Vehicle fault processing method and device, electronic equipment and storage medium

Abstract

The invention provides a vehicle fault processing method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining sensor data of a plurality of sensors of the same type in a vehicle, carrying out the preprocessing of the sensor data, carrying out the fusion of the preprocessed sensor data, and obtaining effective sensor data, performing fault diagnosis on the effective sensor data by adopting a pre-trained fault diagnosis model to obtain a fault event and a fault danger degree, performing fault recovery evaluation on the fault event by adopting a pre-trained fault evaluation model to generate a recovery evaluation result of the fault event, and performing fault diagnosis on the effective sensor data according to the fault event, the fault danger degree and the recovery evaluation result. And generating an actual fault level of the fault event and a corresponding processing strategy, and controlling the vehicle to execute the processing strategy corresponding to the actual fault level. According to the method, the fault event is identified based on the accurate sensor data, the fault severity is comprehensively and objectively evaluated by integrating the multi-dimensional information source, the fault diagnosis accuracy is improved, and the fault processing strategy is automatically executed.

Description

Technical Field

[0001] The present invention relates to the field of vehicle technology, and in particular to a vehicle fault handling method, device, electronic equipment and storage medium. Background Art

[0002] With the rapid development and widespread adoption of electric vehicles, ensuring their safety and reliability has become a core goal of the industry. The operational support of electric vehicles involves multiple aspects, including battery management systems, vehicle troubleshooting, and safety testing. The application of fault diagnosis technology is becoming increasingly critical in ensuring the normal operation of vehicles.

[0003] Currently, electric vehicle fault diagnosis relies on various sensor data, such as battery voltage, current, temperature, and pressure. This means that vehicle fault diagnosis is performed based on the data collected by the sensors. However, since vehicle sensors are susceptible to factors such as performance fluctuations and environmental interference, sensor data can become inaccurate or ineffective, greatly increasing the unreliability of sensor data. Fault diagnosis is highly dependent on the accuracy of sensor data, and errors in sensor data can easily lead to misdiagnosis or missed diagnosis of vehicle faults, affecting the accuracy of vehicle fault diagnosis. In severe cases, vehicle faults cannot be promptly and effectively addressed, further affecting the overall operational reliability of the vehicle. Summary of the Invention

[0004] In view of this, the present invention aims to propose a vehicle fault handling method, device, electronic device and storage medium to solve the problem that vehicle faults are easily misdiagnosed or missed due to errors in sensor data, affecting the accuracy of vehicle fault diagnosis. In severe cases, vehicle faults cannot be handled in a timely and effective manner, affecting the overall operational reliability of the vehicle.

[0005] According to a first aspect of the present invention, a vehicle fault handling method is provided, the method comprising:

[0006] Acquiring sensor data from multiple sensors of the same type in the vehicle, preprocessing the sensor data, and fusing the preprocessed sensor data to obtain valid sensor data;

[0007] Performing fault diagnosis on the valid sensor data using a pre-trained fault diagnosis model to obtain a fault event and a fault risk level corresponding to the abnormal sensor data; wherein the fault diagnosis model is pre-trained using historical valid sensor data and historical fault events;

[0008] Performing a fault recovery assessment on the fault event using a pre-trained fault assessment model to generate a recovery assessment result for the fault event; wherein the fault assessment model is pre-trained using fault recovery features in historical valid sensor data, and the recovery assessment result includes whether the fault is self-recoverable or not;

[0009] generating an actual fault level of the fault event according to the fault event, the fault risk level, and the recovery assessment result, and determining a processing strategy corresponding to the actual fault level;

[0010] The vehicle is controlled to execute a processing strategy corresponding to the actual fault level.

[0011] Optionally, acquiring sensor data from multiple sensors of the same type in the vehicle, preprocessing the sensor data, and fusing the preprocessed sensor data to obtain valid sensor data includes:

[0012] Acquire sensor data from multiple sensors of the same type in a vehicle;

[0013] Aligning multiple sensor data of the same type in time and space, and preprocessing the sensors to remove noise data in the sensor data;

[0014] Filtering the preprocessed sensor data of multiple sensors of the same type to obtain target sensor data of the same type of sensor;

[0015] The target sensor data of multiple types of sensors are weightedly fused to obtain effective sensor data.

[0016] Optionally, the using of a pre-trained fault diagnosis model to perform fault diagnosis on the valid sensor data to obtain a fault event and a fault risk level corresponding to the abnormal sensor data includes:

[0017] Inputting the valid sensor data into a pre-trained fault diagnosis model to determine abnormal sensor data in the valid sensor data;

[0018] Obtaining the data source of the abnormal sensor data, and determining the source sensor and the location of the source sensor;

[0019] The abnormal sensor data and the position of the source sensor are used to locate the fault, and a fault event and a fault risk level are output according to the fault location.

[0020] Optionally, the fault diagnosis model is pre-trained using historical valid sensor data and historical fault events, and the training of the fault diagnosis model includes:

[0021] Obtain historical valid sensor data and historical fault events of multiple sensors of the same type in the vehicle;

[0022] Divide the historical valid sensor data and the historical fault events into data sets to obtain a training set and a test set;

[0023] Using the training set to perform abnormal data recognition training and fault location training on a predetermined learning model, to obtain a fault diagnosis result output by the learning model;

[0024] The test set is used to verify the fault diagnosis result output by the learning model, and the training set is used to iteratively train the learning model to obtain a fault diagnosis model.

[0025] Optionally, the using a pre-trained fault assessment model to perform a fault recovery assessment on the fault event to generate a recovery assessment result of the fault event includes:

[0026] Inputting the fault event determined by the fault diagnosis model into a pre-trained fault assessment model to extract the fault recovery features of the fault event; wherein the fault recovery features include the fault frequency, fault duration, and fault location;

[0027] According to the fault frequency, the fault duration and the fault location, determine whether the fault event is restored within a preset time period, and output a recovery evaluation result of the fault event.

[0028] Optionally, generating an actual fault level of the fault event according to the fault event, the fault risk level, and the recovery assessment result, and determining a processing strategy corresponding to the actual fault level includes:

[0029] Acquire multiple pre-classified fault levels and corresponding relationships between the fault levels and processing strategies; wherein the fault level includes at least one of high-risk non-self-recoverable, low-risk non-self-recoverable, low-risk self-recoverable, and no-risk self-recoverable;

[0030] Matching the fault event, the fault risk level, and the recovery assessment result with pre-classified fault levels to generate an actual fault level of the fault event;

[0031] The processing strategy corresponding to the actual fault level is determined according to the corresponding relationship between the fault level and the processing strategy.

[0032] Optionally, controlling the vehicle to execute a processing strategy corresponding to the actual fault level includes:

[0033] If the actual fault level is determined to be high-risk and cannot be self-recovered, emergency braking of the vehicle is controlled and a danger alarm is issued;

[0034] If it is determined that the actual fault level is low-risk and cannot be self-recovered, controlling the vehicle to limit the target function and sending a first prompt message;

[0035] If it is determined that the actual fault level is low-risk and self-recoverable or non-risk and self-recoverable, the fault location is monitored, and in response to monitoring that the fault event self-recovers within a preset time period, a second prompt message is sent.

[0036] According to a second aspect of the present invention, a vehicle fault handling device is provided, the device comprising:

[0037] a data fusion module, configured to obtain sensor data from multiple sensors of the same type in the vehicle, preprocess the sensor data, and fuse the preprocessed sensor data to obtain valid sensor data;

[0038] a fault diagnosis module, configured to perform fault diagnosis on the valid sensor data using a pre-trained fault diagnosis model to obtain a fault event and a fault risk level corresponding to the abnormal sensor data; wherein the fault diagnosis model is pre-trained using historical valid sensor data and historical fault events;

[0039] a fault assessment module, configured to perform a fault recovery assessment on the fault event using a pre-trained fault assessment model, and generate a recovery assessment result for the fault event; wherein the fault assessment model is pre-trained using fault recovery features in historical valid sensor data, and the recovery assessment result includes whether the fault is self-recoverable or not;

[0040] a strategy determination module, configured to generate an actual fault level of the fault event according to the fault event, the fault risk level, and the recovery assessment result, and to determine a processing strategy corresponding to the actual fault level;

[0041] The fault processing module is used to control the vehicle to execute a processing strategy corresponding to the actual fault level.

[0042] According to another aspect of the present invention, there is provided an electronic device, comprising:

[0043] processor;

[0044] a memory for storing instructions executable by the processor;

[0045] Wherein, the processor is configured to execute the instructions to implement the vehicle fault handling method as described above.

[0046] According to another aspect of the present invention, a readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the vehicle fault handling method described above are implemented.

[0047] The vehicle fault handling method provided by an embodiment of the present invention obtains sensor data from multiple sensors of the same type in a vehicle, preprocesses the sensor data, fuses the preprocessed sensor data to obtain valid sensor data, uses a pretrained fault diagnosis model to perform fault diagnosis on the valid sensor data, obtains the fault event and fault severity corresponding to the abnormal sensor data, uses a pretrained fault assessment model to perform fault recovery assessment on the fault event, generates a recovery assessment result for the fault event, and generates an actual fault level of the fault event based on the fault event, fault severity, and recovery assessment result, determines a processing strategy corresponding to the actual fault level, and controls the vehicle to execute the processing strategy corresponding to the actual fault level. The embodiment of the present invention obtains accurate valid sensor data by integrating sensor data from multiple sensors of the same type. Using a model trained based on the accurate sensor data, the method identifies the fault event, its type, severity, and recovery potential. It then comprehensively and objectively assesses fault severity by integrating multiple information sources, avoiding misjudgments caused by a single indicator. This improves the accuracy of fault diagnosis and automatically executes the fault level-based processing strategy without manual intervention, accelerating fault response, significantly improving vehicle safety and maintenance efficiency, and further enhancing overall vehicle operational reliability.

[0048] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are specifically listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0050] Figure 1 This is a flowchart of a vehicle fault handling method provided by an embodiment of the present invention;

[0051] Figure 2 yes Figure 1 Flowchart of step 101 in the vehicle fault handling method provided by an embodiment of the present invention;

[0052] Figure 3 yes Figure 1 Flowchart of step 102 in the vehicle fault handling method provided by an embodiment of the present invention;

[0053] Figure 4 yes Figure 1 Flowchart of step 103 in the vehicle fault handling method provided by the embodiment of the present invention;

[0054] Figure 5 yes Figure 1 Flowchart of step 104 in the vehicle fault handling method provided by an embodiment of the present invention;

[0055] Figure 6 1 is a schematic structural diagram of a vehicle fault handling device provided by an embodiment of the present invention;

[0056] Figure 7 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, each embodiment of the present invention will be described in detail below with reference to the accompanying drawings. However, it will be understood by those skilled in the art that in each embodiment of the present invention, many technical details are provided to enable the reader to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in this application can be implemented. The division of the following embodiments is for convenience of description and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined with each other and referenced to each other under the premise that there is no contradiction.

[0058] Reference Figure 1 , shows a flowchart of the steps of a vehicle fault handling method provided by an embodiment of the present invention, the method may include:

[0059] Step 101 : acquiring sensor data from multiple sensors of the same type in a vehicle, preprocessing the sensor data, and fusing the preprocessed sensor data to obtain valid sensor data.

[0060] It should be noted that in the embodiments of the present invention, sensor types are selected based on the parameters that the electric vehicle needs to monitor, such as battery temperature, vehicle speed, and the pressure applied to the battery pack. Accordingly, temperature sensors, lidars, acceleration sensors, and pressure sensors are selected. To address the problem that fault diagnosis is highly dependent on the accuracy of sensor data, and that sensor data errors can easily lead to misdiagnosis or missed diagnosis of vehicle faults, seriously affecting the accuracy of vehicle fault diagnosis and further affecting the overall operational reliability of the vehicle, this embodiment deploys multiple sensors of the same type on key components of the electric vehicle to ensure that the sensors cover all key areas and improve the accuracy of certain sensor parameters. The specific number and deployment locations of the sensors are determined based on actual vehicle production requirements and are not specifically limited in this embodiment. The number of sensors collecting the same parameter is at least two, achieving sensor redundancy. This utilizes sensor data collected by multiple sensors of the same type, compares and fuses the data from multiple sensors, and obtains accurate and valid sensor data, further improving the accuracy of fault diagnosis.

[0061] Specifically, the vehicle controller obtains sensor data from multiple sensors of the same type in the vehicle, aligns the multiple sensor data of the same type in time and space, preprocesses the sensors, removes noise data in the sensor data, performs interference filtering on the preprocessed multiple sensor data of the same type, obtains the optimal sensor data of the same type of sensor, and finally performs weighted fusion on the optimal sensor data of multiple types of sensors to obtain effective sensor data.

[0062] Step 102: Use a pre-trained fault diagnosis model to perform fault diagnosis on valid sensor data to obtain the fault event and fault risk level corresponding to the abnormal sensor data;

[0063] Among them, the fault diagnosis model is pre-trained using historical valid sensor data and historical fault events.

[0064] In an embodiment of the present invention, after obtaining effective sensor data of optimal sensor data fusion of multiple sensor data of the same type under multiple types of sensors, a pre-trained fault diagnosis model is used to perform fault diagnosis on the effective sensor data, identify abnormal sensor data, and diagnose the fault event and fault risk level corresponding to the abnormal sensor data.

[0065] Specifically, a pre-trained fault diagnosis model is used to identify abnormal sensor data within valid sensor data, obtain the source of the abnormal sensor data, and locate the sensor that collected the abnormal sensor data. The fault is then located based on the abnormal sensor data and the source sensor's location. The fault event and the severity of the fault are then output based on the location of the fault. Fault severity levels range from non-critical to low-critical to high-critical. The fault diagnosis model outputs a detailed fault event, including the specific location of the fault event, the fault type, and the corresponding severity.

[0066] It should be noted that the fault diagnosis model is pre-trained using historical valid sensor data and historical fault events. Specifically, the fault diagnosis model can be based on learning models such as deep neural networks, random forests, and support vector machines, and is trained using historical valid sensor data and historical fault events. The historical valid sensor data is the sensor data fused during the historical detection period, and the historical fault events are the fault events detected during the historical detection period.

[0067] Step 103: Use the pre-trained fault assessment model to perform a fault recovery assessment on the fault event and generate a recovery assessment result of the fault event;

[0068] Among them, the fault assessment model is pre-trained using the fault recovery features in historical valid sensor data, and the recovery assessment results include whether the fault can be self-recovered or not.

[0069] In an embodiment of the present invention, a pre-trained fault assessment model is used to perform fault recovery assessment on a fault event to generate a recovery assessment result of the fault event. The recovery assessment result includes whether the fault can be self-recovered and whether the fault cannot be self-recovered. Specifically, the fault assessment model is pre-trained using fault recovery features in historical valid sensor data. During actual fault diagnosis, the fault event determined by the fault diagnosis model is input into the pre-trained fault assessment model to extract the fault recovery features of the fault event. The fault recovery features include the frequency of fault occurrence, the duration of the fault, and the location of the fault. Based on the frequency of fault occurrence, the duration of the fault, and the location of the fault, it is determined whether the fault event is recovered within a preset time period to obtain a recovery assessment result of the fault event.

[0070] It should be noted that the specific judgment is whether the fault can self-recover within a predefined period of time, that is, whether it will be eliminated naturally without external intervention, so as to output the recovery evaluation result of the fault event. If the fault frequency is lower than the preset threshold, the fault duration is short, and the fault location is in an easily recoverable area, then it is assessed that the fault can self-recover within the predefined period of time, and it is determined that the fault is self-recoverable. If the fault frequency is higher than the preset threshold, the fault duration is long, and the fault location is in a key component, then it is assessed that the fault cannot self-recover within the predefined period of time, and it is determined that the fault is not self-recoverable.

[0071] Step 104 : generating an actual fault level of the fault event according to the fault event, the fault risk level, and the recovery assessment result, and determining a processing strategy corresponding to the actual fault level.

[0072] In an embodiment of the present invention, the fault event level is comprehensively determined based on the fault event, the fault risk level and the recovery assessment result, so as to determine the processing strategy corresponding to the actual fault level. Specifically, based on multiple pre-divided fault levels and the correspondence between the fault levels and the processing strategies, the fault levels include at least one of high-risk non-self-recoverable, low-risk non-self-recoverable, low-risk self-recoverable and no-risk self-recoverable. The fault event, the fault risk level and the recovery assessment result are matched with the pre-divided fault levels to generate the actual fault level of the fault event. According to the correspondence between the fault levels and the processing strategies, the processing strategy corresponding to the actual fault level is determined.

[0073] Specifically, different fault levels are comprehensively distinguished based on the probability of endangering the safety of vehicle occupants, whether it affects the performance of the vehicle battery or the entire vehicle, and whether the fault event can recover by itself. High-risk non-self-recovery refers to a fault with a high probability of endangering the safety of vehicle occupants, affecting the performance of the vehicle battery or the entire vehicle, and cannot recover by itself; low-risk non-self-recovery refers to a fault with a medium probability of endangering the safety of vehicle occupants, affecting the performance of the vehicle battery or the entire vehicle, and cannot recover by itself; low-risk self-recovery refers to a fault with a medium probability of endangering the safety of vehicle occupants, affecting the performance of the vehicle battery or the entire vehicle, and can recover by itself; non-risk self-recovery refers to a fault that does not affect the performance of the vehicle battery or the entire vehicle and can recover by itself. In this embodiment, the processing strategies may include emergency braking, immediate parking, deceleration, starting the backup system, alarming, monitoring and observation, etc. The fault levels and strategies can be adjusted and optimized according to actual needs to support a variety of scenarios, which will not be elaborated here.

[0074] Step 105: Control the vehicle to execute a processing strategy corresponding to the actual fault level.

[0075] In an embodiment of the present invention, a vehicle controller generates control signals to various target control systems in the vehicle based on the actual fault level of the fault event and the corresponding processing strategy for the actual fault level, thereby controlling the vehicle to execute the processing strategy corresponding to the actual fault level. Specifically, if the actual fault level is determined to be high-risk and non-self-recoverable, the vehicle is controlled to perform emergency braking and issue a hazard warning. If the actual fault level is determined to be low-risk and non-self-recoverable, the vehicle is controlled to restrict target functions and send a prompt message. If the actual fault level is determined to be low-risk and self-recoverable or non-risk and self-recoverable, the fault location is monitored and, in response to monitoring that the fault event has self-recovered within a preset time period, a prompt message is sent.

[0076] The vehicle fault handling method provided by an embodiment of the present invention obtains sensor data from multiple sensors of the same type in a vehicle, preprocesses the sensor data, fuses the preprocessed sensor data to obtain valid sensor data, uses a pretrained fault diagnosis model to perform fault diagnosis on the valid sensor data, obtains the fault event and fault severity corresponding to the abnormal sensor data, uses a pretrained fault assessment model to perform fault recovery assessment on the fault event, generates a recovery assessment result for the fault event, and generates an actual fault level of the fault event based on the fault event, fault severity, and recovery assessment result, determines a processing strategy corresponding to the actual fault level, and controls the vehicle to execute the processing strategy corresponding to the actual fault level. The embodiment of the present invention obtains accurate valid sensor data by integrating sensor data from multiple sensors of the same type. Using a model trained based on the accurate sensor data, the method identifies the fault event, its type, severity, and recovery potential. It then comprehensively and objectively assesses fault severity by integrating multiple information sources, avoiding misjudgments caused by a single indicator. This improves the accuracy of fault diagnosis and automatically executes the fault level-based processing strategy without manual intervention, accelerating fault response, significantly improving vehicle safety and maintenance efficiency, and further enhancing overall vehicle operational reliability.

[0077] Further, refer to Figure 2 , showing Figure 1 A flowchart of step 101 of a vehicle fault handling method is provided. This method is substantially the same as the vehicle fault handling method provided in the first embodiment of the present invention. Step 101 may include:

[0078] Step 1011, acquiring sensor data of multiple sensors of the same type in the vehicle;

[0079] Step 1012: aligning multiple sensor data of the same type in time and space, and preprocessing the sensors to remove noise data in the sensor data;

[0080] Step 1013 , filtering the pre-processed sensor data of the same type to obtain target sensor data of the same type of sensor;

[0081] Step 1014 : Perform weighted fusion on the target sensor data of multiple types of sensors to obtain effective sensor data.

[0082] It should be noted that in an embodiment of the present invention, a vehicle is equipped with multiple sensors of the same type, including multiple temperature sensors, multiple pressure sensors, or acceleration sensors. Specifically, data from each sensor is collected via an on-board communication network (such as a CAN bus), or the sampled values ​​of each sensor are collected in real time using a communication protocol to obtain sensor data from multiple sensors of the same type in the vehicle. Based on the sensor sampling timestamps, the data from different sensors are interpolated or compensated to a unified time base for spatiotemporal alignment. This allows the data from different sensors to be aligned in time and value, facilitating data integration, analysis, and comparison. This allows the sensors to send data at the same frequency and format. The unified format helps reduce the complexity of data processing and improves data processing efficiency.

[0083] Specifically, preprocessing includes noise removal and outlier removal, which can clean the collected raw sensor data to remove noise and outliers that may affect the analysis results. In this embodiment, the comparability of data from different sensors is ensured, possible sensor failures or acquisition errors are identified, sensor data is cleaned, noise and outliers are removed, and data is standardized. Data from each sensor is compared to identify inconsistent or expected data. Thresholds are set to identify abnormal spikes and remove or correct them. The data expectation threshold is determined based on statistical analysis of sensor performance and measurement environment, as well as data distribution under normal operating conditions, and is adjusted based on real-time and historical data.

[0084] In this embodiment, the preprocessed sensor data of multiple similar sensors are filtered to obtain target sensor data of the same type of sensors. That is, multiple similar sensors collect sensor parameters for the same parameter. The multiple similar sensor data are filtered, such as by Kalman filtering, particle filtering, etc., to obtain smooth and stable target sensor data. The target sensor data is valid data of the same parameter collected by the same type of sensors. For example, if multiple temperature sensors are deployed on a battery, then in the same detection time window, the multiple temperature sensors collect multiple battery temperatures. The multiple battery temperatures are filtered to obtain a target battery temperature that is closest to the actual situation.

[0085] For example, using Kalman filtering, the state and error covariance at the next moment are predicted based on the system model. The predicted values ​​are corrected using sensor measurements to obtain the optimal estimate, reducing the impact of random interference. After filtering multiple sensor data, accurate target sensor data is obtained. It should be noted that Kalman filtering is used to estimate the system state and reduce uncertainty by considering measurement noise and process noise. The result obtained by the Kalman filtering algorithm is an updated value of the state estimate, that is, the state estimate at time k. By considering measurement noise and process noise to reduce uncertainty, it provides a more accurate estimate of the state. Particle filtering represents probability distributions by simulating a large number of particles and is suitable for nonlinear and non-Gaussian processes. This embodiment does not specifically limit the filtering algorithm used to filter multiple preprocessed sensor data of the same type.

[0086] In this embodiment, target sensor data from multiple types of sensors are weightedly fused to obtain effective sensor data. That is, a weighted fusion strategy is adopted based on the importance and signal quality of different sensors to synthesize multiple types of target sensor data into one effective sensor data. The weights of various sensor data in the weighted fusion can be set by the dynamic performance of the sensor (such as signal-to-noise ratio, stability) or based on actual monitoring needs.

[0087] The embodiments of the present invention achieve multi-type sensor fusion through multi-sensor data collection, spatiotemporal alignment and filtering processing, reduce the impact of noise, interference or sensor failure, improve the stability and reliability of sensor data, and ensure the accuracy of subsequent fault diagnosis.

[0088] Further, refer to Figure 3 , showing Figure 1 A flowchart of step 102 of a vehicle fault handling method is provided. This method is substantially the same as the vehicle fault handling method provided in the first embodiment of the present invention. Step 102 may include:

[0089] Step 1021 , inputting valid sensor data into a pre-trained fault diagnosis model to determine abnormal sensor data in the valid sensor data;

[0090] Step 1022, obtaining the data source of the abnormal sensor data, and determining the source sensor and the location of the source sensor;

[0091] Step 1023 , using the abnormal sensor data and the position of the source sensor to locate the fault, and outputting a fault event and a fault risk level according to the fault location.

[0092] It should be noted that in the embodiment of the present invention, valid sensor data is input into a pre-trained fault diagnosis model to identify abnormal sensor data, that is, the valid sensor data obtained after fusion is sent to the pre-trained fault diagnosis model, wherein the fault diagnosis model can be based on a learning model such as a deep neural network, a random forest, a support vector machine, etc., trained using historical valid sensor data and historical fault events. The specific training process is not described here in detail. The fault diagnosis model can identify anomalies in each valid sensor data.

[0093] In this embodiment, after identifying abnormal sensor data, the data source of the abnormal sensor data is obtained to determine the source sensor and the location of the source sensor. The source sensor is the sensor that collects the abnormal sensor data. The corresponding sensor location can be directly found based on the sensor ID in the abnormal sensor data fed back by the fault diagnosis model, such as the front left, rear right, engine compartment, etc. of the vehicle. It can also be automatically mapped in combination with the vehicle sensor layout diagram to determine the source sensor and the location of the source sensor.

[0094] Specifically, by utilizing the data characteristics of abnormal sensors, such as the deviation value, abnormal fluctuation degree and source sensor location of abnormal sensor data, combined with the fault diagnosis model to locate the fault, the corresponding fault danger level can be evaluated based on the spatial location of the fault and the degree of data abnormality. The fault danger levels include no danger, low danger, high danger, etc. The fault diagnosis model outputs detailed fault events, namely the specific location of the fault event, the fault type and the corresponding danger level.

[0095] The embodiment of the present invention applies a pre-trained model to valid sensor data, quickly identifies abnormal sensor data, obtains the specific location of the abnormal sensor data, locks the fault point, and realizes automated fault detection, positioning and risk assessment.

[0096] In some embodiments, fault location is performed using abnormal sensor data and the location of the source sensor. After outputting a fault event and the degree of fault danger based on the location of the fault, in order to improve the accuracy of the sensor data, the sensor can be calibrated regularly to ensure the accuracy of its measurement data, and the sensitivity and calibration parameters of the sensor can be automatically adjusted based on the real-time sensor data. The specific operating steps are as follows: determine the calibration cycle based on the frequency of use and environmental conditions of the sensor, calibrate the sensor using a known standard or reference value, compare the sensor output with the standard value, and adjust the sensor output to match the standard value, and record the parameters during the calibration process, such as offset, gain, and nonlinear correction coefficient. Specifically, collect real-time data collected by the sensor under different conditions, extract sensor performance and environmental condition-related features from the collected data, identify the relationship between the sensor output and the environmental conditions, and adjust the sensor sensitivity and calibration parameters to adapt to the current environmental conditions.

[0097] Specifically, in the above embodiment, the fault diagnosis model is pre-trained using historical valid sensor data and historical fault events. The training of the fault diagnosis model may specifically include:

[0098] Sub-step 01: Obtain historical valid sensor data and historical fault events of multiple sensors of the same type in the vehicle;

[0099] Sub-step 02: Divide the historical valid sensor data and historical fault events into data sets to obtain training sets and test sets;

[0100] Sub-step 03: Use the training set to perform abnormal data recognition and fault location training on the predetermined learning model to obtain the fault diagnosis results output by the learning model;

[0101] Sub-step 04: Use the test set to verify the fault diagnosis results output by the learning model, and iteratively use the training set to train the learning model to obtain a fault diagnosis model.

[0102] It should be noted that in the above steps, the fault diagnosis model is pre-trained using historical valid sensor data and historical fault events. Specifically, historical valid data (such as temperature, pressure, vibration, etc.) from multiple sensors of the same type in the vehicle is collected to ensure that the data covers a variety of normal and abnormal operating conditions. Corresponding fault events (such as fault occurrence time, fault type, and fault location) are also collected as label information. The historical valid sensor data and historical fault events are pre-processed (cleaned, normalized, etc.) to partition the dataset into training and test sets. The overall dataset can be divided into training and test sets according to a specific ratio (e.g., 80% training, 20% testing).

[0103] Specifically, select an appropriate learning model, such as a deep neural network, random forest, or support vector machine, and use the training set to train the model for anomaly recognition (identifying normal and abnormal data) and fault location (identifying fault type / fault location). During the training process, cross-validation and hyperparameter tuning can be used to obtain optimal model performance. It should be noted that model parameters can be adjusted to optimize performance based on different learning models, cross-validation can be used to evaluate the stability and generalization ability of the model, and test set data can be used to verify the accuracy of the model. Specifically, indicators such as accuracy, recall, and F1 score are used to evaluate the performance of the model to avoid overfitting. Based on the verification results, the model structure or training parameters are adjusted, and multiple rounds of training are performed until the model performance reaches the expected standard. Ultimately, a mature fault diagnosis model is formed and applied to the analysis of sensor data in actual vehicles.

[0104] The embodiment of the present invention uses a large amount of historical valid sensor data and fault data for training, so that the model can identify different types of faults and anomalies, thereby improving the accuracy of fault diagnosis.

[0105] Further, refer to Figure 4 , showing Figure 1 A flowchart of step 103 in a vehicle fault handling method is provided. This method is substantially the same as the vehicle fault handling method provided in the first embodiment of the present invention. Step 103 may include:

[0106] Step 1031: input the fault event determined by the fault diagnosis model into the pre-trained fault assessment model to extract the fault recovery feature of the fault event;

[0107] Among them, fault recovery characteristics include fault frequency, fault duration and fault location.

[0108] Step 1032 , based on the fault frequency, fault duration, and fault location, determine whether the fault event is recovered within a preset time period, and output a recovery evaluation result of the fault event.

[0109] It should be noted that, in an embodiment of the present invention, the fault event identified by the fault diagnosis model is used as input and passed to a pre-trained fault assessment model for extracting fault recovery features. The fault recovery features include the fault frequency, fault duration and fault location. Based on the fault frequency, fault duration and fault location, it is determined whether the fault event is recovered within a preset time period, and a recovery assessment result of the fault event is output. The fault event includes information such as the fault type, fault location and fault occurrence time. The fault frequency is the number of times a fault occurs repeatedly at a certain location within a certain time window. The fault duration is the time from the occurrence to the elimination of a single fault. The fault location is the specific location where the fault occurs, a sensor point or part number.

[0110] Specifically, the extracted fault recovery features are judged to determine whether the fault can self-recover within a predefined time period (such as 5 minutes), that is, whether it can be eliminated naturally without external intervention, and thus output the recovery assessment result of the fault event. The recovery assessment results include whether the fault is self-recoverable or not. Specifically, if the fault frequency is lower than the preset threshold, the fault duration is short, and the fault location is in an easily recoverable area, such as a backup line, then the fault is judged to be self-recoverable. If the fault frequency is higher than the preset threshold, the fault duration is long, and the fault location is in a critical component, then the fault is judged to be non-self-recoverable.

[0111] The embodiment of the present invention dynamically determines whether the fault has self-repair capabilities by analyzing characteristics such as fault frequency, fault duration, and fault location, so as to integrate the fault recovery capability into the fault level and fault handling strategy, thereby improving the accuracy of fault diagnosis.

[0112] Further, refer to Figure 5 , showing Figure 1 A flowchart of step 104 of a vehicle fault handling method is provided. This method is substantially the same as the vehicle fault handling method provided in the first embodiment of the present invention. Step 104 may include:

[0113] Step 1041: Acquire multiple pre-classified fault levels and corresponding relationships between fault levels and processing strategies;

[0114] The fault level includes at least one of high-risk non-self-recoverable, low-risk non-self-recoverable, low-risk self-recoverable, and no-risk self-recoverable.

[0115] Step 1042 , matching the fault event, fault risk level, and recovery assessment result with the pre-classified fault levels to generate an actual fault level of the fault event;

[0116] Step 1043: Determine the processing strategy corresponding to the actual fault level according to the correspondence between the fault level and the processing strategy.

[0117] It should be noted that in the embodiments of the present invention, multiple pre-classified fault levels and corresponding relationships between fault levels and handling strategies are obtained. The fault levels include at least one of high-risk, non-self-recoverable, low-risk, non-self-recoverable, and non-risk, self-recoverable. High-risk, non-self-recoverable is a serious fault that cannot be self-recovered and must be handled immediately; low-risk, non-self-recoverable is a medium or minor fault that is difficult to repair automatically and is recommended to be handled promptly; low-risk, self-recoverable is a medium or minor fault that is automatically repaired; and non-risk, self-recoverable is a harmless fault that can be self-recovered or does not require handling. Each fault level corresponds to a specific handling strategy.

[0118] Specifically, the corresponding processing strategy is determined according to the current fault level of the vehicle. Different fault levels are distinguished based on the probability of endangering the safety of vehicle occupants, whether it affects the vehicle battery or the performance of the entire vehicle, and whether the fault event can recover on its own. For example, the fault level includes at least one of high-risk non-self-recoverable, low-risk non-self-recoverable, low-risk self-recoverable, and non-risk self-recoverable. High-risk non-self-recoverable is a fault with a high probability of endangering the safety of vehicle occupants, affecting the vehicle battery or the performance of the entire vehicle, and cannot recover on its own, requiring immediate action; low-risk non-self-recoverable is a fault with a medium probability of endangering the safety of vehicle occupants, affecting the vehicle battery or the performance of the entire vehicle, and cannot recover on its own; low-risk self-recoverable is a fault with a medium probability of endangering the safety of vehicle occupants, affecting the vehicle battery or the performance of the entire vehicle, and cannot recover on its own; low-risk self-recoverable is a fault with a medium probability of endangering the safety of vehicle occupants, affecting the vehicle battery or the performance of the entire vehicle, and can recover on its own; and non-risk self-recoverable is a fault that does not affect the vehicle battery or the performance of the entire vehicle and can recover on its own.

[0119] In this embodiment, the diagnostic and assessment information of the fault event, namely the fault event, fault danger level, and recovery assessment results, is matched with the above-mentioned fault level list to determine the actual fault level that the fault event matches. If the fault danger level is high and the fault cannot be self-recovered, it is determined to be a high-risk, non-self-recoverable fault level. If the fault danger level is low and the fault can be self-recovered, it is determined to be non-hazardous and self-recoverable. Based on the fault level, the corresponding processing measures are obtained from the corresponding relationship of the predefined level strategy, and the processing strategy instructions are transmitted to the vehicle control system or maintenance arrangement. The processing strategies may include emergency braking, immediate stop, deceleration, activation of backup systems, alarm, monitoring and observation, etc. The fault level and strategy can be adjusted and optimized according to actual needs to support a variety of scenarios, which will not be detailed here.

[0120] The embodiment of the present invention comprehensively and objectively determines the fault level of a fault event by integrating the fault event, the fault risk level and the recovery assessment results, accurately and quickly judges the fault level, ensures timely identification and response to dangers, and facilitates the adoption of different treatment measures according to different risk levels.

[0121] Specifically, in some embodiments, step 105 controls the vehicle to execute a processing strategy corresponding to the actual fault level, which may specifically include:

[0122] Sub-step 01: If the actual fault level is determined to be high-risk and cannot be self-recovered, emergency braking of the vehicle is controlled and a danger alarm is issued;

[0123] Sub-step 02: If the actual fault level is determined to be low-risk and non-self-recoverable, the vehicle is controlled to limit target functions and a first prompt message is sent;

[0124] Sub-step 03: If the actual fault level is determined to be low-risk and self-recoverable or non-risk and self-recoverable, the fault location is monitored, and in response to monitoring that the fault event self-recovers within a preset time period, a second prompt message is sent.

[0125] Specifically, when the vehicle detects that the fault level is high-risk and cannot be self-recovered, the emergency braking procedure is immediately activated to ensure that the vehicle slows down and stops as quickly as possible to avoid accidents. At the same time, a danger alarm is issued, such as an audible and visual alarm, and the front and rear warning lights are on to alert passengers and surrounding vehicles. Among them, emergency braking control is implemented by the vehicle's control unit, and the alarm information is notified to the owner through the on-board display, buzzer or wireless network; when the vehicle detects a low-risk and cannot be self-recovered, the vehicle restricts some target functions. The target function restriction is implemented by the vehicle control strategy, such as disabling certain automatic driving capabilities, shutting down the fault-related subsystems, restricting certain vehicle operations, etc. At the same time, the first A prompt message, which can be conveyed to the driver through the vehicle display, voice prompt, etc. The first prompt message is used to remind the user of the fault level and some system function limitations; if the vehicle detects low-risk and self-recovery or no-risk and self-recovery, the fault location is monitored, and in response to monitoring that the fault event self-recovers within the preset time, a second prompt message is sent. Specifically, the monitoring is carried out to see whether it is automatically recovered within the preset time (the fault event eliminates itself or the data returns to normal). If the monitoring confirms that the event self-recovers within the preset time during the period, the second prompt message is sent. The second prompt message is used to remind the user of the fault level and that the fault has been automatically repaired.

[0126] For example, the specific handling strategy for high-risk non-self-recovery can be to stop the vehicle immediately, ensure the safety of people, and park the vehicle in a safe location; emergency response, initiate emergency response procedures, such as contacting rescue services; the specific handling strategy for low-risk non-self-recovery can be to control vehicle functions, such as reducing speed or limiting power output, promptly notify the driver, inform the driver that the vehicle has performance problems, and recommend repairs as soon as possible; the handling strategy for low-risk self-recovery and non-risk self-recovery can be to continuously monitor the fault status, record relevant data, and remind the driver of the fault recovery status.

[0127] The embodiment of the present invention adopts differentiated response measures for faults of different fault levels, reasonably handles vehicle faults according to the fault level, and uses automated processing strategies to shorten reaction time, reduce human delays, and improve emergency processing speed.

[0128] Reference Figure 6 , shows a schematic structural diagram of a vehicle fault handling device provided by an embodiment of the present invention, the device comprising:

[0129] A data fusion module 201 is used to obtain sensor data from multiple sensors of the same type in the vehicle, preprocess the sensor data, and fuse the preprocessed sensor data to obtain valid sensor data;

[0130] A fault diagnosis module 202 is configured to perform fault diagnosis on the valid sensor data using a pre-trained fault diagnosis model to obtain a fault event and a fault risk level corresponding to the abnormal sensor data; wherein the fault diagnosis model is pre-trained using historical valid sensor data and historical fault events;

[0131] A fault assessment module 203 is configured to perform a fault recovery assessment on the fault event using a pre-trained fault assessment model, and generate a recovery assessment result for the fault event; wherein the fault assessment model is pre-trained using fault recovery features in historical valid sensor data, and the recovery assessment result includes whether the fault is self-recoverable or not self-recoverable;

[0132] A strategy determination module 204 is configured to generate an actual fault level of the fault event according to the fault event, the fault risk level, and the recovery assessment result, and to determine a processing strategy corresponding to the actual fault level;

[0133] The fault processing module 205 is used to control the vehicle to execute a processing strategy corresponding to the actual fault level.

[0134] Furthermore, the data fusion module 201 includes:

[0135] A first acquisition submodule is used to acquire sensor data of multiple sensors of the same type in the vehicle;

[0136] A preprocessing submodule, configured to align multiple similar sensor data in time and space, and preprocess the sensors to remove noise data from the sensor data;

[0137] The filtering submodule is used to filter the preprocessed sensor data of multiple sensors of the same type to obtain target sensor data of the same type of sensor;

[0138] The fusion submodule is used to perform weighted fusion on the target sensor data of multiple types of sensors to obtain effective sensor data.

[0139] Furthermore, the fault diagnosis module 202 includes:

[0140] a first determining submodule, configured to input the valid sensor data into a pre-trained fault diagnosis model to determine abnormal sensor data in the valid sensor data;

[0141] A second determining submodule is configured to obtain a data source of the abnormal sensor data, and determine a source sensor and a location of the source sensor;

[0142] The fault location submodule is used to locate the fault using the abnormal sensor data and the position of the source sensor, and output a fault event and a fault risk level according to the fault location.

[0143] Furthermore, the fault diagnosis module is specifically configured such that the training of the fault diagnosis model includes:

[0144] A second acquisition submodule is used to acquire historical valid sensor data and historical fault events of multiple sensors of the same type in the vehicle;

[0145] A partitioning submodule, configured to partition the historical valid sensor data and the historical fault events into data sets to obtain a training set and a test set;

[0146] A training submodule is used to perform abnormal data recognition training and fault location training on a predetermined learning model using the training set to obtain a fault diagnosis result output by the learning model;

[0147] The verification submodule is used to verify the fault diagnosis result output by the learning model using the test set, and iteratively train the learning model using the training set to obtain a fault diagnosis model.

[0148] Furthermore, the fault assessment module 203 includes:

[0149] An extraction submodule, configured to input the fault event determined by the fault diagnosis model into a pre-trained fault assessment model to extract the fault recovery features of the fault event; wherein the fault recovery features include the fault frequency, fault duration, and fault location;

[0150] The evaluation submodule is used to determine whether the fault event is recovered within a preset time period based on the fault frequency, the fault duration and the fault location, and output a recovery evaluation result of the fault event.

[0151] Furthermore, the strategy determination module 204 includes:

[0152] A third acquisition submodule is configured to acquire a plurality of pre-classified fault levels and a correspondence between the fault levels and processing strategies; wherein the fault level includes at least one of high-risk non-self-recoverable, low-risk non-self-recoverable, low-risk self-recoverable, and no-risk self-recoverable;

[0153] a generating submodule, configured to match the fault event, the fault risk level, and the recovery assessment result with pre-classified fault levels to generate an actual fault level of the fault event;

[0154] The third determining submodule is configured to determine the processing strategy corresponding to the actual fault level according to the correspondence between the fault level and the processing strategy.

[0155] Furthermore, the fault handling module 205 includes:

[0156] A first processing submodule is configured to control the vehicle to perform emergency braking and issue a danger alarm if it is determined that the actual fault level is high-risk and cannot be self-recovered;

[0157] A second processing submodule is configured to control the vehicle to limit target functions and send a first prompt message if it is determined that the actual fault level is low-risk and cannot be self-recovered;

[0158] The third processing submodule is used to monitor the fault location if it is determined that the actual fault level is low-risk and self-recoverable or non-risk and self-recoverable, and send a second prompt message in response to monitoring that the fault event self-recovers within a preset time period.

[0159] The vehicle fault handling device provided by an embodiment of the present invention obtains sensor data from multiple sensors of the same type in a vehicle, preprocesses the sensor data, and fuses the preprocessed sensor data to obtain valid sensor data. A pretrained fault diagnosis model is then used to perform fault diagnosis on the valid sensor data, obtaining the fault event and fault severity corresponding to the abnormal sensor data. A pretrained fault assessment model is then used to perform a fault recovery assessment on the fault event, generating a recovery assessment result for the fault event. Based on the fault event, fault severity, and recovery assessment result, the actual fault level of the fault event is determined, a processing strategy corresponding to the actual fault level is determined, and the vehicle is controlled to execute the processing strategy corresponding to the actual fault level. By integrating sensor data from multiple sensors of the same type to obtain accurate valid sensor data, the embodiment of the present invention uses a model trained based on the accurate sensor data to identify the fault event, its type, severity, and recovery potential. This system integrates multiple information sources to comprehensively and objectively assess fault severity, avoiding misjudgments caused by a single indicator. This improves the accuracy of fault diagnosis and automatically executes the fault level-based processing strategy without manual intervention, accelerating fault response, significantly improving vehicle safety and maintenance efficiency, and further enhancing overall vehicle operational reliability.

[0160] Reference Figure 7 , an embodiment of the present invention further provides an electronic device, such as Figure 7As shown, it includes a processor 301, a communication interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communication interface 302, and the memory 303 communicate with each other through the communication bus 304.

[0161] Processor 301, memory 303 for storing processor-executable instructions;

[0162] The processor 301 is configured to execute the instructions to implement the vehicle fault handling method described below:

[0163] Acquiring sensor data from multiple sensors of the same type in the vehicle, preprocessing the sensor data, and fusing the preprocessed sensor data to obtain valid sensor data;

[0164] Performing fault diagnosis on the valid sensor data using a pre-trained fault diagnosis model to obtain a fault event and a fault risk level corresponding to the abnormal sensor data; wherein the fault diagnosis model is pre-trained using historical valid sensor data and historical fault events;

[0165] Performing a fault recovery assessment on the fault event using a pre-trained fault assessment model to generate a recovery assessment result for the fault event; wherein the fault assessment model is pre-trained using fault recovery features in historical valid sensor data, and the recovery assessment result includes whether the fault is self-recoverable or not;

[0166] generating an actual fault level of the fault event according to the fault event, the fault risk level, and the recovery assessment result, and determining a processing strategy corresponding to the actual fault level;

[0167] The vehicle is controlled to execute a processing strategy corresponding to the actual fault level.

[0168] The communication bus mentioned in the terminal can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is used in the figure, but this does not mean that there is only one bus or only one type of bus.

[0169] The communication interface is used for communication between the above terminal and other devices.

[0170] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0171] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0172] In another embodiment provided by the present invention, a computer-readable storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, the vehicle fault handling method in the above embodiment is implemented.

[0173] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0174] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0175] Each embodiment in this specification is described in a related manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiment is generally similar to the method embodiment, so the description is relatively simple. For related parts, refer to the description of the method embodiment.

[0176] The above description is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention are included in the scope of protection of the present invention.

Claims

1. A vehicle fault handling method, characterized in that: The method comprises: Acquiring sensor data from multiple sensors of the same type in the vehicle, preprocessing the sensor data, and fusing the preprocessed sensor data to obtain valid sensor data; Performing fault diagnosis on the valid sensor data using a pre-trained fault diagnosis model to obtain a fault event and a fault risk level corresponding to the abnormal sensor data; wherein the fault diagnosis model is pre-trained using historical valid sensor data and historical fault events; Performing a fault recovery assessment on the fault event using a pre-trained fault assessment model to generate a recovery assessment result for the fault event; wherein the fault assessment model is pre-trained using fault recovery features in historical valid sensor data, and the recovery assessment result includes whether the fault is self-recoverable or not; generating an actual fault level of the fault event according to the fault event, the fault risk level, and the recovery assessment result, and determining a processing strategy corresponding to the actual fault level; The vehicle is controlled to execute a processing strategy corresponding to the actual fault level.

2. The method according to claim 1, characterized in that The acquiring of sensor data from a plurality of sensors of the same type in the vehicle, preprocessing the sensor data, and fusing the preprocessed sensor data to obtain valid sensor data includes: Acquire sensor data from multiple sensors of the same type in a vehicle; Aligning multiple sensor data of the same type in time and space, and preprocessing the sensors to remove noise data in the sensor data; Filtering the preprocessed sensor data of multiple sensors of the same type to obtain target sensor data of the same type of sensor; The target sensor data of multiple types of sensors are weightedly fused to obtain effective sensor data.

3. The method according to claim 1, characterized in that The method of using a pre-trained fault diagnosis model to perform fault diagnosis on the valid sensor data to obtain a fault event and a fault risk level corresponding to the abnormal sensor data includes: Inputting the valid sensor data into a pre-trained fault diagnosis model to determine abnormal sensor data in the valid sensor data; Obtaining the data source of the abnormal sensor data, and determining the source sensor and the location of the source sensor; The abnormal sensor data and the position of the source sensor are used to locate the fault, and a fault event and a fault risk level are output according to the fault location.

4. The method according to claim 3, characterized in that The fault diagnosis model is pre-trained using historical valid sensor data and historical fault events. The training of the fault diagnosis model includes: Obtain historical valid sensor data and historical fault events of multiple sensors of the same type in the vehicle; Divide the historical valid sensor data and the historical fault events into data sets to obtain a training set and a test set; Using the training set to perform abnormal data recognition training and fault location training on a predetermined learning model, to obtain a fault diagnosis result output by the learning model; The test set is used to verify the fault diagnosis result output by the learning model, and the training set is used to iteratively train the learning model to obtain a fault diagnosis model.

5. The method according to claim 1, wherein The using of the pre-trained fault assessment model to perform a fault recovery assessment on the fault event to generate a recovery assessment result of the fault event includes: Inputting the fault event determined by the fault diagnosis model into a pre-trained fault assessment model to extract the fault recovery features of the fault event; wherein the fault recovery features include the fault frequency, fault duration, and fault location; According to the fault frequency, the fault duration and the fault location, determine whether the fault event is restored within a preset time period, and output a recovery evaluation result of the fault event.

6. The method according to claim 1, characterized in that Generating an actual fault level of the fault event according to the fault event, the fault risk level, and the recovery assessment result, and determining a processing strategy corresponding to the actual fault level, includes: Acquire multiple pre-classified fault levels and corresponding relationships between the fault levels and processing strategies; wherein the fault level includes at least one of high-risk non-self-recoverable, low-risk non-self-recoverable, low-risk self-recoverable, and no-risk self-recoverable; Matching the fault event, the fault risk level, and the recovery assessment result with pre-classified fault levels to generate an actual fault level of the fault event; The processing strategy corresponding to the actual fault level is determined according to the corresponding relationship between the fault level and the processing strategy.

7. The method according to claim 6, characterized in that The controlling the vehicle to execute a processing strategy corresponding to the actual fault level includes: If the actual fault level is determined to be high-risk and cannot be self-recovered, emergency braking of the vehicle is controlled and a danger alarm is issued; If it is determined that the actual fault level is low-risk and cannot be self-recovered, controlling the vehicle to limit the target function and sending a first prompt message; If it is determined that the actual fault level is low-risk and self-recoverable or non-risk and self-recoverable, the fault location is monitored, and in response to monitoring that the fault event self-recovers within a preset time period, a second prompt message is sent.

8. A vehicle fault handling device, characterized in that: The device comprises: a data fusion module, configured to obtain sensor data from multiple sensors of the same type in the vehicle, preprocess the sensor data, and fuse the preprocessed sensor data to obtain valid sensor data; a fault diagnosis module, configured to perform fault diagnosis on the valid sensor data using a pre-trained fault diagnosis model to obtain a fault event and a fault risk level corresponding to the abnormal sensor data; wherein the fault diagnosis model is pre-trained using historical valid sensor data and historical fault events; a fault assessment module, configured to perform a fault recovery assessment on the fault event using a pre-trained fault assessment model, and generate a recovery assessment result for the fault event; wherein the fault assessment model is pre-trained using fault recovery features in historical valid sensor data, and the recovery assessment result includes whether the fault is self-recoverable or not; a strategy determination module, configured to generate an actual fault level of the fault event according to the fault event, the fault risk level, and the recovery assessment result, and to determine a processing strategy corresponding to the actual fault level; The fault processing module is used to control the vehicle to execute a processing strategy corresponding to the actual fault level.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to execute the instructions to implement the vehicle fault handling method according to any one of claims 1 to 7.

10. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the vehicle fault handling method according to any one of claims 1 to 7 is implemented.

Images