Vehicle fault analysis method and device based on extreme scene and storage medium

Abstract

The invention discloses a vehicle fault analysis method and device based on an extreme scene and a storage medium, and the method comprises the steps: receiving fault data information containing fault time, and obtaining fault data corresponding to the fault time; preprocessing the fault data by adopting an intelligent agent algorithm to obtain fault preprocessing data; and inputting the fault preprocessing data into a preset fault analysis model, and outputting a fault analysis result which comprises a fault type, a fault equipment name and a fault triggering reason. By implementing the method, the fault analysis result of the vehicle in the extreme scene can be accurately obtained, and the fault analysis result comprises the fault type, the fault equipment name and the fault triggering reason.

Description

Technical Field

[0001] This invention relates to the field of automotive technology, and in particular to a vehicle fault analysis method, device, and storage medium based on extreme scenarios. Background Technology

[0002] As private cars become increasingly common, the number of vehicles on the road is increasing significantly, and the number of vehicle collisions and thermal runaways is also rising accordingly. In extreme scenarios such as collisions and thermal runaways, the time between accidents is often very short, and the vehicle's electrical circuits and electronic components may be damaged. This can lead to the risk of loss or corruption of vehicle-side signal log data, and further difficulties in collecting this data subsequently.

[0003] In extreme scenarios such as vehicle collisions and thermal runaway, traditional data collection and analysis methods mainly rely on vehicle manufacturers' personnel to go to the site, power on the vehicle, and export the data. This significantly reduces data analysis efficiency and creates a bottleneck for locating and troubleshooting vehicle problems. Therefore, there is an urgent need for a vehicle fault analysis method based on extreme scenarios. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a vehicle fault analysis method, computer equipment, computer-readable storage medium and computer program product based on extreme scenarios, so as to accurately determine the fault type of a vehicle in extreme scenarios.

[0005] The technical solution of this invention provides a vehicle fault analysis method based on extreme scenarios, including:

[0006] Upon receiving fault data information containing the fault time, obtain the fault data corresponding to the fault time;

[0007] The fault data is preprocessed using an intelligent agent algorithm to obtain fault preprocessed data;

[0008] The fault preprocessing data is input into a preset fault analysis model, and the fault analysis results are output. The fault analysis results include the fault type, the name of the faulty device, and the cause of the fault.

[0009] In one of the alternative technical solutions, the preprocessing of the fault data using an intelligent agent algorithm to obtain fault preprocessed data includes:

[0010] The fault data is aligned using a time-domain alignment algorithm to obtain window period data.

[0011] The window period data is subjected to noise filtering to obtain smooth data;

[0012] The smoothed data is reassembled according to the maximum timing jitter being less than a preset bit period transmission time to obtain the fault preprocessing data.

[0013] In one of the alternative technical solutions, the fault analysis model is obtained using the following method:

[0014] Build an open-source model for deep exploration;

[0015] The low-rank adaptive algorithm was used to adjust the open-source deep search model to obtain an initial fault analysis model;

[0016] Historical fault data is input into the initial fault analysis model for training to obtain the fault analysis model.

[0017] In one of the alternative technical solutions, the step of adjusting the open-source deep search model using a low-rank adaptation algorithm to obtain an initial fault analysis model includes:

[0018] The open-source depth-finding model is adjusted using the following formula:

[0019]

[0020] Where ΔW is the fine-tuning data matrix; W0 is the original weight matrix of the pre-trained model; B is the randomly initialized trainable dimensionality reduction mapping matrix; A is the all-zero initialized trainable dimensionality increase mapping matrix; and W′ is the target fine-tuning matrix.

[0021] In one of the alternative technical solutions, the step of inputting historical fault data into the initial fault analysis model for training to obtain the fault analysis model further includes:

[0022] Input the prompt template into the fault analysis model to obtain the initial fault repair suggestion model;

[0023] Input the historical fault repair suggestion data into the initial fault repair suggestion model to obtain the target fault repair suggestion model.

[0024] In one of the alternative technical solutions, the step of inputting the fault preprocessing data into a preset fault analysis model and outputting the fault analysis results further includes:

[0025] The fault analysis results are input into the target fault repair suggestion model, and the fault repair suggestions corresponding to the fault analysis results are output.

[0026] In one of the alternative technical solutions, the step of inputting the fault preprocessing data into a preset fault analysis model and outputting the fault analysis results further includes:

[0027] If the fault analysis result is adopted, the fault analysis result will be returned to the fault analysis model.

[0028] The present invention also provides a computer device, including a memory, a processor, and a computer program on the memory.

[0029] The processor executes the computer program to implement the steps of the vehicle fault analysis method based on extreme scenarios as described above.

[0030] The present invention also provides a computer-readable storage medium having a computer program / instruction stored thereon.

[0031] When the computer program / instructions are executed by the processor, they implement the steps of the vehicle fault analysis method based on extreme scenarios as described above.

[0032] The present invention also provides a computer program product, including a computer program / instructions.

[0033] When the computer program / instructions are executed by the processor, they implement the steps of the vehicle fault analysis method based on extreme scenarios as described above.

[0034] The above technical solution has the following beneficial effects: by receiving fault data information containing fault time, obtaining fault data corresponding to the fault time, using intelligent agent algorithm to preprocess the fault data to obtain fault preprocessed data, inputting the fault preprocessed data into a preset fault analysis model, and outputting fault analysis results, thereby achieving accurate fault analysis results of vehicles in extreme scenarios, including fault type, fault device name and fault triggering cause. Attached Figure Description

[0035] The disclosure of this invention will become more readily understood by referring to the accompanying drawings. It should be understood that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. In the drawings:

[0036] Figure 1 A flowchart illustrating a vehicle fault analysis method under extreme scenarios, provided as an embodiment of the present invention;

[0037] Figure 2 A flowchart illustrating a vehicle fault analysis method under extreme scenarios, provided as another embodiment of the present invention;

[0038] Figure 3 This is a schematic diagram of the structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0039] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0040] It is readily understood that, based on the technical solution of this invention, various structural and implementation methods can be interchanged by those skilled in the art without altering the essential spirit of the invention. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this invention and should not be considered as the entirety of the invention or as limitations or restrictions on the technical solution of the invention.

[0041] The directional terms such as up, down, left, right, front, back, front, back, top, and bottom mentioned or possibly used in this specification are defined relative to the structures shown in the accompanying drawings. They are relative concepts and may therefore vary depending on their location and usage. Therefore, these or other directional terms should not be interpreted as restrictive.

[0042] like Figure 1 As shown, an embodiment of the present invention provides a vehicle fault analysis method based on extreme scenarios, comprising:

[0043] Step S101: Receive fault data information containing the fault time, and obtain the fault data corresponding to the fault time;

[0044] Step S102: The fault data is preprocessed using an intelligent agent algorithm to obtain fault preprocessed data;

[0045] Step S103: Input the fault preprocessing data into a preset fault analysis model and output the fault analysis results, which include the fault type, fault device name and fault triggering cause.

[0046] Specifically, when the system goes live in the cloud, the management platform specifies the vehicle VIN code list, vehicle-side components, vehicle-side data types, the time period for uploaded data, and trigger conditions (signals, trigger conditions, trigger thresholds), etc. After configuring the above information, the vehicle manufacturer's engineers click "Policy Deployment." The platform will then deploy the configured policy to the designated vehicle-side telematics box (TBOX) for the specified VIN code via the Message Queuing Telemetry Transport (MQTT) channel.

[0047] The vehicle controller (including the Electronic Stability Control (EBS) and Body Control Module (BCM) transmits signal data to the Intelligent Cruise Control (ICC) via CANFD. The ICC stores the mirrored signal data locally on its hard drive. When the vehicle-side TBOX receives the configuration rules from the cloud via the vehicle-cloud command channel, it transmits the command to the ICC, which then parses the rules using its built-in rule processing engine. When the signal emitted by the vehicle matches the trigger signal emitted by the cloud and reaches the configured trigger threshold, the vehicle captures and uploads the signal data file stored on the ICC's hard drive according to the configured time interval. At this time, the controller executes step S101 to receive fault data information including the fault time and obtain the fault data corresponding to the fault time; the trigger thresholds for collision and thermal runaway are calculated according to the following formula:

[0048] (1) Dynamic threshold decision model

[0049] The model uses the sliding window method to calculate the standard deviation of acceleration σ in real time.

[0050]

[0051] Where N is the total number of signals within the sliding window time; i is the current signal index; x i This represents the actual data value of the signal. This is the average value of all signals within the sliding window.

[0052] When σ exceeds the preset baseline value, the high-precision signal mirroring data upload mode is activated.

[0053] (2) Thermal runaway prediction equation

[0054]

[0055] Where ΔT is the temperature change; Δt is the time change; k1 is the weighting coefficient for voltage change; ΔV is the voltage change; k2 is the weighting coefficient for air pressure change; P gas This represents the change in air pressure.

[0056] Wherein, k1 is preferably 0.15-0.35; k2 is preferably 0.05-0.12.

[0057] When TCRI > 85, an L1 level alarm is triggered.

[0058] Then, step S102 is executed, using an intelligent agent algorithm to preprocess the fault data to obtain fault preprocessed data. The time point of the accident in the signal mirror data is determined based on the fault time in the uploaded message at the time of the vehicle accident. Based on this time point, data preprocessing is performed according to the analysis strategy configured in the background, such as analyzing data within the time range of 3 minutes before and 1 minute after the accident.

[0059] Finally, step S103 is executed to input the fault preprocessing data into the fault analysis model. The fault analysis model compares the fault preprocessing data with the preset standard data range value, and determines the fault type, fault device name, fault triggering cause, etc. based on the abnormal data, and obtains the fault analysis result. Thus, the fault analysis result of the vehicle under extreme scenarios is analyzed, including the fault type, fault device name, and fault triggering cause.

[0060] The vehicle fault analysis method based on extreme scenarios provided in this embodiment receives fault data information containing fault time, obtains fault data corresponding to the fault time, preprocesses the fault data using an intelligent agent algorithm to obtain fault preprocessed data, inputs the fault preprocessed data into a preset fault analysis model, and outputs fault analysis results, thereby accurately obtaining the vehicle fault analysis results under extreme scenarios, including fault type, fault device name, and fault triggering cause.

[0061] In one embodiment, the fault analysis model is obtained using the following method:

[0062] Build an open-source model for deep exploration;

[0063] The low-rank adaptive algorithm was used to adjust the open-source deep search model to obtain an initial fault analysis model;

[0064] Historical fault data is input into the initial fault analysis model for training to obtain the fault analysis model.

[0065] Specifically, the DeepSeek open-source model is fine-tuned using a low-level (Lora) algorithm (with a fine-tuning level defined at 1.5b) to obtain an initial fault analysis model. Historical fault data is then input into this initial model for training, resulting in a new fault analysis model. When the fault analysis model receives fault preprocessing data, it matches and compares the signal data pairs (Data(canId, value, time)) in the preprocessing data with the existing standard data pairs in the fault analysis model, from oldest to newest, for the same canId value. This comparison results list is then used to perform fault analysis. The historical fault data includes historical fault types, historical fault device names, and historical fault triggering reasons.

[0066] In one embodiment, to further improve accuracy, the low-rank adaptation algorithm is used to adjust the depth-finding open-source model to obtain an initial fault analysis model, including:

[0067] The open-source depth-finding model is adjusted using the following formula:

[0068]

[0069] Where ΔW is the fine-tuning data matrix; W0 is the original weight matrix of the pre-trained model; B is the randomly initialized trainable dimensionality reduction mapping matrix; A is the all-zero initialized trainable dimensionality increase mapping matrix; and W′ is the target fine-tuning matrix.

[0070] Specifically, the following steps are used to fine-tune the Deep Search open-source model:

[0071] 1) Determine the parameters required by the PyTorch framework in the Depth Search open-source model, including the rank r and the scaling factor a.

[0072] 2) Configure the rank r and scaling factor a in the PyTorch framework's parameter configuration file. Preferably, r is 8 and a is 16.

[0073] 3) During runtime, the Deep Search open-source model will perform dimensional transformation based on the configured rank r and scaling factor a to obtain the dimensionality reduction mapping matrix B and the dimensionality increase mapping matrix A. After multiplying them, the fine-tuning data matrix ΔW is obtained. The original weight matrix W0 and the fine-tuning data matrix ΔW are added together to obtain the target fine-tuning matrix W′ after fine-tuning.

[0074] In one embodiment, the step of inputting historical fault data into the initial fault analysis model for training to obtain the fault analysis model further includes:

[0075] Input the prompt template into the fault analysis model to obtain the initial fault repair suggestion model;

[0076] Input the historical fault repair suggestion data into the initial fault repair suggestion model to obtain the target fault repair suggestion model.

[0077] Specifically, a prompt template is input into the fault analysis model, and the model is trained by feeding it historical fault repair suggestion data. This enables the fault analysis model to make solution suggestions based on the analysis results, thus obtaining the target fault repair suggestion model.

[0078] In one embodiment, the step of inputting the fault preprocessing data into a preset fault analysis model and outputting the fault analysis results further includes:

[0079] The fault analysis results are input into the target fault repair suggestion model, and the fault repair suggestions corresponding to the fault analysis results are output.

[0080] Specifically, the fault analysis results are matched with keywords in the Prompt template. The matched templates are then used to populate the data, and fault repair suggestions corresponding to the fault analysis results are output for easy maintenance.

[0081] The target fault repair suggestion model matches the analyzed accident causes and classification information to output a template. By replacing the placeholders %s in the output template, the analyzed fault causes, conclusions, and repair suggestions are filled into the template to generate specific output information. For example, the output template for a collision scenario is as follows:

[0082] Scene: Collision Scene

[0083] Fault type: %s

[0084] Cause of failure: The fault occurred in controller %s. %s caused %s to malfunction, triggering a collision.

[0085] Analysis process: Analysis of the %s signal revealed that its value significantly deviated from the normal range (%s). This deviation indicates an abnormality in %s, which in turn leads to an abnormal %s signal, resulting in insufficient braking force and a collision.

[0086] Analysis conclusion: Based on the above analysis, the collision accident was caused by %s.

[0087] Repair suggestion: It is recommended to pay attention to the %s signal module of the controller %s.

[0088] like Figure 2 As shown, another embodiment of the present invention provides a vehicle fault analysis method based on extreme scenarios, comprising:

[0089] Step S201: Receive fault data information containing the fault time, and obtain the fault data corresponding to the fault time;

[0090] Step S202: Use a time-domain alignment algorithm to align the fault data to obtain window period data;

[0091] Specifically, the data for the time period requiring accident analysis is aligned using a time-domain alignment algorithm across multiple signal streams. Model input s i Given the true value of the function, the reference value of the ref function, and the time window length, find the optimal t value with the smallest error within the total sequence from i=1 to i=N to obtain the optimal window period data. This can be achieved using the following formula:

[0092]

[0093] Step S203: Filter the window period data for noise to obtain smoothed data;

[0094] Specifically, noise is filtered from the data within the optimal window period. By inputting parameters such as the window radius M, the offset k, and the weight of the h(k) window function, the smoothed signal data is obtained through calculation. This can be achieved using the following formula:

[0095]

[0096] Step S204: Reassemble the smoothed data according to the transmission time when the maximum timing jitter is less than the preset bit period to obtain fault preprocessing data;

[0097] Specifically, during the acquisition, transmission, and disk writing of signal data, anomalies such as signal timing misalignment and fragmented transmission of long data frames may occur. Therefore, the smoothed data after noise filtering is reassembled according to the requirement that the maximum timing jitter is less than 3 bit periods. This resolves the signal misalignment issue and yields fault preprocessed data, which removes abnormal data such as sampling anomalies and signal misalignment. This can be achieved using the following formula:

[0098]

[0099] Step S205: Input the fault preprocessing data into the preset fault analysis model and output the fault analysis results;

[0100] Step S206: If the fault analysis result is adopted, the fault analysis result is returned to the fault analysis model.

[0101] Specifically, by constructing a channel for feeding the analysis results and the fault analysis model into training, the fault analysis model can be self-trained and evolved, achieving positive feedback between the analysis results and the model training, as detailed below:

[0102] The analysis results are input into the fault analysis model, which is mainly achieved by calling the model data feeding training interface. The fault analysis model receives the analysis results (signal data pairs) and combines them with existing data to train itself, generating a new result set. The new result set will be used for subsequent fault data analysis and processing, thereby further improving accuracy.

[0103] The vehicle fault analysis method based on extreme scenarios provided in this embodiment preprocesses fault data by using a time-domain alignment algorithm, noise filtering, and message reassembly to obtain fault preprocessed data. Abnormal data such as sampling anomalies and signal misalignments are removed. Furthermore, the adopted fault analysis results are called back to the fault analysis model to generate a new result set. The new result set will be used for subsequent fault data analysis and processing, thereby further improving accuracy.

[0104] like Figure 3 As shown, a hardware structure diagram of an electronic device for vehicle fault analysis based on extreme scenarios provided by an embodiment of the present invention includes:

[0105] At least one processor 301; and,

[0106] Memory 302 is communicatively connected to at least one processor 301; wherein,

[0107] The memory 302 stores instructions that can be executed by at least one processor 301, which enables the at least one processor 301 to perform the vehicle fault analysis method based on extreme scenarios as described in any of the above method embodiments.

[0108] Figure 3 Take processor 301 as an example.

[0109] The electronic device is preferably an electronic control unit (ECU).

[0110] The electronic device may also include an input device 303 and an output device 304.

[0111] The processor 301, memory 302, input device 303 and output device 304 can be connected by a bus or other means. The figure shows an example of connection by bus.

[0112] Memory 302, as a non-volatile computer-readable storage medium, can be used to obtain non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions / modules corresponding to the vehicle fault analysis method based on extreme scenarios in the embodiments of this application, for example, Figures 1-2 The method flow is shown. The processor 301 executes various functional applications and data processing by running non-volatile software programs, instructions, and modules acquired in the memory 302, thereby realizing the vehicle fault analysis method based on extreme scenarios in the above embodiments.

[0113] The memory 302 may include an acquisition program area and an acquisition data area, wherein the acquisition program area may acquire an operating system and an application program required for at least one function; the acquisition data area may acquire data created based on the use of the vehicle fault analysis method under extreme scenarios, etc. Furthermore, the memory 302 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 302 may optionally include memory remotely located relative to the processor 301, and these remote memories may be connected via a network to the apparatus performing the vehicle fault analysis method under extreme scenarios. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0114] The input device 303 can receive user clicks and generate signal inputs related to user settings and function controls for vehicle fault analysis methods based on extreme scenarios. The output device 304 may include a display device such as a screen.

[0115] When the one or more modules are accessed in the memory 302 and are run by the one or more processors 301, the vehicle fault analysis method based on extreme scenarios in any of the above method embodiments is executed.

[0116] The above-described product can perform the methods provided in the embodiments of this application, and has the corresponding functional modules and beneficial effects for performing the methods. Technical details not described in detail in this embodiment can be found in the methods provided in the embodiments of this application.

[0117] When the computer device disclosed in this invention is running, it can execute all the steps of the above-mentioned vehicle fault analysis method based on extreme scenarios. By receiving fault data information containing fault time, it obtains fault data corresponding to the fault time, uses an intelligent agent algorithm to preprocess the fault data to obtain fault preprocessed data, inputs the fault preprocessed data into a preset fault analysis model, and outputs fault analysis results, thereby accurately obtaining the fault analysis results of the vehicle in extreme scenarios, including fault type, fault device name, and fault triggering cause.

[0118] An embodiment of the present invention provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor 301, implements the steps of any of the aforementioned vehicle fault analysis methods based on extreme scenarios.

[0119] In the context of this disclosure, a storage medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. The storage medium can be a machine-readable signal medium or a machine-readable storage medium. Optionally, the storage medium can be a non-transitory computer-readable storage medium, such as a ROM, random access memory (RAM), compact disc ROM (CD-ROM), magnetic tape, floppy disk, and optical data storage device.

[0120] One embodiment of the present invention provides a computer program product, including a computer program / instructions, which, when executed by a processor, implements the steps of any of the aforementioned vehicle fault analysis methods based on extreme scenarios.

[0121] By running the aforementioned computer program product, all steps of the vehicle fault analysis method based on extreme scenarios described above can be executed. By receiving fault data information containing fault time, fault data corresponding to the fault time is obtained. The fault data is preprocessed using an intelligent agent algorithm to obtain fault preprocessed data. The fault preprocessed data is then input into a preset fault analysis model to output fault analysis results, thereby accurately obtaining the vehicle fault analysis results under extreme scenarios, including fault type, fault device name, and fault triggering cause.

[0122] The above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, and are not intended to limit them. Although the embodiments of the present invention have been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A vehicle fault analysis method based on extreme scenarios, characterized in that, include: Upon receiving fault data information containing the fault time, obtain the fault data corresponding to the fault time; The fault data is preprocessed using an intelligent agent algorithm to obtain fault preprocessed data; The fault preprocessing data is input into a preset fault analysis model, and the fault analysis results are output. The fault analysis results include the fault type, the name of the faulty device, and the cause of the fault.

2. The vehicle fault analysis method based on extreme scenarios as described in claim 1, characterized in that, The fault data is preprocessed using an intelligent agent algorithm to obtain fault preprocessed data, including: The fault data is aligned using a time-domain alignment algorithm to obtain window period data. The window period data is subjected to noise filtering to obtain smooth data; The smoothed data is reassembled according to the maximum timing jitter being less than a preset bit period transmission time to obtain the fault preprocessing data.

3. The vehicle fault analysis method based on extreme scenarios as described in claim 1, characterized in that, The fault analysis model was obtained using the following method: Build an open-source model for deep exploration; The low-rank adaptive algorithm was used to adjust the open-source deep search model to obtain an initial fault analysis model; Historical fault data is input into the initial fault analysis model for training to obtain the fault analysis model.

4. The vehicle fault analysis method based on extreme scenarios as described in claim 3, characterized in that, The low-rank adaptive algorithm is used to adjust the open-source deep search model to obtain an initial fault analysis model, including: The open-source depth-finding model is adjusted using the following formula: Where ΔW is the fine-tuned data matrix; W0 is the original weight matrix of the pre-trained model; B is the randomly initialized trainable dimensionality reduction mapping matrix; A is the all-zero initialized trainable dimensionality increase mapping matrix; W ′ Fine-tune the matrix for the target.

5. The vehicle fault analysis method based on extreme scenarios as described in claim 3, characterized in that, The step of inputting historical fault data into the initial fault analysis model for training to obtain the fault analysis model further includes: Input the prompt word template into the fault analysis model to obtain the initial fault repair suggestion model; Input the historical fault repair suggestion data into the initial fault repair suggestion model to obtain the target fault repair suggestion model.

6. The vehicle fault analysis method based on extreme scenarios as described in claim 5, characterized in that, The step of inputting the fault preprocessing data into a preset fault analysis model and outputting fault analysis results further includes: The fault analysis results are input into the target fault repair suggestion model, and the fault repair suggestions corresponding to the fault analysis results are output.

7. The vehicle fault analysis method based on extreme scenarios as described in any one of claims 1-6, characterized in that, The step of inputting the fault preprocessing data into a preset fault analysis model and outputting fault analysis results further includes: If the fault analysis result is adopted, the fault analysis result will be returned to the fault analysis model.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the vehicle fault analysis method based on extreme scenarios as described in any one of claims 1-7.

9. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instruction is executed by the processor, it implements the steps of the vehicle fault analysis method based on extreme scenarios as described in any one of claims 1 to 7.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the vehicle fault analysis method based on extreme scenarios as described in any one of claims 1-7.

Images