Diagnostic script generation method and apparatus, electronic device, computer-readable storage medium

Abstract

The present disclosure provides a diagnostic script generation method and device, relates to the technical field of computers, in particular to the technical field of large models, deep learning and the like. The specific implementation scheme is as follows: acquiring vehicle fault information; generating a fault correlation factor based on the vehicle fault information and a factor large model; and generating a diagnostic script based on the fault correlation factor and a diagnostic large model, thereby improving user experience.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of computers, in particular to the technical field of large models, deep learning, image processing, and the like, and more particularly to a diagnosis script generation method and device, an electronic device, a computer readable storage medium, and a computer program product. BACKGROUND

[0002] In the field of vehicle remote diagnosis, a diagnosis script is generally used, which is a script used to guide the communication between a diagnostic tool and a vehicle control unit and to perform a specific diagnosis task in the vehicle diagnosis process; an operator selects a ready diagnosis script in the cloud according to diagnosis requirements, and sends the diagnosis script to the vehicle end for execution, thereby realizing the diagnosis of vehicle end faults. SUMMARY

[0003] The present disclosure provides a diagnosis script generation method and device, an electronic device, a computer readable storage medium, and a computer program product.

[0004] According to a first aspect, a diagnosis script generation method is provided, which includes: obtaining vehicle fault information; generating a fault-related factor based on the vehicle fault information and a factor large model; and generating a diagnosis script based on the fault-related factor and a diagnosis large model.

[0005] According to a second aspect, a diagnosis script generation device is provided, which includes: an obtaining unit configured to obtain vehicle fault information; a factor generation unit configured to generate a fault-related factor based on the vehicle fault information and a factor large model; and a script generation unit configured to generate a diagnosis script based on the fault-related factor and a diagnosis large model.

[0006] According to a third aspect, an electronic device is provided, which includes: at least one processor; and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in any implementation manner of the first aspect.

[0007] According to a fourth aspect, a non-transitory computer readable storage medium storing computer instructions is provided, the computer instructions being used to cause a computer to perform the method described in any implementation manner of the first aspect.

[0008] According to a fifth aspect, a computer program product is provided, which includes a computer program, and the computer program, when executed by a processor, implements the method described in any implementation manner of the first aspect.

[0009] The embodiment of the disclosure provides a diagnostic script generation method and device, first, vehicle fault information is acquired; second, based on the vehicle fault information and the factor large model, a fault correlation factor is generated; finally, based on the fault correlation factor and the diagnosis large model, a diagnostic script is generated. Therefore, based on the vehicle fault information, the fault correlation factor is automatically generated, the diagnostic script is generated based on the fault correlation factor, and the diagnostic script generation efficiency is improved; without the need for operation personnel to find information, the user experience is improved.

[0010] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the disclosure, nor is it intended to limit the scope of the disclosure. Other features of the disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0011] The accompanying drawings are used to better understand the scheme, and do not constitute a limitation on the disclosure. Among them:

[0012] Figure 1 is a flowchart of one embodiment of the diagnostic script generation method according to the disclosure;

[0013] Figure 2 is a structural schematic diagram of the diagnostic script generation process of the disclosure;

[0014] Figure 3 is a flowchart of another embodiment of the diagnostic script generation method according to the disclosure;

[0015] Figure 4 is a flowchart of another embodiment of the diagnostic script generation method according to the disclosure;

[0016] Figure 5 is a structural schematic diagram of one embodiment of the diagnostic script generation device according to the disclosure;

[0017] Figure 6 is a block diagram of an electronic device for implementing the diagnostic script generation method of the embodiment of the disclosure. DETAILED DESCRIPTION

[0018] The exemplary embodiments of the disclosure are described below in conjunction with the accompanying drawings, including various details of the embodiments of the disclosure to help understanding, which should be considered only as exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the disclosure. Also, in order to be clear and concise, the description in the following description omits the description of well-known functions and structures.

[0019] The core purpose of the automobile remote diagnosis process is to diagnose and analyze the fault and troubleshoot the problem. The vehicle fault information reported when the fault occurs mainly includes DTC (Diagnostic Trouble Code), and the part information (part name, hardware version, software version, etc.) of the fault. The diagnostic script is also for a specific DTC, and the related signals are obtained in the vehicle. The conclusion is obtained after comprehensive judgment and analysis based on the value of the signal reading.

[0020] The vehicle manufacturer will have a fault detection manual for all parts and all possible DTC faults, and will also accumulate many past case records for troubleshooting different faults.

[0021] Currently, the diagnostic script in the industry is manually written by diagnostic operation personnel and uploaded to the platform. A large amount of existing materials (fault troubleshooting manual and past cases) need to be consulted during the manual writing process, and the writing is done in combination with the fault troubleshooting and maintenance experience of the writer. The completely manual writing of the diagnostic script has high technical requirements for the script writer, and the script writer needs to consult the fault troubleshooting manual and past cases during the script writing process. The general fault troubleshooting manual is very complex, and the document volume is large, so the manual consultation workload is large. There are many past fault troubleshooting and maintenance cases, and the manual troubleshooting workload is huge.

[0022] In view of the defects in the prior art, the present disclosure provides a diagnostic script generation method, which can automatically generate a diagnostic script and improve user experience, Figure 1 An embodiment of the diagnostic script generation method according to the present disclosure is shown in flowchart 100, which includes the following steps:

[0023] Step 101, obtaining vehicle fault information.

[0024] In this embodiment, the execution subject on which the diagnostic script generation method runs can be a cloud, such as Figure 2 As shown in the figure, the cloud is a remote server, and the cloud server realizes vehicle fault information processing, diagnostic script generation, and delivery functions.

[0025] In this embodiment, the vehicle fault information is the information reported when the vehicle fault occurs, and the vehicle fault information includes DTC and part information of the fault. The part information includes part name, hardware version, software version, etc. The vehicle fault information can be information stored independently by the cloud, such as Figure 2 The vehicle fault information can also be fault information G sent by the vehicle diagnostic program of the vehicle to the execution subject when the fault occurs.

[0026] In the technical solutions of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of vehicle fault information are performed after authorization and comply with relevant laws and regulations. The vehicle fault information related to the user is obtained after the user's permission, and the generation of the diagnostic script related to the user is a process performed under the condition of privacy protection.

[0027] In step 102, a fault correlation factor is generated based on the vehicle fault information and the factor large model.

[0028] In this embodiment, the fault correlation factor is a factor associated with the vehicle fault information. The fault correlation factor can include a fault detection process sub-factor and a detection or correlation sub-factor in the fault detection process. The fault detection process sub-factor is used to implement the detection process of the fault corresponding to the vehicle fault information. The detection or correlation sub-factor can be an instruction involved in the monitoring process corresponding to the fault detection process sub-factor.

[0029] In this embodiment, the factor large model is a generative large model. By inputting the factor prompt word for generating the fault correlation factor and the vehicle fault information into the factor large model, the fault correlation factor output by the factor large model can be obtained. The factor large model generates a factor script by analyzing the vehicle fault information under the prompt of the factor prompt word. The step 102 includes inputting the fault correlation factor and the factor generation prompt word into the factor large model to obtain the fault correlation factor output by the factor large model.

[0030] In step 103, a diagnostic script is generated based on the fault correlation factor and the diagnostic large model.

[0031] In this embodiment, the diagnostic large model is a generative large model. By inputting the diagnostic prompt word for generating the script and the fault correlation factor into the diagnostic large model, the diagnostic script output by the diagnostic large model can be obtained. The diagnostic large model generates a diagnostic script by analyzing the fault correlation factor under the prompt of the diagnostic prompt word.

[0032] In this embodiment, the diagnostic script is a script for diagnosing the fault corresponding to the vehicle fault information. By inputting the fault correlation factor into the diagnostic large model, the diagnostic script output by the diagnostic large model can be obtained.

[0033] The diagnostic script generation method provided by the embodiments of the present disclosure first acquires vehicle fault information. Then, based on the vehicle fault information and the factor large model, a fault correlation factor is generated. Finally, based on the fault correlation factor and the diagnostic large model, a diagnostic script is generated. Thus, based on the vehicle fault information, the fault correlation factor is automatically generated, and based on the fault correlation factor, the diagnostic script is generated, which improves the efficiency of generating the diagnostic script. The operator does not need to search for information, and the user experience is improved.

[0034] In some embodiments of the present disclosure, the generating the fault-related factor based on the vehicle fault information and the factor large model comprises: detecting whether there is a diagnostic script matched with the vehicle fault information based on a diagnostic script library; and in response to no diagnostic script matched with the vehicle fault information being detected, generating the fault-related factor based on the vehicle fault information and the factor large model.

[0035] In the embodiment, as shown in FIG. 1, the diagnostic script library is a database in which a plurality of diagnostic scripts are stored in the cloud, and each diagnostic script in the diagnostic script library can have corresponding fault information. An execution subject on which the diagnostic script generation method runs compares the fault information in the diagnostic script library with the vehicle fault information to detect whether there is a diagnostic script matched with the vehicle fault information in the diagnostic script library. When the similarity between the fault information in the diagnostic script library and the vehicle fault information is greater than or equal to a similarity threshold (for example, 90%), the diagnostic script corresponding to the fault information in the diagnostic script library is determined as the diagnostic script matched with the vehicle fault information. When the similarity between the fault information in the diagnostic script library and the vehicle fault information is less than the similarity threshold, it is determined that no diagnostic script matched with the vehicle fault information is detected. Figure 2 The diagnostic script generation method provided by the optional implementation manner detects whether there is a diagnostic script matched with the vehicle fault information based on the diagnostic script library, and in response to no diagnostic script matched with the vehicle fault information being detected, generates the fault-related factor based on the vehicle fault information, which provides a means for directly obtaining the diagnostic script and improves the reliability of the diagnostic script.

[0036] Optionally, the diagnostic script generation method further comprises: in response to detecting that the diagnostic script library has the diagnostic script matched with the vehicle fault information, directly adopting the diagnostic script.

[0037] Optionally, before the generating the fault-related factor based on the vehicle fault information, the diagnostic script generation method further comprises: in response to no diagnostic script matched with the vehicle fault information being detected from the diagnostic script library, detecting a diagnostic script matched with the vehicle fault information from information other than the diagnostic script library, and if no diagnostic script matched with the vehicle fault information is detected, generating the fault-related factor based on the vehicle fault information.

[0038]

[0039] ​In some optional implementations of the present disclosure, the detection of whether there is a diagnostic script matching the vehicle fault information based on the diagnostic script library comprises: performing vectorization processing on the diagnostic scripts in the diagnostic script library to obtain a first vector library; performing vectorization processing on the vehicle fault information to obtain a vehicle fault vector; matching the vehicle fault vector with the vectors in the first vector library; and in response to a successful matching of the vehicle fault vector with the vectors in the first vector library, determining that there is a diagnostic script matching the vehicle fault information.

[0040] In the optional implementation, as shown in Figure 2 the diagnostic scripts in the diagnostic script library are subjected to vectorization processing, and the vehicle fault information is subjected to vectorization processing, and the vectors in the first vector library obtained are matched with the vehicle fault vector, thereby achieving the vector retrieval in Figure 2 Through the vector retrieval, the diagnostic scripts related to the vehicle fault information in the diagnostic script library can be searched in all directions, and the accuracy and comprehensiveness of the diagnostic script search are improved.

[0041] The diagnostic script generation method provided by the optional implementation comprises: performing vectorization processing on the diagnostic scripts in the diagnostic script library to obtain a first vector library; performing vectorization processing on the vehicle fault information to obtain a vehicle fault vector; matching the vehicle fault vector with the vectors in the first vector library; and in response to a successful matching of the vehicle fault vector with the vectors in the first vector library, determining that there is a diagnostic script matching the vehicle fault information. Thus, through the vector retrieval, the accuracy and comprehensiveness of the detection of whether there is a diagnostic script matching the vehicle fault information are improved.

[0042] In some optional implementations of the present disclosure, the detection of whether there is a diagnostic script matching the vehicle fault information based on the diagnostic script library comprises: performing vectorization processing on the diagnostic scripts in the diagnostic script library to obtain a first vector library; performing vectorization processing on the vehicle fault information to obtain a vehicle fault vector; matching the vehicle fault vector with the vectors in the first vector library; and in response to a successful matching of the vehicle fault vector with the vectors in the first vector library, determining that there is a diagnostic script matching the vehicle fault information.

[0043] The method for detecting a diagnostic script provided by the optional implementation comprises: performing vectorization processing on the diagnostic scripts in the diagnostic script library to obtain a first vector library; performing vectorization processing on the vehicle fault information to obtain a vehicle fault vector; matching the vehicle fault vector with the vectors in the first vector library; and in response to a successful matching of the vehicle fault vector with the vectors in the first vector library, determining that there is a diagnostic script matching the vehicle fault information. Thus, through the vector matching, the diagnostic script matching the vehicle fault information can be comprehensively determined, and the reliability and accuracy of the detection of the diagnostic script matching the vehicle fault information are improved.

[0044] Optionally, the detecting whether there is a diagnostic script matched with the vehicle fault information based on the diagnostic script library comprises: matching the diagnostic script in the diagnostic script library with the vehicle fault vector; and in response to a successful matching of the vehicle fault vector with the diagnostic script in the diagnostic script library, determining that there is a diagnostic script matched with the vehicle fault information.

[0045] Optionally, the detecting whether there is a diagnostic script matched with the vehicle fault information based on the diagnostic script library comprises: determining a fault instance matched with the diagnostic script in the diagnostic script library to obtain a fault instance library; and matching the fault instance in the fault instance library with the vehicle fault information, and in response to a successful matching of the fault instance with the vehicle fault information, determining that there is a diagnostic script matched with the vehicle fault information.

[0046] In some embodiments of the present disclosure, the vehicle fault information is fault information sent by a target vehicle, and the diagnostic script generation method further comprises: sending the diagnostic script to the target vehicle. Specifically, Figure 3 A flow 300 of another embodiment of the diagnostic script generation method according to the present disclosure is shown, and the diagnostic script generation method comprises the following steps:

[0047] Step 301: obtaining vehicle fault information.

[0048] In this embodiment, the vehicle fault information is fault information sent by a target vehicle to an execution subject on which the diagnostic script generation method runs after a fault occurs.

[0049] Step 302: generating a fault correlation factor based on the vehicle fault information and a factor large model.

[0050] Step 303: generating a diagnostic script based on the fault correlation factor and a diagnostic large model.

[0051] It should be understood that the operations and features in the above steps 301-303 correspond to the operations and features in the steps 101-103 respectively, and therefore the description of the operations and features in the steps 101-103 also applies to the steps 301-303, which will not be repeated here.

[0052] Step 304: sending the diagnostic script to the target vehicle.

[0053] In this embodiment, after the target vehicle obtains the diagnostic script issued by the execution subject, the target vehicle obtains vehicle signals related to the diagnostic script on the vehicle itself, and gives a diagnostic conclusion based on a comprehensive analysis of the vehicle according to the values of the vehicle signals.

[0054] The diagnostic script generation method provided in the embodiment can send the diagnostic script to the target vehicle after obtaining the diagnostic script, so that the target vehicle can obtain the diagnostic script in time and diagnose itself, thereby improving the efficiency of vehicle diagnosis.

[0055] Optionally, the diagnostic script generation method further includes: detecting whether the diagnostic script is a script executed after power-off; and in response to the diagnostic script being a script executed after power-off, sending power-off prompt information to the target vehicle.

[0056] The diagnostic script generation method provided in the optional implementation manner can send power-off prompt information to the target vehicle after detecting that the diagnostic script is a script executed after power-off, so that the target vehicle can be powered off in time after receiving and running the diagnostic script, thereby improving the reliability of running the diagnostic script.

[0057] In some embodiments of the present disclosure, the diagnostic script generation method further includes: receiving a diagnostic result sent by the target vehicle, and sending the diagnostic result and the diagnostic script to the client; receiving feedback information related to the diagnostic result and the diagnostic script sent by the client; and in response to the feedback information indicating that the vehicle diagnosis is qualified, adding the diagnostic script to the diagnostic script library.

[0058] In the embodiment, the diagnostic result is obtained by the target vehicle after obtaining the diagnostic script, by acquiring vehicle signals related to the diagnostic script on the vehicle, and by giving a diagnostic conclusion based on a comprehensive analysis of the vehicle according to the values of the vehicle signals, for example, the diagnostic result is that the engine is not easy to start or the steering wheel shakes at high speed. Specifically, as shown in Figure 2 The diagnostic result J is information sent by the vehicle diagnosis program of the target vehicle to the cloud after obtaining the diagnostic script Z issued by the cloud.

[0059] In the embodiment, specifically, after obtaining the diagnostic script issued by the execution subject, the target vehicle acquires vehicle signals related to the diagnostic script on the vehicle, and gives a diagnostic result based on a comprehensive analysis of the vehicle according to the values of the vehicle signals.

[0060] In the embodiment, the feedback information includes: vehicle diagnosis qualified or vehicle diagnosis unqualified. Sending the diagnostic result and the diagnostic script to the client can enable the operator to analyze the diagnostic script and the diagnostic result, determine whether the diagnostic script meets the diagnostic requirements, and whether the diagnostic script and the diagnostic result correspond to each other, generate feedback information of vehicle diagnosis qualified when the diagnostic script meets the diagnostic requirements and the diagnostic script and the diagnostic result correspond to each other, and generate feedback information of vehicle diagnosis unqualified when the diagnostic script does not meet the diagnostic requirements or the diagnostic script and the diagnostic result do not correspond to each other.

[0061] The diagnostic script generation method provided in the embodiment receives a diagnostic result sent by a target vehicle, and sends the diagnostic result and the diagnostic script to a client; receives feedback information related to the diagnostic result and the diagnostic script sent by the client; and in response to the feedback information indicating that the vehicle diagnosis is qualified, adds the diagnostic script to a diagnostic script library, enriches the diagnostic script library, and improves the reliability of the diagnostic script.

[0062] In some embodiments of the present disclosure, the diagnostic script generation method further includes: in response to the feedback information indicating that the vehicle diagnosis is unqualified, obtaining an optimized script based on the feedback information; and sending the optimized script to the target vehicle as the diagnostic script. Specifically, Figure 4 Flow 400 illustrating still another embodiment of the diagnostic script generation method according to the present disclosure is shown, which includes the following steps:

[0063] Step 401, vehicle fault information is obtained, and then step 402 is performed.

[0064] Step 402, a fault correlation factor is generated based on the vehicle fault information and a factor large model, and then step 403 is performed.

[0065] Step 403, a diagnostic script is generated based on the fault correlation factor and a diagnostic large model, and then step 404 is performed.

[0066] Step 404, the diagnostic script is sent to the target vehicle, and then step 405 is performed.

[0067] It should be understood that the operations and features in the above steps 401-404 correspond to the operations and features in steps 301-304, respectively, and therefore the description of the operations and features in steps 301-304 also applies to steps 401-404, which will not be repeated here.

[0068] Step 405, the diagnostic result sent by the target vehicle is received, and the diagnostic result and the diagnostic script are sent to a client, and then step 406 is performed.

[0069] In the embodiment, the diagnostic result is a result obtained by the target vehicle performing the diagnostic script on its own vehicle. Specifically, after obtaining the diagnostic script issued by the execution subject, the target vehicle obtains vehicle signals related to the diagnostic script on its own vehicle, and gives a diagnostic result based on the comprehensive analysis of the values of the vehicle signals.

[0070] In this embodiment, the diagnostic result and the diagnostic script are sent to the client, so that the operation personnel can analyze the diagnostic script and the diagnostic result, determine whether the diagnostic script meets the diagnostic requirement, and whether the diagnostic script corresponds to the diagnostic result. When the diagnostic script meets the diagnostic requirement and the diagnostic script corresponds to the diagnostic result, feedback information indicating that the generated diagnostic information is qualified is given. When the diagnostic script does not meet the diagnostic requirement or the diagnostic script does not correspond to the diagnostic result, feedback information indicating that the generated diagnostic information is unqualified is given.

[0071] In step 406, the feedback information sent by the client is received, and then step 407 is performed.

[0072] In this embodiment, the feedback information is information related to the diagnostic result and the diagnostic script, and the feedback information includes: vehicle diagnosis qualified or vehicle diagnosis unqualified. When the diagnostic result corresponds to the diagnostic script and the diagnostic script meets the diagnostic requirement, it is determined that the diagnosis of the target vehicle is qualified, and feedback information indicating that the diagnosis is qualified is given. When the diagnostic result does not correspond to the diagnostic script or the diagnostic script does not meet the diagnostic requirement, it is determined that the diagnosis of the target vehicle is unqualified.

[0073] In step 407, it is determined whether the feedback information indicates that the diagnostic result is qualified. If the feedback information indicates that the diagnostic result is qualified, step 408 is performed. Otherwise, step 410 is performed.

[0074] In step 408, the diagnostic script is added to the diagnostic script library, and then step 409 is performed.

[0075] In step 409, the process ends.

[0076] In step 410, based on the feedback information, an optimized script is obtained, and the optimized script is used as the diagnostic script, and step 404 is performed.

[0077] In this embodiment, the optimized script is a script obtained by optimizing (for example, modifying instructions, replacing instructions) the diagnostic script.

[0078] Alternatively, the optimized script can also be a script obtained from the client. The operation personnel optimizes the diagnostic script on the client to obtain the optimized script, encapsulates the optimized script and other information into the feedback information, and feeds back the feedback information to the execution subject, so that the execution subject can obtain the optimized script of the diagnostic script from a specific area of the feedback information when the feedback information indicates that the vehicle diagnosis is unqualified.

[0079] The diagnostic script generation method provided in this embodiment responds to the feedback information indicating that the vehicle diagnosis is unqualified, obtains an optimized script based on the feedback information, and sends the optimized script as a diagnostic script to the target vehicle, thereby improving the reliability of the diagnostic script generation.

[0080] In some optional implementations of the present disclosure, the above factor large model comprises: a first large model, and the generating of the fault-related factor based on the vehicle fault information and the factor large model comprises: matching the vehicle fault information with fault repair information; and in response to a successful matching of the vehicle fault information with the fault repair information, inputting the first large model based on the vehicle fault information, the fault repair information, and a selected prompt word to obtain the fault-related factor output by the first large model.

[0081] In the optional implementation, the first large model is a pre-trained multi-modal large model, and the first large model is a model for obtaining the fault-related factor from relevant data selected from a diagnosis database. The multi-modal information and the prompt word are input into the first large model to obtain the fault-related factor output by the first large model. The diagnosis database is a database related to vehicle diagnosis, and the data in the diagnosis database comprises: fault repair information, which is multi-modal information related to vehicle diagnosis and repair, and the fault repair information comprises: a fault repair manual and a DTC repair case.

[0082] In the optional implementation, when the vehicle fault information is successfully matched with the fault repair information, it is determined that the vehicle fault information has information related to the vehicle fault information, and at this time, the diagnosis script can be obtained by analyzing the vehicle fault information.

[0083] In the optional implementation, the selected prompt word is prompt information for prompting the large model to select relevant information from pre-prepared materials, and when the pre-prepared accurate materials are the vehicle fault information, the selected prompt word is used to prompt the large model to select the fault-related factor from the vehicle fault information.

[0084] The method for generating a fault-related factor provided in the optional implementation is used to input the first large model based on the vehicle fault information, the fault repair information, and a selected prompt word after the vehicle fault information is successfully matched with the fault repair information, to obtain the fault-related factor output by the first large model, thereby providing a reliable implementation means for obtaining the fault-related factor, and improving the reliability and accuracy of the fault-related factor.

[0085] In an optional implementation, when the vehicle fault information and the fault repair information are both text data, the first large model is a large language model.

[0086] In an optional implementation, when the vehicle fault information and the fault repair information are both images, videos, or audios, the first large model is an image, video, or audio large model.

[0087] In some optional implementations of the present disclosure, the above-mentioned factor large model also includes: a second large model, which generates a fault correlation factor based on the vehicle fault information and the factor large model, and further includes: in response to the vehicle fault information and the fault repair information not being successfully matched, inputting the vehicle fault information and the generated prompt word into the second large model to obtain the fault correlation factor output by the second large model.

[0088] In this optional implementation, the second largest model is a pre-trained multimodal large model, and the second largest model is a model that can directly generate fault correlation factors.

[0089] In this optional implementation, the generated prompt word is prompt information that prompts the large model to generate relevant information. When the relevant information is a fault correlation factor, the generated prompt word is used to prompt the large model to generate the fault correlation factor.

[0090] The method for generating a fault correlation factor provided by this optional implementation method inputs the vehicle fault information and the generated prompt word into the second largest model after the vehicle fault information and the fault repair information are not successfully matched, and obtains the fault correlation factor output by the second largest model, thereby providing another reliable implementation means for obtaining the fault correlation factor and improving the reliability and accuracy of obtaining the fault correlation factor.

[0091] In an optional implementation, when the vehicle fault information is text data, the second largest model is a large language model.

[0092] In an optional implementation, when the vehicle fault information is an image, video or audio, the second largest model is an image, video or audio large model.

[0093] In an optional implementation, when the vehicle fault information is multimodal data, the second largest model is a multimodal large model.

[0094] In some optional implementations of the present disclosure, the above-mentioned fault correlation factors include: fault detection process sub-factors. Based on the fault correlation factors and the diagnostic big model, generating a diagnostic script includes: inputting the fault detection process sub-factors and diagnostic prompt words into the diagnostic big model to obtain the diagnostic script output by the diagnostic big model.

[0095] In this optional implementation, the fault detection process sub-factor represents the process steps involved in detecting the fault corresponding to the vehicle fault information. This sub-factor can be used to determine the specific process by which the vehicle detects the fault. This sub-factor can be represented using various data modalities, such as images and text.

[0096] In this optional implementation, the diagnostic big model is a big model that generates a diagnostic script from the fault correlation factors. When the fault correlation factors are multimodal data, the diagnostic big model is a multimodal big model.

[0097] In another optional implementation, the diagnostic large model is a large model for generating a diagnostic script based on the fault correlation factor. When the fault correlation factor is text data, the diagnostic large model is a large language model.

[0098] In another optional implementation, the diagnostic large model is a large model for generating a diagnostic script based on the fault correlation factor. When the fault correlation factor is an image, a video, or audio, the diagnostic large model is an image, a video, or audio large model.

[0099] In this optional implementation, the diagnostic prompt word is a prompt word for prompting the diagnostic large model to generate a diagnostic script.

[0100] The method for generating a diagnostic script provided in this optional implementation inputs the fault detection process sub-factor and the diagnostic prompt word into the diagnostic large model to obtain a diagnostic script output by the diagnostic large model, thereby providing a reliable implementation means for obtaining a diagnostic script and improving the reliability and accuracy of obtaining a diagnostic script.

[0101] In some optional implementations of the present disclosure, the fault correlation factor further includes an instruction key sub-factor. Generating the diagnostic script based on the fault correlation factor and the diagnostic large model includes: obtaining an instruction list based on the instruction key sub-factor and a script instruction library; and inputting the instruction list, the fault detection process sub-factor, and the diagnostic prompt word into the diagnostic large model to obtain the diagnostic script output by the diagnostic large model.

[0102] In this optional implementation, the instruction key sub-factor is a factor related to an instruction for implementing the fault detection process sub-factor. Specifically, the instruction key sub-factor includes: an in-vehicle CAN signal related to a fault corresponding to vehicle fault information; a vehicle state related to the fault; an Ethernet message related to the fault; and an acquisition condition, a time period, and a frequency corresponding to each of the above factors (the in-vehicle CAN signal, the vehicle state, and the Ethernet message).

[0103] In this optional implementation, the script instruction library is a database for storing script instructions, and the script instructions in the script instruction library can be queried through an index. The script instructions are instructions related to scripts of a vehicle.

[0104] In this optional implementation, the instruction list is a list including at least one script instruction. The script instructions in the instruction list are all script instructions related to the instruction key sub-factor.

[0105] The method for obtaining a diagnosis script instruction provided by the optional implementation provides an instruction list based on an instruction key sub-factor and a script instruction library; the instruction list, a fault detection process sub-factor, and a diagnosis prompt word are collectively input into a diagnosis large model to obtain a diagnosis script output by the diagnosis large model. Compared with directly providing a diagnosis script generated by a fault detection process sub-factor, a diagnosis prompt word, and a diagnosis large model, the large model can more clearly constitute a script instruction of the diagnosis script, and generate the diagnosis script based on the script instructions, thereby further improving the accuracy of diagnosis script generation.

[0106] In some optional implementations of the present disclosure, the instruction key sub-factor includes an instruction keyword and an acquisition attribute of the instruction keyword, and obtaining the instruction list based on the instruction key sub-factor and the script instruction library includes: acquiring an instruction set related to the instruction keyword from the script instruction library based on the acquisition attribute; performing vectorization processing on the instruction set to obtain an instruction vector set; matching the instruction vector set with a vector of the instruction keyword; and obtaining the instruction list based on a matching result of the instruction vector set and the vector of the instruction keyword.

[0107] In the optional implementation, the instruction keyword is a word related to an instruction of the fault detection process sub-factor, such as a CAN signal in a vehicle related to a fault of vehicle fault information; a vehicle state related to the fault; and an Ethernet message related to the fault.

[0108] In the optional implementation, the acquisition attribute of the instruction keyword is attribute information of the instruction keyword, for example, an acquisition condition, a time period, and a frequency of acquiring the instruction keyword.

[0109] In the optional implementation, the matching of the instruction vector set with the vector of the instruction keyword is performed by referring to the vector index shown in Figure 2

[0110] The method for obtaining an instruction list provided by the optional implementation acquires an instruction set related to an instruction keyword from a script instruction library based on an acquisition attribute; performs vectorization processing on the instruction set to obtain an instruction vector set; matches the instruction vector set with a vector of the instruction keyword; and obtains the instruction list based on a matching result of the instruction vector set and the vector of the instruction keyword. The acquisition attribute and the instruction keyword are used to limit the instruction list, thereby improving the accuracy of the obtained instruction list.

[0111] In some optional implementations of the present disclosure, the diagnosis large model is a code generation model after fine-tuning, and the fine-tuning includes inputting script syntax information into the code generation model, so that the code generation model can output a code corresponding to the diagnosis script.

[0112] ​In this optional implementation, large model fine-tuning refers to adjusting the parameters of a large language model to adapt to a specific task. This process is accomplished by training the large model on a dataset specific to the task, aiming to optimize the model's performance on the specific task. The importance of fine-tuning lies in enabling the large model to adapt more finely to specific tasks, reducing inference costs, and improving the model's performance in specific domains.

[0113] In this optional implementation, the code generation model is a large language model that can generate script code. The fine-tuning of the above code generation model refers to adjusting the parameters of the large language model while also inputting script syntax information to the code generation model to adapt to the specific diagnostic script generation task. The script syntax information refers to an executable file written in a specific descriptive language according to a certain format. This language is usually used to control the behavior of software applications and can be interpreted or compiled for execution when called.

[0114] Since scripts have corresponding writing rules, in order to adapt to these writing rules, script syntax information is input to the code generation model during the fine-tuning process of the code generation model, which enables the code generation model to output the diagnostic script corresponding to the current script syntax information, improving the accuracy of diagnostic script generation.

[0115] The diagnostic large model provided in this optional implementation inputs syntax information to the code generation model during the fine-tuning of the code generation model, improving the accuracy of the diagnostic large model in generating diagnostic scripts.

[0116] As can be seen from the above embodiments, the diagnostic generation method provided by the present disclosure can be a complete script automatic generation and optimization process, and can complete the automatic construction process method of the script library, ultimately realizing the logical closed loop of script automatic retrieval, generation, optimization, and storage into the library. The complete process schematic diagram is shown in Figure 2

[0117] 1) After the vehicle reports vehicle fault information, the cloud uses vector retrieval technology to filter diagnostic scripts that match the current vehicle fault information based on the existing diagnostic script library (including diagnostic scripts, corresponding DTCs, and fault phenomena). If a match is successful, the diagnostic script is directly returned, allowing the vehicle's vehicle diagnostic program to directly execute the diagnostic script.

[0118] Among them, the vector retrieval technology vectorizes all diagnostic scripts, corresponding DTCs, and fault phenomena in the existing diagnostic script library, and then compares the similarity of the vectors corresponding to the DTCs and fault phenomena in the vehicle fault information uploaded by the vehicle. If the similarity is higher than a certain threshold, the script with the highest similarity is selected for successful matching. Otherwise, it is considered that there is no successfully matched script.

[0119] ​2) When the vector retrieval technology is not matched successfully, based on the factor big model, analyze the "fault correlation factor" corresponding to the vehicle fault information. The "fault correlation factor" includes two parts: the first part and the second part. Each part includes but is not limited to the following description:

[0120] First part: CAN signal involved in the fault corresponding to the vehicle fault information; vehicle state involved in the fault; CAN and Ethernet message involved in the fault; acquisition conditions, time period and frequency corresponding to each factor.

[0121] Second part: fault detection process.

[0122] 3) In the process of analyzing "fault correlation factors", the fault repair manual and DTC repair case provided by the host factory are used as input of the factor big model, and appropriate prompt words are constructed to make the factor big model output each "fault correlation factor" required. This process can be implemented using a fine-tuned model, or by using a vector retrieval technology plus a big model summary.

[0123] In addition to using fault repair manual and DTC repair case for analysis, for faults that cannot be matched, a suitable prompt word can be constructed to query a general big model, or other data sources can be used to obtain the required information.

[0124] 4) Based on the vector retrieval technology, the script instructions for acquiring each factor in the first part of the current "fault correlation factor" on the vehicle end are obtained from the script instruction library.

[0125] The cloud end contains a script instruction library, and the vehicle diagnostic control program includes a vehicle diagnostic program that executes all instructions. The script instruction library includes the definition and function description of all diagnostic scripts, and through the matching of fault correlation factors and diagnostic script descriptions, it is determined which script instruction should be used for each factor.

[0126] 5) Based on the "fault detection process" sub-factor in the fault diagnosis factor and the instruction list, a diagnostic script is constructed using a diagnostic big model. The diagnostic big model used in this step is a fine-tuned code generation model, and the fine-tuning process requires the definition of all script syntax.

[0127] 6) The diagnostic script generated in step 5) is issued to the target vehicle, and the vehicle diagnostic program of the target vehicle executes the diagnostic script.

[0128] 7) The target vehicle returns the diagnostic results of the execution to the cloud end, and enters a process for manual judgment.

[0129] 8) The generated diagnostic script and the executed diagnostic results are analyzed by a human, and if they do not meet the diagnostic requirements, the script is optimized and then issued to the vehicle end for execution again until the expected results are obtained.

[0130] 9) The artificially confirmed script is added to the script library, and the next time a similar problem is encountered, the script can be directly returned from the script library for automatic execution.

[0131] The diagnostic script generation method provided by the present disclosure can automatically retrieve the required script from the script library and issue it to the vehicle end for execution, realizing automatic diagnosis function; for diagnostic scripts that are not in the diagnostic script library, diagnostic scripts can be automatically generated, without experienced R&D personnel spending a lot of time consulting a large number of fault maintenance manuals and fault maintenance cases, improving the diagnostic script generation efficiency; the automatically generated diagnostic script will be reviewed and optimized by artificial and then stored in the library, forming a logical closed loop, and improving the optimization effect of the diagnostic script.

[0132] Further reference Figure 5 , as an implementation of the method shown in the above figures, the present disclosure provides an embodiment of a diagnostic script generation device, which corresponds to the method embodiment shown in Figure 1 , and the device can be implemented in various electronic devices.

[0133] As shown in Figure 5 , the diagnostic script generation device 500 provided by the present embodiment includes an acquisition unit 501, a factor generation unit 502, and a script generation unit 503. The acquisition unit 501 can be configured to acquire vehicle fault information. The factor generation unit 502 can be configured to generate a fault-related factor based on the vehicle fault information and a factor large model. The script generation unit 503 can be configured to generate a diagnostic script based on the fault-related factor and a diagnostic large model.

[0134] In the present embodiment, the diagnostic script generation device 500 includes the acquisition unit 501, the factor generation unit 502, and the script generation unit 503, and the specific processing thereof and the technical effects brought thereby can be respectively referred to the related descriptions of the steps 101, 102, and 103 in the corresponding embodiments, which will not be repeated here. Figure 1

[0135] In some optional implementation manners of the present embodiment, the factor generation unit 502 is configured to detect whether there is a diagnostic script matching the vehicle fault information based on the diagnostic script library; and in response to no diagnostic script matching the vehicle fault information being detected, generate a fault-related factor based on the vehicle fault information.

[0136] ​In some optional implementation manners of the present embodiment, the factor generation unit 502 is further configured to: perform vectorization processing on the diagnosis scripts in the diagnosis script library to obtain a first vector library; perform vectorization processing on the vehicle fault information to obtain a vehicle fault vector; match the vehicle fault vector with the vectors in the first vector library; and in response to the vehicle fault vector matching successfully with the vectors in the first vector library, determine that the diagnosis script has a match with the vehicle fault information.

[0137] In some optional implementation manners of the present embodiment, the vehicle fault information is fault information sent by the target vehicle, and the apparatus 500 further includes a sending unit (not shown in the figure), configured to send the diagnosis script to the target vehicle.

[0138] In some optional implementation manners of the present embodiment, the apparatus 500 further includes an adding unit (not shown in the figure), configured to: receive diagnosis results sent by the target vehicle, and send the diagnosis results and the diagnosis script to the client; receive feedback information sent by the client, the feedback information being related to the diagnosis results and the diagnosis script; and in response to the feedback information indicating that the vehicle diagnosis is qualified, add the diagnosis script to the diagnosis script library.

[0139] In some optional implementation manners of the present embodiment, the apparatus 500 further includes an optimization unit (not shown in the figure), configured to: in response to the feedback information indicating that the vehicle diagnosis is unqualified, obtain an optimized script based on the feedback information; and send the optimized script to the target vehicle as the diagnosis script.

[0140] In some optional implementation manners of the present embodiment, the factor large model includes a first large model, and the factor generation unit 502 is configured to: match the vehicle fault information with the fault repair information; and in response to the vehicle fault information matching successfully with the fault repair information, input the vehicle fault information, the fault repair information and the selected prompt word into the first large model to obtain a fault correlation factor output by the first large model.

[0141] In some optional implementation manners of the present embodiment, the factor large model further includes a second large model, and the factor generation unit 502 is configured to: in response to the vehicle fault information and the fault repair information not matching successfully, input the vehicle fault information and the generated prompt word into the second large model to obtain a fault correlation factor output by the second large model.

[0142] In some optional implementation manners of the present embodiment, the fault correlation factor includes a fault detection process sub-factor, and the script generation unit 503 is configured to: input the fault detection process sub-factor and the diagnosis prompt word into a diagnosis large model to obtain a diagnosis script output by the diagnosis large model.

[0143] In some optional implementations of the present disclosure, the fault correlation factor further includes an instruction key sub-factor, and the script generation unit 503 is further configured to: obtain an instruction list based on the instruction key sub-factor and the script instruction library; and input the instruction list, the fault detection process sub-factor, and the diagnosis prompt word into the diagnosis large model to obtain a diagnosis script output by the diagnosis large model.

[0144] In some optional implementations of the present disclosure, the instruction key sub-factor includes an instruction key word and an acquisition attribute of the instruction key word, and the script generation unit 503 is further configured to: obtain an instruction set related to the instruction key word from the script instruction library based on the acquisition attribute; perform vectorization processing on the instruction set to obtain an instruction vector set; match the instruction vector set with a vector of the instruction key word; and obtain an instruction list based on a matching result of the instruction vector set and the vector of the instruction key word.

[0145] In some optional implementations of the present disclosure, the diagnosis large model is a code generation model that is fine-tuned, and the fine-tuning includes inputting script syntax information into the code generation model, so that the code generation model can output a code corresponding to the diagnosis script.

[0146] The diagnosis script generation apparatus provided by the embodiments of the present disclosure first acquires vehicle fault information by the acquisition unit 501, then generates a fault correlation factor based on the vehicle fault information and a factor large model by the factor generation unit 502, and finally generates a diagnosis script based on the fault correlation factor and a diagnosis large model by the script generation unit 503. In this way, the fault correlation factor is automatically generated based on the vehicle fault information, and the diagnosis script is generated based on the fault correlation factor, thereby improving the diagnosis script generation efficiency and improving the user experience without the need for an operator to search for information.

[0147] In the technical solutions of the present disclosure, the collection, storage, use, processing, transmission, provision, and disclosure of user personal information comply with relevant laws and regulations and do not violate public order and good customs.

[0148] According to the embodiments of the present disclosure, the present disclosure further provides an electronic device, a readable storage medium, and a computer program product.

[0149] Figure 6A schematic block diagram of an example electronic device 40600 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0150] like Figure 6 As shown, the device 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. Various programs and data required for the operation of the device 600 can also be stored in the RAM 603. The computing unit 601, the ROM 602, and the RAM 40603 are connected to each other via a bus 40604. An input / output (I / O) interface 40605 is also connected to the bus 604.

[0151] Multiple components in device 600 are connected to I / O interface 40605, including an input unit 606, such as a keyboard, mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a magnetic disk, optical disk, etc.; and a communication unit 609, such as a network card, modem, wireless communication transceiver, etc. The communication unit 609 allows device 600 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0152] The computing unit 601 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs various methods and processes described above, such as the diagnostic script generation method. For example, in some embodiments, the diagnostic script generation method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded onto the RAM 603 and executed by the computing unit 601, one or more steps of the diagnostic script generation method described above can be performed. Alternatively, in other embodiments, the computing unit 601 can be configured to perform the diagnostic script generation method by any other suitable means, such as by means of firmware.

[0153] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (PLD), a computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0154] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable diagnostic script generation apparatus to produce a machine, such that the program code, when executed by the processor or controller, implements the functions / operations specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0155] In the context of this disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0156] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0157] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a client computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0158] The computer system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0159] It should be understood that the various forms of flow shown above can be used to reorder, add, or delete steps. For example, the steps described in the present disclosure can be performed in parallel, in series, or in a different order, as long as the desired results of the technology disclosed in the present disclosure can be achieved, which is not limited herein.

[0160] The specific implementation described above does not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A diagnostic script generation method, comprising: obtaining vehicle fault information; generating fault correlation factors based on the vehicle fault information and a factor large model; wherein the factor large model comprises a first large model; including: matching the vehicle fault information with fault maintenance information; in response to successful matching of the vehicle fault information with the fault maintenance information, inputting the vehicle fault information, the fault maintenance information and selected prompt words into the first large model to obtain fault correlation factors output by the first large model; generating a diagnostic script based on the fault correlation factors and a diagnostic large model; wherein the fault correlation factors include fault detection process sub-factors; including: inputting the fault detection process sub-factors and diagnostic prompt words into a diagnostic large model to obtain a diagnostic script output by the diagnostic large model.

2. The method of claim 1, wherein generating fault correlation factors based on the vehicle fault information and a factor large model comprises: detecting whether there is a diagnostic script matching the vehicle fault information based on a diagnostic script library; in response to no diagnostic script matching the vehicle fault information being detected, generating fault correlation factors based on the vehicle fault information and a factor large model.

3. The method of claim 2, wherein, The detection of whether there is a diagnostic script matching the vehicle fault information based on a diagnostic script library comprises: vectorizing the diagnostic scripts in the diagnostic script library to obtain a first vector library; vectorizing the vehicle fault information to obtain a vehicle fault vector; matching the vehicle fault vector with the vectors in the first vector library; in response to successful matching of the vehicle fault vector with the vectors in the first vector library, determining that there is a diagnostic script matching the vehicle fault information.

4. The method of claim 1, wherein the vehicle fault information is fault information sent by a target vehicle, and the method further comprises: sending the diagnostic script to the target vehicle.

5. The method of claim 4, further comprising: receiving a diagnostic result sent by the target vehicle and sending the diagnostic result and the diagnostic script to a client; receiving feedback information sent by the client, the feedback information being related to the diagnostic result and the diagnostic script respectively; in response to the feedback information indicating that the vehicle diagnosis is qualified, adding the diagnostic script to a diagnostic script library.

6. The method of claim 5, further comprising: in response to the feedback information indicating that the vehicle diagnosis is unqualified, obtaining an optimized script based on the feedback information; sending the optimized script as a diagnostic script to the target vehicle.

7. The method of claim 1, wherein, The factor large model further comprises a second large model, and generating fault correlation factors based on the vehicle fault information and a factor large model further comprises: in response to the vehicle fault information and the fault maintenance information not matching successfully, inputting the vehicle fault information and a generated prompt word into the second large model to obtain fault correlation factors output by the second large model.

8. The method of claim 1, wherein, The fault correlation factors further comprise instruction key sub-factors, and generating a diagnostic script based on the fault correlation factors and a diagnostic large model further comprises: obtain an instruction list based on the instruction key sub-factor and a script instruction library; input the instruction list, the fault detection process sub-factor and a diagnosis prompt word into the diagnosis large model to obtain a diagnosis script output by the diagnosis large model.

9. The method of claim 8, wherein, The instruction key sub-factor includes an instruction keyword and an acquisition attribute of the instruction keyword, and obtaining the instruction list based on the instruction key sub-factor and the script instruction library includes: acquiring an instruction set related to the instruction keyword from the script instruction library based on the acquisition attribute; performing vectorization processing on the instruction set to obtain an instruction vector set; matching the instruction vector set with a vector of the instruction keyword; obtaining the instruction list based on a matching result of the instruction vector set and the vector of the instruction keyword.

10. The method of any one of claims 1, 8-9, wherein, The diagnosis large model is a code generation model after fine-tuning; the fine-tuning includes inputting script syntax information into the code generation model, so that the code generation model can output a code corresponding to the diagnosis script.

11. A diagnosis script generation apparatus, the apparatus comprising: an acquisition unit configured to acquire vehicle fault information; a factor generation unit configured to generate a fault association factor based on the vehicle fault information and a factor large model; The factor large model includes a first large model, and the factor generation unit is configured to: match the vehicle fault information with fault repair information; and in response to successful matching of the vehicle fault information and the fault repair information, input the vehicle fault information, the fault repair information and a selected prompt word into the first large model to obtain a fault association factor output by the first large model. a script generation unit configured to generate a diagnosis script based on the fault association factor and a diagnosis large model; the fault association factor includes a fault detection process sub-factor, and the script generation unit is configured to input the fault detection process sub-factor and a diagnosis prompt word into a diagnosis large model to obtain a diagnosis script output by the diagnosis large model.

12. The apparatus of claim 11, wherein the factor generation unit is configured to: detect whether there is a diagnosis script matching the vehicle fault information based on a diagnosis script library; and in response to no diagnosis script matching the vehicle fault information being detected, generate a fault association factor based on the vehicle fault information.

13. The apparatus of claim 12, wherein, The factor generation unit is further configured to: perform vectorization processing on diagnosis scripts in a diagnosis script library to obtain a first vector library; perform vectorization processing on the vehicle fault information to obtain a vehicle fault vector; match the vehicle fault vector with vectors in the first vector library; and in response to successful matching of the vehicle fault vector with a vector in the first vector library, determine that there is a diagnosis script matching the vehicle fault information.

14. The apparatus of claim 11, the vehicle failure information being failure information sent by a target vehicle, the apparatus further comprising: a sending unit configured to send the diagnosis script to the target vehicle.

15. The apparatus of claim 14, wherein, The device further comprises an adding unit configured to: receive the diagnostic result sent by the target vehicle, and send the diagnostic result and the diagnostic script to the client; receive feedback information sent by the client, the feedback information being related to the diagnostic result and the diagnostic script; and in response to the feedback information indicating that the vehicle diagnosis is qualified, add the diagnostic script to a diagnostic script library.

16. The apparatus of claim 15, wherein, The device further comprises an optimizing unit configured to: in response to the feedback information indicating that the vehicle diagnosis is unqualified, obtain an optimized script based on the feedback information; and send the optimized script to the target vehicle as a diagnostic script.

17. The apparatus of claim 11, wherein, The factor large model further comprises a second large model, and the factor generation unit is configured to: in response to the vehicle fault information and the fault repair information not being successfully matched, input the vehicle fault information and a generated prompt word into the second large model to obtain a fault correlation factor output by the second large model.

18. The apparatus of claim 11, wherein, The fault correlation factor further comprises an instruction key sub-factor, and the script generation unit is further configured to: obtain an instruction list based on the instruction key sub-factor and a script instruction library; and input the instruction list, the fault detection process sub-factor, and a diagnostic prompt word into the diagnostic large model to obtain a diagnostic script output by the diagnostic large model.

19. The apparatus of claim 18, wherein, The instruction key sub-factor comprises an instruction key word and an acquisition attribute of the instruction key word, and the script generation unit is further configured to: based on the acquisition attribute, obtain an instruction set related to the instruction key word from a script instruction library; perform vectorization processing on the instruction set to obtain an instruction vector set; match the instruction vector set with a vector of the instruction key word; and based on a matching result of the instruction vector set and the vector of the instruction key word, obtain an instruction list.

20. The apparatus of one of claims 11, 18-19, wherein, The diagnostic large model is a code generation model that has been fine-tuned; the fine-tuning comprises inputting script syntax information into the code generation model, so that the code generation model can output a code corresponding to a diagnostic script.

21. An electronic device, comprising: comprise: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-10.

22. A non-transitory computer-readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to enable the computer to perform the method of any one of claims 1-10.

23. A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1-10.

Images