the target statement are extracted. These keywords are the core words in the statement and represent the main semantics. Then, the first score calculated previously, which reflects the degree of association between each keyword and the instruction database, is used. According to the keywords and scores, a weighted search algorithm is adopted, and keywords with a higher degree of association are given a greater search weight to search for matching instructions in the instruction database, and k instruction search results are obtained. For example, if "engine" and "fault" are voice keywords and the first score of "engine" is very high, the search will focus more on instructions related to engine faults.

[0091] In the embodiments of the present application, when the second score is low, by obtaining sensor data and integrating historical interaction data, additional information dimensions are provided for target statements with ambiguous semantics. The fusion of these multi-source data can effectively supplement the missing semantics, enabling the system to more accurately understand the user's intention and avoid incorrect instruction matching caused by ambiguous statements. For example, when the user says "something doesn't seem right" during vehicle driving, by combining sensor data and finding abnormal engine speed, it can be accurately determined that the user may be referring to an engine problem rather than other aspects. When the second score is high, search is performed based on voice keywords and the first score, leveraging the association degree between the keywords and the instruction database, which can accurately locate the instruction that matches the user's intention in the database, improving the accuracy and efficiency of instruction matching.

[0092] In the embodiments of the present application, due to the diversity and ambiguity of natural language, by setting different processing paths, the system can flexibly adjust the search strategy according to the accuracy of the statement. For less accurate statements, semantic expansion is carried out with the help of multi-source data, and for accurate statements, direct search is performed based on keywords and scores, enhancing the system's adaptability to various natural language inputs. Integrating historical interaction data and real-time sensor data enables the system to learn the user's usage habits and combine the real-time state of the vehicle to make more intelligent judgments, which not only improves the success rate of instruction matching but also provides more personalized services for users, enhancing the user experience.

[0093] Optionally, step S106, pushing diagnostic information, may include the following steps:

[0094] Step S501: Obtain the information length of the diagnostic information;

[0095] Step S502: Determine the information level of the diagnostic information; the information level is used to reflect the importance or urgency of the diagnostic information;

[0096] Step S503: Determine the voice broadcast requirement score of the diagnostic information according to the information length and the information level;

[0097] Step S504: If the voice broadcast requirement score is higher than or equal to the preset voice broadcast requirement score threshold, extract the voice broadcast content from the diagnostic information;

[0098] Step S505: Push the voice broadcast content through voice;

[0099] Step S506: If the voice broadcast requirement score is lower than the voice broadcast requirement score threshold, generate a diagnostic analysis report according to the diagnostic information;

[0100] Step S507: Push the diagnostic analysis report through text.

[0101] In a possible embodiment, character statistics are performed on the diagnostic information to obtain the number of characters in the diagnostic information, and this length reflects the complexity of the diagnostic information. Then, according to the content involved in the diagnostic information and the actual application scenario, a series of rules are preset to determine the information level. For example, for vehicle diagnostic information, if it involves faults in safety-critical systems such as brakes and steering, the information level is set to "high"; if it is only some non-critical components, such as problems with the interior reading lights, the information level is set to "low"; for problems that involve a decrease in vehicle performance but do not affect safety, the information level can be set to "medium". The specific diagnostic information is matched with the preset rules, and key content such as the faulty components and the degree of fault impact mentioned in the diagnostic information is analyzed to determine its corresponding information level. For example, if the diagnostic information shows "abnormal brake system pressure", through rule matching, the information level can be determined to be "high".

[0102] In a possible embodiment, a scoring model based on the information length and the information level is established. The weighted summation method can be used to assign different weights to the information length and the information level respectively. For example, the information level "high" corresponds to a weight of 0.7, "medium" corresponds to a weight of 0.5, and "low" corresponds to a weight of 0.3. In terms of the information length, the information length is divided into several intervals, such as short, medium, and long, which correspond to weights of 0.3, 0.5, and 0.7 respectively. According to the above model, the weights corresponding to the information length and the information level of the diagnostic information are multiplied and then added together to obtain the voice broadcast requirement score. For example, if the information level of a certain diagnostic information is "high" and the information length is 80 characters, the voice broadcast requirement score is 0.64.

[0103] In a possible embodiment, when the voice broadcast requirement score is higher than or equal to the preset voice broadcast requirement score threshold, it indicates that the diagnostic information is relatively important and suitable for voice broadcast. The key content is extracted from the diagnostic information as the voice broadcast content, such as extracting the core parts such as the fault description and the recommended operation. Then, the extracted content is converted into a voice signal through voice synthesis technology and pushed through the in-vehicle audio system or other specified voice playback devices, enabling the user to quickly obtain important information. If the voice broadcast requirement score is lower than the voice broadcast requirement score threshold, it means that the diagnostic information may be relatively less important or more suitable to be presented in text form. According to the structure and content of the diagnostic information, a detailed diagnostic analysis report is generated. The report content can include the fault phenomenon, cause analysis, possible solutions, etc. It is pushed to the user through the text display interfaces such as the vehicle's display screen and the mobile phone APP, facilitating the user to carefully consult and analyze.

[0104] In the embodiments of the present application, by comprehensively considering the information length and information level to determine the voice broadcast requirement score, the most suitable presentation method can be intelligently selected for different diagnostic information. For important and urgent information, voice broadcast can enable users to obtain key information in the first time without being distracted to view the text, which is especially suitable for driving scenarios to ensure that users can timely understand important situations. For relatively less important or information-rich content that is more suitable for detailed reading, it is presented in the form of a text report to facilitate users to view and deeply analyze as needed. Moreover, this method avoids information overload that may be caused by voice broadcasting all diagnostic information, and also avoids the situation where important information may be ignored when presented only in text form. According to different scoring results, information is pushed in a targeted manner, improving the accuracy and efficiency of information transmission and ensuring that users can quickly and effectively obtain the required information. Considering that users have different requirements for information acquisition methods in different scenarios, by dynamically adjusting the information presentation method, the personalized needs of users can be better met. For example, during driving, users tend to quickly understand important information through voice, while after parking, users may be more willing to carefully read the text report to deeply understand the vehicle condition. This adaptive information push method enhances the user experience of interacting with the vehicle in different scenarios and increases user satisfaction with the vehicle diagnostic system. Similarly, selecting voice or text push according to the information characteristics helps to reasonably utilize the hardware resources of the vehicle. Voice broadcast requires the use of an audio output device, while the text report mainly relies on a display device. By reasonably allocating resources, over-occupation or waste of resources is avoided, ensuring the efficient operation of the vehicle system.

[0105] Optionally, the vehicle diagnosis method based on the on-vehicle automatic diagnosis system provided by the embodiments of the present application may further include the following steps:

[0106] Step S601: Obtain the remote control data of the target vehicle; the remote control data is used to perform at least one of the following functions: vehicle self-check, in-vehicle temperature adjustment, vehicle displacement;

[0107] Step S602: Determine the second target instruction according to the remote control data and the instruction database;

[0108] Step S603: Determine the remote control issuance time, remote control issuance location, and remote control issuance device according to the remote control data;

[0109] Step S604: Determine the instruction execution time of the second target instruction according to the remote control issuance time, remote control issuance location, and remote control issuance device;

[0110] Step S605: Control the target vehicle to execute the second target instruction according to the instruction execution time.

[0111] In a possible embodiment, the target vehicle establishes a stable data connection with a remote control server or a user device through an in-vehicle communication module. These communication modules transmit data according to corresponding communication protocols to ensure the accuracy and security of the data. The vehicle receives remote control data sent from the remote control end, parses the received data, and extracts information such as control instruction content and control parameters according to a predefined data format. For example, if it is control data for adjusting the in-vehicle temperature, the target temperature value will be parsed out. If it is a vehicle self-check instruction, relevant parameters such as the scope and type of the self-check will be identified.

[0112] In a possible embodiment, the parsed remote control data is matched with an instruction database, which stores various remote control instructions and their corresponding execution logics and parameter requirements. By comparing key information in the control data, such as instruction keywords and function identifiers, with the records in the database, the corresponding instruction is found. For example, when the control data received contains the keyword "vehicle self-check", the corresponding vehicle self-check instruction and its detailed execution steps are searched for in the instruction database.

[0113] In a possible embodiment, if there are multiple matching instructions or further screening is required, it is judged according to other parameters in the control data, such as priority and applicable scenarios. For example, when control data for vehicle self-check and in-vehicle temperature adjustment are received simultaneously, but the current state of the vehicle is more suitable for self-check first, the vehicle self-check instruction is determined as the second target instruction.

[0114] In a possible embodiment, the remote control issuance time is extracted from the packet header or metadata of the remote control data. This time information can be a timestamp accurate to seconds or even milliseconds, recording the moment when the remote control instruction is sent from the control end. The location where the remote control is issued is obtained by using the positioning function of the remote control device. If the control instruction is sent by a mobile phone program, the mobile phone obtains the current location information through its own positioning module and includes it in the remote control data and sends it to the vehicle. After the vehicle receives the data, the location information therein, such as longitude and latitude coordinates, is extracted. And, according to the source address or device identification information of the data transmission, the remote control issuance device is determined. For example, by identifying the network protocol address or the unique device identifier when the mobile phone program communicates with the vehicle, it is judged that the control instruction is sent from a specific mobile phone device of a certain user.

[0115] In a possible embodiment, according to the preset time rules and strategies, combined with information such as the remote control issuing time, issuing location, and issuing device, the instruction execution time of the second target instruction is determined. For example, for some emergency instructions, such as the emergency braking instruction when the vehicle is stolen, it is executed immediately regardless of when and where it is issued. For some non-emergency in-vehicle temperature adjustment instructions, if the issuing location is far from the vehicle and it is expected that the user will take some time to reach the vehicle, it can be set to be executed a few minutes before the user arrives to create a comfortable in-vehicle environment in advance. When specifically calculating the instruction execution time, appropriate adjustments are made according to factors such as network transmission delay and the current task status of the vehicle. For example, if the network transmission delay is high, the instruction execution time is appropriately delayed to ensure that the vehicle executes the instruction after receiving complete and accurate control data. At the same time, if the vehicle is currently executing other important tasks, the new instruction execution time is postponed until the current task is completed.

[0116] In a possible embodiment, when the determined instruction execution time arrives, the electronic control unit of the vehicle sends the second target instruction to the corresponding execution component. For example, for the vehicle self-check instruction, the electronic control unit sends the instruction to each sensor and control module of the vehicle to start a comprehensive self-check program. For the in-vehicle temperature adjustment instruction, the electronic control unit controls the air conditioning system to adjust the cooling or heating power to reach the set in-vehicle temperature. During the process of the vehicle executing the instruction, the execution status is monitored in real time, and the execution result is fed back to the remote control end. For example, after the vehicle self-check is completed, the self-check report is sent back to the remote control device through the communication module, and the self-check report can include information such as whether a fault is detected and the fault location, so that the user can understand the execution situation of the instruction.

[0117] See Figure 3 , Figure 3 is a schematic diagram of a data interaction method based on intelligent control by the vehicle owner provided by an embodiment of the present application. As Figure 3 shown, the vehicle owner issues a command through the remote control device, and the remote control device communicates with the target vehicle through the server to control the target vehicle to execute relevant instructions.

[0118] In the embodiments of the present application, by obtaining remote control data, remote control of various functions such as vehicle self-check, in-vehicle temperature adjustment, and vehicle relocation can be realized, meeting the diverse needs of users in different scenarios. For example, users can adjust the in-vehicle temperature in advance before going home to enjoy a comfortable environment as soon as they get in the car, or when the vehicle has an abnormality, remotely start the vehicle self-check function to timely understand the vehicle condition. By comprehensively considering information such as the remote control sending time, location, and device to determine the instruction execution time, personalized services are realized, and the instruction execution timing can be flexibly adjusted according to the actual situation and needs of users, improving the user experience. For example, when the user remotely starts the vehicle relocation function, according to the user's current location and the expected arrival time, the vehicle is reasonably arranged to complete the relocation before the user arrives, saving the user's waiting time. By matching with the instruction database to determine the second target instruction, it is ensured that only legal and valid instructions can be executed, preventing illegal instructions or incorrect instructions from damaging the vehicle. At the same time, screening among multiple matching instructions ensures the rationality and priority of instruction execution. When determining the instruction execution time, factors such as network transmission delay and the current task status of the vehicle are considered to avoid vehicle failures or safety accidents caused by improper instruction execution. The vehicle feeds back the execution result to the remote control end during the instruction execution process, enabling the user to timely understand the vehicle status and the instruction execution situation. The entire remote control process enhances the interaction between the user and the vehicle. The user can control and manage the vehicle anytime and anywhere, improving the user's sense of control and the convenience of use of the vehicle.

[0119] Optionally, step S604 of determining the instruction execution time of the second target instruction according to the remote control sending time, remote control sending location, and remote control sending device may include the following steps:

[0120] Step S701: Determine the first time required for the execution of the second target instruction;

[0121] Step S702: Determine the target user corresponding to the remote control data according to the remote control sending location and remote control sending device;

[0122] Step S703: Obtain the location data of the target user;

[0123] Step S704: Determine the second time for the target user to reach the target vehicle according to the location data;

[0124] Step S705: Determine the instruction execution time according to the remote control sending time, the first time, and the second time.

[0125] In a possible embodiment, the second target instruction is analyzed in depth to clarify its specific operation requirements and processes. For example, if the second target instruction is "start the vehicle engine preheating and adjust the interior temperature of the vehicle to 26°C", for the engine preheating, query the vehicle technical manual or set relevant parameters through the vehicle engine control system to determine the time required for the engine preheating to reach the appropriate state, assuming it is 3 minutes. For adjusting the interior temperature of the vehicle to 26°C, estimate the required time through calculation or empirical formula based on the performance parameters of the vehicle air conditioning system and the current interior temperature of the vehicle, assuming it is 5 minutes. Take the longer of the two, that is, 5 minutes as the first time required for the execution of this instruction. And determine the influence of the current device state of the vehicle and external environmental factors on the instruction execution time. If the vehicle battery power is low, the engine start may take longer, and accordingly increase the first time. If the external environmental temperature is extremely low, the difficulty of adjusting the interior temperature increases, and the first time is also appropriately extended.

[0126] In a possible embodiment, remote control issuing devices usually have unique identifiers. When a user registers or first uses the remote control function, these device identifiers are bound to the user information and stored in the user information database. When receiving remote control data, extract the remote control issuing device identifier contained in the data and match it with the records in the user information database to determine the target user who initiates the remote control. If there are multiple possible users through device identifier matching, for example, the same user uses multiple devices for remote control, further confirm by combining the remote control issuing location information. Compare the common location information recorded in the database for different users, such as home address, work address, etc., and select the user who is most matched with the remote control issuing location as the target user.

[0127] In a possible embodiment, if the target user uses a device with a positioning function for remote control, after the vehicle control system or the server receives the remote control data containing location information, the data is parsed to extract the location data of the target user.

[0128] In a possible embodiment, use the map navigation service to plan a path with the current location of the target user and the location of the target vehicle as the starting point and the ending point. The navigation service can calculate multiple possible driving routes according to real-time traffic conditions information, such as road congestion, traffic accidents, etc., and estimate the time required to reach the target vehicle along each route under the current traffic conditions. Select the shortest or most reasonable time from the multiple estimated arrival times as the second time for the target user to reach the target vehicle. For example, if there is a route with a shorter distance but serious congestion and a route with a slightly longer distance but smooth traffic conditions, after comprehensive consideration, select the arrival time corresponding to the route with smooth traffic conditions as the second time.

[0129] In a possible embodiment, the instruction execution time is determined through a certain calculation logic according to the remote control issuance time, the first time, and the second time. If it is required that the vehicle just finishes executing the second target instruction when the target user arrives at the vehicle, then the instruction execution time is the remote control issuance time plus the second time minus the first time.

[0130] In a possible embodiment, refer to Figure 4 , Figure 4 which is a schematic diagram of an application scenario of a vehicle diagnosis method based on intelligent control of the vehicle owner provided by an embodiment of the present application. As Figure 4 shown, the instruction execution distance is determined according to the position data of the target user and the first time required for executing the second target instruction. When the distance between the target user and the target vehicle is less than the instruction execution distance, the second target instruction starts to be executed.

[0131] In the embodiment of the present application, by accurately calculating the instruction execution time, it is ensured that when the user arrives at the vehicle, the vehicle has completed corresponding operations, such as preheating the engine well, adjusting the interior temperature well, etc., providing a comfortable and convenient travel experience for the user, reducing the user's waiting time, and making the user feel the intelligence and considerate service of the vehicle control system. Moreover, according to the specific position and expected arrival time of each user to determine the instruction execution time, personalized service is realized, meeting the needs of different users in different scenarios, improving the user's satisfaction with the vehicle remote control system. Considering the time required for instruction execution and the user's arrival time, executing instructions too early or too late is avoided, reducing unnecessary energy consumption of the vehicle and the operation time of equipment. The accurate time planning helps the vehicle control system reasonably arrange tasks, avoid instruction execution conflicts, and improve the operation efficiency and stability of the entire remote control system. Considering multiple factors such as the remote control issuance time, the instruction execution time, the user's position, and the arrival time comprehensively to determine the final instruction execution time reflects the powerful data analysis and decision-making ability of the system, making the vehicle remote control system more intelligent and flexible, and capable of adapting to complex and changeable actual application scenarios. During the process of determining the time, dynamic information such as traffic conditions is obtained in real time, enabling the system to adjust the instruction execution time in a timely manner according to the actual situation, better coping with various emergencies, and providing reliable services for users.

[0132] In a possible embodiment, refer to Figure 5 , Figure 5 which is a flowchart of a vehicle diagnosis method based on a voice assistant and intelligent control of the vehicle owner provided by an embodiment of the present application. After the vehicle owner issues a voice command, the command is parsed through the voice assistant module, vehicle data is collected through the OBD-II interface, intelligent diagnosis and health report generation are performed after the data verification and analysis module, and then intelligent health reminder push is performed to enable the vehicle owner to receive voice or application feedback, or the vehicle is operated to execute the command through the remote control module.

[0133] In a possible embodiment, refer to Figure 6 , Figure 6 which is a timing diagram of a vehicle diagnosis method based on a voice assistant and intelligent control of the vehicle owner provided by an embodiment of the present application. The vehicle owner issues a voice command to the voice assistant module, such as "What's wrong with my car?", the voice assistant module requests vehicle diagnosis data from the OBD-II interface to obtain fault codes and real-time data, the OBD-II interface sends the data to the data verification and analysis module, verifies and analyzes the data through the data verification and analysis module, generates a diagnosis report and a health reminder, feeds back a fault analysis and health advice through the health report and reminder module, the vehicle owner issues a remote control command through the remote control module, such as: "Start self-check", after performing the remote control operation, communicate with the vehicle through the OBD-II interface to perform a self-check operation.

[0134] In a possible embodiment, intelligent analysis algorithms and health report generation can also be performed according to the diagnosis information, protecting machine learning or intelligent analysis algorithms for analyzing vehicle data, as well as health reports and personalized push mechanisms generated based on these algorithms. This technology can actively push accurate maintenance and repair suggestions to the vehicle owner through the analysis of fault codes, driving behaviors, and repair histories.

[0135] In summary, in the embodiment of the present application, first, voice data, personnel data, and historical interaction data of the target vehicle are obtained, where the personnel data is used to reflect at least one of the following personnel information: the number of personnel, personnel identity, personnel location, personnel behavior, personnel acoustic characteristics, and the historical interaction data is used to reflect at least one of the following interaction information: historical driving habits, historical voice command records. Then, the voice data is parsed according to the personnel data and historical interaction data to obtain a target statement, and the preset instruction database is searched according to the target statement to obtain a first target instruction corresponding to the target statement. Next, vehicle data of the target vehicle is obtained according to the first target instruction, and the target vehicle is diagnosed according to the vehicle data to obtain diagnosis information. Finally, the diagnosis information is pushed. Thus, by obtaining voice data and determining the first target instruction corresponding to the voice data, diagnosing according to the first target instruction to obtain diagnosis information, and pushing the diagnosis information, when the vehicle owner needs to diagnose the vehicle, the manual operation steps of the vehicle owner can be simplified, and the vehicle diagnosis efficiency can be improved.

[0136] The method of the embodiment of the present invention is elaborated in detail above, and the device of the embodiment of the present invention is provided below.

[0137] Refer to Figure 7 , Figure 7 which is a schematic structural diagram of a vehicle diagnosis device based on an on-board diagnostic system provided by an embodiment of the present application. As Figure 7As shown, the vehicle diagnostic device 800 based on the on-vehicle automatic diagnostic system includes an acquisition unit 801 and a processing unit 802;

[0138] The acquisition unit 801 is configured to acquire voice data, personnel data, and historical interaction data of the target vehicle; the personnel data is used to reflect at least one of the following personnel information: the number of personnel, personnel identity, personnel location, personnel behavior, and personnel acoustic characteristics; the historical interaction data is used to reflect at least one of the following interaction information: historical driving habits and historical voice command records;

[0139] The processing unit 802 is configured to parse the voice data according to the personnel data and the historical interaction data to obtain a target statement;

[0140] Search a preset instruction database according to the target statement to obtain a first target instruction corresponding to the target statement;

[0141] Obtain vehicle data of the target vehicle according to the first target instruction;

[0142] Diagnose the target vehicle according to the vehicle data to obtain diagnostic information;

[0143] Push the diagnostic information.

[0144] In a possible embodiment, when parsing the voice data according to the personnel data and the historical interaction data to obtain a target statement, the processing unit 802 is specifically configured to:

[0145] Decompose the voice data into m voice segments according to the personnel data and the historical interaction data; m is a positive integer; the m voice segments correspond to m personnel;

[0146] Determine the acoustic state sequence of each of the m voice segments to obtain m acoustic state sequences; the m acoustic state sequences are used to reflect the voice change information of the m voice segments;

[0147] Determine the voice recognition result corresponding to each of the m voice segments according to the m acoustic state sequences to obtain m voice recognition results;

[0148] Determine the time sequence and logical association relationship of the m voice recognition results;

[0149] Determine the instruction authority of each of the m personnel to obtain m instruction authorities;

[0150] Determine the instruction probability of each of the m voice recognition results according to the m instruction authorities, time sequence, and logical association relationship to obtain m instruction probabilities;

[0151] Determine the voice recognition result corresponding to the maximum value among the m instruction probabilities as the target statement.

[0152] In a possible embodiment, the preset instruction database is searched according to the target statement, and a first target instruction corresponding to the target statement is obtained. The processing unit 802 is specifically configured to:

[0153] Extract n voice keywords from the target statement; n is a positive integer;

[0154] Determine a first score for each voice keyword among the n voice keywords to obtain n first scores; the n first scores are used to reflect the degree of association between each voice keyword among the n voice keywords and the instruction database;

[0155] Determine the syntactic relationship between the n voice keywords;

[0156] Determine a second score of the target statement according to the n first scores and the syntactic relationship; the second score is used to reflect the accurate reference degree of the target statement;

[0157] Search the instruction database according to the second score and the target statement to obtain k instruction search results; k is a natural number;

[0158] Determine the instruction level of each instruction search result among the k instruction search results to obtain k instruction levels; the k instruction levels are used to reflect the instruction execution order relationship between the k instruction search results;

[0159] Determine the first target instruction from the k instruction search results according to the k instruction levels.

[0160] In a possible embodiment, when searching the instruction database according to the second score and the target statement to obtain k instruction search results, the processing unit 802 is specifically configured to:

[0161] If the second score is lower than the preset accuracy score threshold, obtain the sensor data of the target vehicle; the sensor data includes at least one of the following: vehicle speed, steering wheel angle, driving time;

[0162] Semantically expand the target statement according to the sensor data and the historical interaction data to obtain an expanded statement;

[0163] Search the instruction database according to the expanded statement to obtain k instruction search results;

[0164] If the second score is higher than or equal to the accuracy score threshold, search the instruction database according to the n first scores and the n voice keywords to obtain k instruction search results.

[0165] In a possible embodiment, when pushing diagnostic information, the processing unit 802 is specifically configured to:

[0166] Obtain the information length of the diagnostic information;

[0167] Determine the information level of the diagnostic information; the information level is used to reflect the importance or urgency of the diagnostic information;

[0168] Determine the voice broadcast requirement score of the diagnostic information according to the information length and the information level;

[0169] If the voice broadcast requirement score is higher than or equal to the preset voice broadcast requirement score threshold, extract the voice broadcast content from the diagnostic information;

[0170] Push the voice broadcast content through voice;

[0171] If the voice broadcast requirement score is lower than the voice broadcast requirement score threshold, generate a diagnostic analysis report according to the diagnostic information;

[0172] Push the diagnostic analysis report through text.

[0173] In a possible embodiment, the processing unit 802 is further configured to:

[0174] Obtain the remote control data of the target vehicle; the remote control data is used to perform at least one of the following functions: vehicle self-check, in-vehicle temperature adjustment, vehicle relocation;

[0175] Determine the second target instruction according to the remote control data and the instruction database;

[0176] Determine the remote control issuance time, remote control issuance location and remote control issuance device according to the remote control data;

[0177] Determine the instruction execution time of the second target instruction according to the remote control issuance time, remote control issuance location and remote control issuance device;

[0178] Control the target vehicle to execute the second target instruction according to the instruction execution time.

[0179] In a possible embodiment, to determine the instruction execution time of the second target instruction according to the remote control issuance time, remote control issuance location and remote control issuance device, the processing unit 802 is specifically configured to:

[0180] Determine the first time required for the execution of the second target instruction;

[0181] Determine the target user corresponding to the remote control data according to the remote control issuance location and the remote control issuance device;

[0182] Obtain the location data of the target user;

[0183] Determine the second time for the target user to reach the target vehicle according to the location data;

[0184] Determine the instruction execution time according to the remote control issuing time, the first time, and the second time.

[0185] Refer to Figure 8 , Figure 8 which is a schematic structural diagram of an electronic device provided by an embodiment of the present application. As Figure 8 shown, the electronic device 900 includes a transceiver 901, a processor 902, and a memory 903, which are connected to each other through a bus 904. The memory 903 is used to store computer programs and data, and can transmit the data stored in the memory 903 to the processor 902. Among them, the electronic device can be the vehicle diagnostic device 800 based on the on-board automatic diagnostic system described above, and the processor 902 can be the obtaining unit 801 and the processing unit 802 described above.

[0186] The processor 902 is used to read the computer program in the memory 903 and perform the following operations:

[0187] Obtain voice data, personnel data, and historical interaction data of the target vehicle; the personnel data is used to reflect at least one of the following personnel information: the number of personnel, the identity of the personnel, the location of the personnel, the behavior of the personnel, the acoustic characteristics of the personnel; the historical interaction data is used to reflect at least one of the following interaction information: historical driving habits, historical voice instruction records;

[0188] Analyze the voice data according to the personnel data and the historical interaction data to obtain a target statement;

[0189] Search a preset instruction database according to the target statement to obtain a first target instruction corresponding to the target statement;

[0190] Obtain vehicle data of the target vehicle according to the first target instruction;

[0191] Diagnose the target vehicle according to the vehicle data to obtain diagnostic information;

[0192] Push the diagnostic information.

[0193] In a possible embodiment, when analyzing the voice data according to the personnel data and the historical interaction data to obtain a target statement, the processor 902 is specifically used to perform the following operations:

[0194] Decompose the voice data into m voice segments according to the personnel data and the historical interaction data; m is a positive integer; the m voice segments correspond to m personnel;

[0195] Determine the acoustic state sequence of each voice segment in the m voice segments to obtain m acoustic state sequences; the m acoustic state sequences are used to reflect the voice change information of the m voice segments;

[0196] Determine the speech recognition result corresponding to each speech segment among the m speech segments according to the m acoustic state sequences, and obtain m speech recognition results;

[0197] Determine the time sequence and logical association relationship of the m speech recognition results;

[0198] Determine the instruction permission of each person among the m persons, and obtain m instruction permissions;

[0199] Determine the instruction probability of each speech recognition result among the m speech recognition results according to the m instruction permissions, time sequence and logical association relationship, and obtain m instruction probabilities;

[0200] Determine the speech recognition result corresponding to the maximum value among the m instruction probabilities as the target statement.

[0201] In a possible embodiment, search the preset instruction database according to the target statement to obtain a first target instruction corresponding to the target statement. The processor 902 is specifically configured to perform the following operations:

[0202] Extract n speech keywords from the target statement; n is a positive integer;

[0203] Determine the first score of each speech keyword among the n speech keywords, and obtain n first scores; the n first scores are used to reflect the association degree of each speech keyword among the n speech keywords with the instruction database;

[0204] Determine the syntactic relationship between the n speech keywords;

[0205] Determine the second score of the target statement according to the n first scores and syntactic relationship; the second score is used to reflect the accurate reference degree of the target statement;

[0206] Search the instruction database according to the second score and the target statement to obtain k instruction search results; k is a natural number;

[0207] Determine the instruction level of each instruction search result among the k instruction search results, and obtain k instruction levels; the k instruction levels are used to reflect the instruction execution sequence relationship among the k instruction search results;

[0208] Determine the first target instruction from the k instruction search results according to the k instruction levels.

[0209] In a possible embodiment, search the instruction database according to the second score and the target statement to obtain k instruction search results. The processor 902 is specifically configured to perform the following operations:

[0210] If the second score is lower than the preset accuracy score threshold, obtain the sensor data of the target vehicle; the sensor data includes at least one of the following: vehicle speed, steering wheel angle, driving time;

[0211] Semantically expand the target statement according to the sensor data and historical interaction data to obtain an expanded statement;

[0212] Search the instruction database according to the expanded statement to obtain k instruction search results;

[0213] If the second score is higher than or equal to the accuracy score threshold, search the instruction database according to the n first scores and n voice keywords to obtain k instruction search results.

[0214] In a possible embodiment, to push diagnostic information, the processor 902 is specifically configured to perform the following operations:

[0215] Obtain the information length of the diagnostic information;

[0216] Determine the information level of the diagnostic information; the information level is used to reflect the importance or urgency of the diagnostic information;

[0217] Determine the voice broadcast requirement score of the diagnostic information according to the information length and the information level;

[0218] If the voice broadcast requirement score is higher than or equal to the preset voice broadcast requirement score threshold, extract the voice broadcast content from the diagnostic information;

[0219] Push the voice broadcast content through voice;

[0220] If the voice broadcast requirement score is lower than the voice broadcast requirement score threshold, generate a diagnostic analysis report according to the diagnostic information;

[0221] Push the diagnostic analysis report through text.

[0222] In a possible embodiment, the processor 902 is further configured to perform the following operations:

[0223] Obtain the remote control data of the target vehicle; the remote control data is used to perform at least one of the following functions: vehicle self-check, in-vehicle temperature adjustment, vehicle relocation;

[0224] Determine the second target instruction according to the remote control data and the instruction database;

[0225] Determine the remote control issuance time, remote control issuance location, and remote control issuance device according to the remote control data;

[0226] Determine the instruction execution time of the second target instruction according to the remote control issuance time, remote control issuance location, and remote control issuance device;

[0227] Execute the second target instruction for the target vehicle according to the instruction execution time.

[0228] In a possible embodiment, determine the instruction execution time of the second target instruction according to the remote control sending time, remote control sending location, and remote control sending device. The processor 902 is specifically configured to perform the following operations:

[0229] Determine the first time required for the execution of the second target instruction;

[0230] Determine the target user corresponding to the remote control data according to the remote control sending location and remote control sending device;

[0231] Obtain the location data of the target user;

[0232] Determine the second time for the target user to reach the target vehicle according to the location data;

[0233] Determine the instruction execution time according to the remote control sending time, the first time, and the second time.

[0234] The embodiment of the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement some or all of the steps of any one of the vehicle diagnosis methods based on an on-vehicle automatic diagnosis system as described in the above method embodiments.

[0235] The embodiment of the present application further provides a computer program product. The computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps of any one of the vehicle diagnosis methods based on an on-vehicle automatic diagnosis system as described in the above method embodiments.

[0236] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present application is not limited by the described action sequence, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0237] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0238] In several embodiments provided in this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of the device or module can be in electrical or other forms.

[0239] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they can be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0240] In addition, in each embodiment of this application, the functional modules can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in one module. The above integrated modules can be implemented in the form of hardware or in the form of software program modules.

[0241] If the above integrated module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. 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 this 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 enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of this application. And the aforementioned memory includes: USB flash drives, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), mobile hard disks, magnetic disks, or optical discs and other media that can store program codes.

[0242] The above has introduced the embodiments of this application in detail. Specific examples are used in this article to elaborate on the principle and implementation manner of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. A vehicle diagnosis method based on an on-board automatic diagnosis system, characterized in that: include: Acquire voice data, personnel data and historical interaction data of the target vehicle; the personnel data is used to reflect at least one of the following personnel information: number of personnel, personnel identity, personnel location, personnel behavior, and personnel acoustic characteristics; The historical interaction data is used to reflect at least one of the following interaction information: historical driving habits, historical voice command records; Parsing the voice data according to the personnel data and the historical interaction data to obtain a target sentence; Searching a preset instruction database according to the target sentence to obtain a first target instruction corresponding to the target sentence; Acquiring vehicle data of the target vehicle according to the first target instruction; diagnose the target vehicle according to the vehicle data to obtain diagnostic information; The diagnostic information is pushed.

2. The method according to claim 1, characterized in that The step of parsing the voice data according to the personnel data and the historical interaction data to obtain a target sentence includes: Decomposing the voice data into m voice segments according to the personnel data and the historical interaction data; m is a positive integer; and the m voice segments correspond to m personnel; Determine the acoustic state sequence of each of the m speech segments to obtain m acoustic state sequences; the m acoustic state sequences are used to reflect the speech change information of the m speech segments; Determine a speech recognition result corresponding to each of the m speech segments according to the m acoustic state sequences, to obtain m speech recognition results; Determining the time sequence and logical association relationship of the m speech recognition results; Determine the instruction authority of each of the m personnel to obtain m instruction authorities; Determine the instruction probability of each of the m speech recognition results according to the m instruction permissions, the time sequence and the logical association relationship to obtain m instruction probabilities; The target sentence is determined by the speech recognition result corresponding to the maximum value among the m instruction probabilities.

3. The method according to claim 1, characterized in that The step of searching a preset instruction database according to the target sentence to obtain a first target instruction corresponding to the target sentence includes: Extracting n phonetic keywords from the target sentence; n is a positive integer; Determine a first score of each of the n voice keywords to obtain n first scores; the n first scores are used to reflect the degree of association between each of the n voice keywords and the instruction database; Determining the syntactic relationship between the n phonetic keywords; Determining a second score of the target sentence according to the n first scores and the syntactic relationship; the second score is used to reflect the reference accuracy of the target sentence; Search the instruction database according to the second score and the target sentence to obtain k instruction search results, where k is a natural number; Determine the instruction level of each instruction search result in the k instruction search results to obtain k instruction levels; the k instruction levels are used to reflect the instruction execution order relationship between the k instruction search results; The first target instruction is determined from the k instruction search results according to the k instruction levels.

4. The method according to claim 3, characterized in that The step of searching the instruction database according to the second score and the target sentence to obtain k instruction search results includes: If the second score is lower than a preset accuracy score threshold, obtaining sensor data of the target vehicle; the sensor data includes at least one of the following: vehicle speed, steering wheel angle, and driving time; semantically expand the target sentence according to the sensor data and the historical interaction data to obtain an expanded sentence; Search the instruction database according to the extended statement to obtain the k instruction search results; If the second score is higher than or equal to the accuracy score threshold, the instruction database is searched according to the n first scores and the n voice keywords to obtain the k instruction search results.

5. The method according to claim 1, characterized in that The pushing of the diagnostic information includes: Obtaining the information length of the diagnostic information; Determining an information level of the diagnostic information; the information level is used to reflect the importance or urgency of the diagnostic information; Determining a voice broadcast requirement score for the diagnostic information according to the information length and the information level; If the voice broadcast demand score is higher than or equal to a preset voice broadcast demand score threshold, extracting voice broadcast content from the diagnostic information; Pushing the voice broadcast content via voice; If the voice broadcast demand score is lower than the voice broadcast demand score threshold, generating a diagnostic analysis report according to the diagnostic information; The diagnostic analysis report is pushed via text.

6. The method according to any one of claims 1 to 5, characterized in that: The method further comprises: Acquiring remote control data of the target vehicle; the remote control data is used to perform at least one of the following functions: vehicle self-check, in-vehicle temperature adjustment, and vehicle relocation; Determine a second target instruction according to the remote control data and the instruction database; Determine the remote control sending time, remote control sending location and remote control sending device according to the remote control data; Determining the instruction execution time of the second target instruction according to the remote control issuing time, the remote control issuing location and the remote control issuing device; The target vehicle is controlled to execute the second target command according to the command execution time.

7. The method according to claim 6, characterized in that The determining the instruction execution time of the second target instruction according to the remote control issuing time, the remote control issuing location and the remote control issuing device includes: determining a first time required for execution of the second target instruction; Determine a target user corresponding to the remote control data according to the remote control sending location and the remote control sending device; Acquire location data of the target user; Determining a second time when the target user arrives at the target vehicle according to the location data; The instruction execution time is determined according to the remote control sending time, the first time and the second time.

8. A vehicle diagnostic device based on an on-board automatic diagnostic system, characterized in that: The device comprises an acquisition unit and a processing unit; The acquisition unit is used to acquire voice data, personnel data and historical interaction data of the target vehicle; the personnel data is used to reflect at least one of the following personnel information: number of personnel, personnel identity, personnel location, personnel behavior, and personnel acoustic characteristics; The historical interaction data is used to reflect at least one of the following interaction information: historical driving habits, historical voice command records; The processing unit is used to parse the voice data according to the personnel data and the historical interaction data to obtain a target sentence; Searching a preset instruction database according to the target sentence to obtain a first target instruction corresponding to the target sentence; Acquiring vehicle data of the target vehicle according to the first target instruction; diagnose the target vehicle according to the vehicle data to obtain diagnostic information; The diagnostic information is pushed.

9. An electronic device, characterized in that: include: A processor and a memory, the processor is connected to the memory, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor is caused to perform the method according to any one of claims 1 to 7.

Images