Fault monitoring methods, devices, equipment and storage media
By employing fault rules and diagnostic models in intelligent driving systems to filter and process vehicle data, and utilizing machine learning and deep learning technologies, the problems of misjudgment and missed judgment in fault monitoring in intelligent driving systems have been solved, thereby improving the fault monitoring rate and system reliability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHONGQING CHANGAN TECH CO LTD
- Filing Date
- 2023-06-25
- Publication Date
- 2026-07-17
Smart Images
Figure CN116767247B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of fault monitoring technology, specifically to a fault monitoring method, device, equipment, and storage medium. Background Technology
[0002] Automobiles and other similar devices, as a means of human transportation, have brought great convenience to human travel. The widespread use of automobiles has promoted social development and improved people's quality of life.
[0003] Furthermore, with the rapid development of artificial intelligence and big data technologies, intelligent driving systems are becoming a hot topic in the automotive industry. Traditional driving methods require human intervention, posing safety risks. In contrast, big data-based intelligent driving systems can achieve autonomous driving through data collection, processing, and analysis, improving driving safety, efficiency, and comfort.
[0004] However, fault monitoring in intelligent driving systems still presents challenges. Existing fault monitoring technologies primarily rely on sensor data and human judgment, which can easily lead to misjudgments and missed detections, resulting in malfunctions occurring during vehicle operation that cannot be identified and addressed in a timely manner. Summary of the Invention
[0005] One objective of this application is to provide a fault monitoring method, which can effectively improve the fault monitoring rate of intelligent driving systems and further enhance the safety and reliability of intelligent driving; a second objective of this application is to provide a fault monitoring device; a third objective of this application is to provide an equipment; and a fourth objective of this application is to provide a storage medium.
[0006] To achieve the above objectives, firstly, this application provides a fault monitoring method for intelligent driving, the fault monitoring method comprising:
[0007] Receive vehicle data sent by the vehicle;
[0008] Based on the fault rules satisfied by the vehicle data and the fault rule configuration information, the target fault coarse classification type corresponding to the vehicle data is determined; wherein, the fault rule configuration information includes the correspondence between multiple fault rules and fault coarse classification types;
[0009] Select fault data corresponding to the target fault coarse classification from the vehicle data;
[0010] The fault data is processed to determine the target feature data;
[0011] The target feature data is input into the fault diagnosis model to predict the fault diagnosis information corresponding to the target feature data; wherein, the fault diagnosis information includes fault sub-types, and the fault coarse-classification type includes at least one fault sub-type.
[0012] Furthermore, the vehicle data is time-series data, and the vehicle data includes at least one of the following:
[0013] The sensor data collected by the vehicle's sensors, the vehicle's driving data, and the camera data collected by the vehicle's camera unit.
[0014] Further, the step of processing the fault data to determine target feature data includes:
[0015] The fault data is subjected to a setting process to determine initial feature data; wherein, the setting process includes data cleaning, feature extraction, and data clustering;
[0016] Anomaly detection is performed on the initial diagnostic data, and the data containing anomalies in the initial feature data are identified as target feature data.
[0017] Furthermore, the fault diagnosis model is determined in the following way:
[0018] Obtain a training sample set; wherein the training sample set includes multiple training sample pairs, each training sample pair includes an input sample and an output sample, the output sample includes a fault diagnosis information sample, the input sample includes a target feature data sample, and the target feature data sample is obtained by processing the fault data sample corresponding to the fault diagnosis information sample.
[0019] The original model is trained based on the training sample set to determine the fault diagnosis model.
[0020] Further, the step of training the original model based on the training sample set to determine the fault diagnosis model includes:
[0021] Based on the training sample set, the original model is trained using an adaptive learning rate scheduling strategy to determine the fault diagnosis model.
[0022] Furthermore, the original model includes an attention mechanism.
[0023] Furthermore, the fault diagnosis information also includes the fault severity corresponding to the fault sub-type.
[0024] To achieve the above objectives, secondly, this application also provides a fault monitoring device, the fault monitoring device comprising:
[0025] The receiving module is used to receive the vehicle data sent by the vehicle.
[0026] The data processing module is used to determine the target coarse fault classification type corresponding to the vehicle data based on the fault rules satisfied by the vehicle data and the fault rule configuration information; wherein, the fault rule configuration information includes the correspondence between multiple fault rules and fault coarse classification types;
[0027] It is also used to select fault data from the vehicle data that corresponds to the target fault coarse classification type;
[0028] It is also used to process the fault data to determine target feature data;
[0029] It is also used to input the target feature data into a fault diagnosis model to predict the fault diagnosis information corresponding to the target feature data; wherein, the fault diagnosis information includes fault subdivision types, and the fault coarse classification type includes at least one fault subdivision type.
[0030] To achieve the above objectives, in a third aspect, this application also provides an apparatus comprising: a processor and a memory, wherein the processor is configured to execute a control program stored in the memory to implement the fault monitoring method described above.
[0031] To achieve the above objectives, in a fourth aspect, this application also provides a storage medium storing one or more programs that can be executed by one or more processors to implement the fault monitoring method described above.
[0032] The beneficial effects of this application are:
[0033] In this application, vehicles employing intelligent driving technology can transmit vehicle data to a cloud platform device. The cloud platform device then first judges the vehicle data based on fault rules to predict the coarse types of possible faults. Next, it uses these coarse fault types to filter the vehicle data, obtaining fault data that may cause faults. Finally, based on a fault diagnosis model, it predicts the final detailed fault type. This effectively improves the fault detection rate of the intelligent driving system, further enhancing the safety and reliability of intelligent driving. Furthermore, because the data is filtered based on the coarse fault types before using the fault diagnosis model, the data processing load of the fault diagnosis model is reduced, increasing data processing speed and further improving the efficiency of fault monitoring. Attached Figure Description
[0034] Figure 1 This diagram illustrates a flowchart of a fault monitoring method provided in an embodiment of this application.
[0035] Figure 2 This diagram illustrates the structure of a fault monitoring device according to an embodiment of this application.
[0036] Figure 3 This illustration shows a structural diagram of a device provided in an embodiment of this application;
[0037] in:
[0038] 10. Receiving module; 20. Data processing module;
[0039] 100. Device; 101. Processor; 102. Memory; 1021. Operating system; 1022. Application program; 103. User interface; 104. Network interface; 105. Bus system. Detailed Implementation
[0040] The embodiments of this application will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be understood that the preferred embodiments are only for illustrating this application and are not intended to limit the scope of protection of this application.
[0041] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. Therefore, the drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.
[0042] The embodiments of this application will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be understood that the preferred embodiments are only for illustrating this application and are not intended to limit the scope of protection of this application.
[0043] To facilitate understanding of the embodiments of this application, the following will provide further explanation and description with reference to the accompanying drawings and specific embodiments. These embodiments do not constitute a limitation on the embodiments of this application.
[0044] This embodiment provides a fault monitoring method that can be applied to equipment. (See reference...) Figure 1 As shown, the method may include:
[0045] S110, Receive vehicle data sent by the vehicle;
[0046] S120. Based on the fault rules satisfied by the vehicle data and the fault rule configuration information, determine the target coarse fault classification type corresponding to the vehicle data; wherein, the fault rule configuration information includes the correspondence between multiple fault rules and fault coarse classification types.
[0047] S130. Select fault data corresponding to the target fault coarse classification type from the set vehicle data;
[0048] S140. Perform data processing on the fault data to determine the target characteristic data;
[0049] S150. Input the target feature data into the fault diagnosis model to predict the fault diagnosis information corresponding to the target feature data; wherein, the fault diagnosis information includes fault sub-types, and the fault coarse-types include at least one fault sub-type.
[0050] In step S110, the vehicle data may include at least one of the following: sensor data collected by the vehicle's sensors, vehicle driving data, camera data collected by the vehicle's camera unit, etc.
[0051] Vehicles equipped with intelligent driving technology can be equipped with multiple sensors, such as speed sensors, steering sensors, acceleration sensors, braking sensors, and engine speed sensors. These sensors can collect real-time data on vehicle operation. It should be noted that higher sensor accuracy results in more accurate data, better ensuring the reliability of fault monitoring. The data collected by the sensors can be transmitted to the onboard computer system.
[0052] The camera unit may include a camera, which may be a high-resolution camera to improve the accuracy of the camera data.
[0053] This application provides a reliable data foundation for data mining and fault diagnosis by comprehensively and accurately collecting various data from intelligent driving systems, thereby improving the reliability of fault monitoring.
[0054] Vehicles can transmit vehicle data to a cloud platform device via a cloud network. This cloud platform device can be, for example, a cloud server. The cloud network can be a high-speed, stable wireless network, such as a 5G network, to ensure real-time data transmission and security. After receiving the vehicle data, the cloud server can use distributed storage technology to store the data, ensuring data security and stability, and enabling data backup and recovery. Using a cloud platform for data storage and processing allows for real-time data processing and querying, improving system efficiency and response speed.
[0055] In step S120, the fault rules can be some preset conditions. Faults that meet the same preset conditions can be grouped into the same type of fault. Specifically, based on vehicle data, it can be determined what kind of fault occurred, and then the faults are classified according to the aforementioned fault rules to determine the coarse fault type.
[0056] The fault rule configuration information includes multiple fault rules and multiple coarse fault classification types, and also includes a one-to-one correspondence between the multiple fault rules and the multiple coarse fault classification types. Therefore, once the fault rules that the vehicle data meets are determined, the coarse fault classification type corresponding to the fault rule can be selected from the fault rule configuration information, and the selected coarse fault classification type is determined as the target coarse fault classification type corresponding to the vehicle data.
[0057] In step S130, after the cloud platform device determines the target fault coarse classification type corresponding to the vehicle data, it can filter the vehicle data based on the target fault coarse classification type, select the data in the vehicle data that is related to the target fault coarse classification type, and record the selected data as fault data.
[0058] This method, based on big data fault monitoring, can quickly detect and diagnose faults in intelligent driving systems, thereby improving the reliability and safety of the system.
[0059] In step S140, after the cloud platform device obtains the fault data from the vehicle data, it can process the fault data, that is, preprocess the fault data to obtain the target feature data corresponding to the fault data, so as to facilitate subsequent fault diagnosis based on the fault diagnosis model.
[0060] The process involves first setting and processing the fault data to determine the initial feature data, and then performing anomaly detection on the initial feature data to identify the data with anomalies as the target feature data.
[0061] The processing steps can include data cleaning, feature extraction, and data clustering, among others.
[0062] In the data cleaning section, methods such as data deduplication, missing value handling, and outlier handling can be employed. Data deduplication refers to removing duplicate data from the collected data to avoid the impact of duplicate data. Missing value handling refers to filling in or deleting missing values from the collected data. Outlier handling refers to detecting and handling outliers in the collected data, which can be based on statistical methods such as the 3σ principle and box plots, and is not limited to these methods.
[0063] Feature extraction refers to extracting useful features from data. In this method, it refers to feature extraction from cleaned data. Specific features, such as acceleration and steering angle, can be extracted from the data using feature extraction algorithms.
[0064] Regarding data clustering, the number of clusters can be determined through experimentation to achieve better clustering results. The clustering method can be chosen based on the specific circumstances, using appropriate algorithms such as K-means and hierarchical clustering, without any specific limitations.
[0065] Anomaly detection refers to the process of detecting anomalies in data, specifically in the initial feature data. This process identifies anomalous data within the initial feature data as target feature data, thereby uncovering potential anomalies and faults.
[0066] In step S150, after the cloud platform device obtains the target feature data, it can process the target feature data based on the fault diagnosis model. That is, the target feature data is input into the fault diagnosis model, and the fault diagnosis model can output the fault diagnosis information corresponding to the target feature data, so as to realize fault prediction based on the target feature data.
[0067] The fault diagnosis information may include fault sub-categories, and each coarse fault category may include at least one fault sub-category. The fault diagnosis information may also include the fault severity corresponding to each sub-category, to better achieve fault monitoring. Of course, the fault diagnosis information may also include other fault-related information, without limitation.
[0068] Among them, the fault diagnosis model can use models such as support vector machine (SVM) and artificial neural network (ANN) to predict the fault type, and can predict the fault situation (e.g., fault severity) by analyzing the trend, periodicity, and seasonality of time series data.
[0069] Among them, the fault diagnosis model can first preprocess the input data, such as cleaning, denoising and calibrating the data, to improve the accuracy and reliability of subsequent analysis.
[0070] Furthermore, fault diagnosis models can incorporate ensemble learning methods (such as voting, stacking, and boosting) to improve the accuracy and robustness of fault detection and judgment. Additionally, fault diagnosis models can integrate expert domain knowledge, incorporating parameters derived from the experience and judgment of domain experts into the model to enhance the reliability of detection and judgment. Interpretable models can also be employed to better understand the model's decision-making process and fault judgment criteria, increasing trust in the model's output and providing valuable clues for further fault diagnosis and repair.
[0071] Once the fault diagnosis information is determined, it can be analyzed to identify the cause of the fault and the solution, and then fed back to the control module of the intelligent driving system to achieve automated repair.
[0072] In this method, vehicles using intelligent driving technology can transmit vehicle data to a cloud platform device. The cloud platform device then first judges the vehicle data based on fault rules to predict the possible coarse fault types. Next, it uses these coarse fault types to filter the vehicle data, obtaining fault data that may cause faults. Finally, based on a fault diagnosis model, it predicts the final detailed fault type, thereby effectively improving the fault detection rate of the intelligent driving system and further enhancing the safety and reliability of intelligent driving. Furthermore, because the data is filtered based on the coarse fault types before using the fault diagnosis model, the data processing load of the fault diagnosis model can be reduced, improving data processing speed and further enhancing the efficiency of fault monitoring.
[0073] This embodiment provides a fault monitoring method applicable to equipment. In this method, the fault diagnosis model can be determined in the following way:
[0074] S210. Obtain a training sample set; wherein, the training sample set includes multiple training sample pairs, each training sample pair includes an input sample and an output sample, the output sample includes a fault diagnosis information sample, the input sample includes a target feature data sample, and the target feature data sample is obtained by data processing of the fault data sample corresponding to the fault diagnosis information sample;
[0075] S220. Train the original model based on the training sample set to determine the fault diagnosis model.
[0076] In step S210, the data acquisition process was optimized during the determination of the training sample set, increasing the diversity and coverage of the data to better reflect the fault conditions of different vehicles. In this application, the size of the training dataset was expanded, ensuring that it includes vehicle data of different types and brands to better cover different fault types.
[0077] Furthermore, in constructing the training sample set, in addition to labeled data for normal and abnormal states, more detailed labeled data, such as fault types and their severity, were introduced to improve the model's ability to identify and judge faults. Moreover, time-series data was used as training input to capture the dynamic changes and trends in vehicle states. Understandably, richer and more diverse training data can improve the model's generalization ability and accuracy.
[0078] In constructing training sample pairs, the fault data sample corresponding to the fault diagnosis information sample can be determined first. Then, data processing, such as setting processing and anomaly detection, can be performed on the fault data sample to obtain the target feature data sample. The target feature data sample can then be used as the input sample, and the fault diagnosis information sample can be used as the corresponding output sample, thus completing the construction of a training sample pair. Based on the same method, multiple training sample pairs can be constructed to complete the construction of the training sample set.
[0079] In step S220, the model architecture of the original model can be incorporating an attention mechanism to enhance the model's focus on key features and weight adjustment, thereby better capturing abnormal patterns and fault signals.
[0080] During training, a larger training sample set can be used, along with a more optimized optimizer and learning rate scheduling strategy. Regularization techniques can be incorporated to prevent overfitting, and cross-validation and model selection can be performed to improve the model's efficiency and accuracy. Additionally, adaptive learning rate scheduling strategies, such as learning rate decay and dynamic learning rates, can be employed to train the original model, thereby optimizing the training process and convergence speed.
[0081] In this method, fault monitoring technology based on big data can quickly detect and diagnose faults in intelligent driving systems, thereby improving the reliability and safety of the system.
[0082] This method employs machine learning, deep learning, and data mining techniques to analyze and mine complex data, enabling the identification of fault characteristics and patterns, and improving the accuracy of fault monitoring and diagnosis. This method can effectively improve the fault detection rate and reliability of intelligent driving systems, and has broad application prospects and market value.
[0083] This embodiment provides a fault monitoring device, which can be used in equipment, such as a cloud-based big data platform device. The cloud-based big data platform device can perform distributed storage and efficient processing of big data. This device can be used to implement the aforementioned fault monitoring method. (Reference) Figure 2 As shown, for example, the device may include a receiving module 10 and a data processing module 20, wherein, during the implementation of the above method,
[0084] The receiving module 10 is used to receive the vehicle setting data sent by the vehicle;
[0085] The data processing module 20 is used to determine the target coarse fault classification type corresponding to the vehicle data based on the fault rules satisfied by the vehicle data and the fault rule configuration information; wherein, the fault rule configuration information includes the correspondence between multiple fault rules and fault coarse classification types.
[0086] It is also used to select fault data from vehicle data that corresponds to the target fault coarse classification type;
[0087] It is also used to process fault data and determine target characteristic data;
[0088] It is also used to input target feature data into a fault diagnosis model to predict fault diagnosis information corresponding to the target feature data; wherein, the fault diagnosis information includes fault sub-types, and the fault coarse-types include at least one fault sub-type.
[0089] This embodiment provides a fault monitoring device that can be used in equipment. In this device, the vehicle data is time-series data, and the vehicle data includes at least one of the following:
[0090] Sensor data collected by the vehicle's sensors, vehicle driving data, and video data collected by the vehicle's camera unit.
[0091] This embodiment provides a fault monitoring device that can be used in equipment. (Reference) Figure 2 As shown, in this device, the data processing module 20 can be used for:
[0092] The fault data is processed to determine the initial characteristic data; the processing includes data cleaning, feature extraction and data clustering.
[0093] Anomaly detection is performed on the initial diagnostic data, and data with anomalies in the initial feature data are identified as target feature data.
[0094] This embodiment provides a fault monitoring device that can be used in equipment. In this device, the fault diagnosis model is determined in the following way:
[0095] Obtain a training sample set; wherein, the training sample set includes multiple training sample pairs, each training sample pair includes an input sample and an output sample, the output sample includes a fault diagnosis information sample, the input sample includes a target feature data sample, and the target feature data sample is obtained by data processing of the fault data sample corresponding to the fault diagnosis information sample;
[0096] The original model is trained based on the training sample set to determine the fault diagnosis model.
[0097] This embodiment provides a fault monitoring device that can be used in equipment. In this device, an original model is trained based on the training sample set to determine the fault diagnosis model, including:
[0098] Based on the training sample set, an adaptive learning rate scheduling strategy is used to train the original model in order to determine the fault diagnosis model.
[0099] This embodiment provides a fault monitoring device that can be used in equipment. In this device, the original model includes an attention mechanism.
[0100] This embodiment provides a fault monitoring device that can be used in equipment. In this device, fault diagnosis information also includes the fault severity corresponding to the fault sub-type.
[0101] This embodiment provides a device. This device can be used as a cloud-based big data platform, and its specific configuration is not limited.
[0102] refer to Figure 3 As shown, the device 100 includes at least one processor 101, a memory 102, at least one network interface 104, and other user interfaces 103. The various components in the device 100 are coupled together via a bus system 105. It is understood that the bus system 105 is used to implement communication between these components. In addition to a data bus, the bus system 105 also includes a power bus, a control bus, and a status signal bus. However, for clarity, all buses are referred to as bus system 105.
[0103] The user interface 103 may include a display, keyboard, or clicking device (e.g., mouse, trackball, touchpad, or touchscreen).
[0104] It is understood that the memory 102 in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), and Direct Rambus RAM (DRRAM). The memory 102 described herein is intended to include, but is not limited to, these and any other suitable types of memory.
[0105] In some implementations, memory 102 stores elements, executable units or data structures, or subsets thereof, or extended sets thereof: operating system 1021 and application program 1022.
[0106] The operating system 1021 includes various system programs, such as a framework layer, a core library layer, and a driver layer, used to implement various basic business functions and handle hardware-based tasks. The application program 1022 includes various applications, such as a media player and a browser, used to implement various application functions. Programs implementing the methods of this application embodiment can be included in the application program 1022.
[0107] In this embodiment of the application, the processor 101 executes the methods provided in each method embodiment by calling the program or instructions stored in the memory 102, specifically the program or instructions stored in the application program 1022.
[0108] The methods disclosed in the embodiments of this application can be applied to or implemented by the processor 101. The processor 101 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware or by instructions in the form of software in the processor 101. The processor 101 may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software units in the decoding processor. The software units may be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 102. Processor 101 reads the information in memory 102 and performs the above method in conjunction with its hardware.
[0109] It is understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit can be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers, microprocessors, other electronic units for performing the functions described herein, or combinations thereof.
[0110] For software implementation, the techniques described herein can be implemented by units that perform the functions described herein. The software code can be stored in memory and executed by a processor. The memory can be implemented in the processor or external to the processor.
[0111] This application also provides a storage medium (computer-readable storage medium). This storage medium stores one or more programs. The storage medium may include volatile memory, such as random access memory; it may also include non-volatile memory, such as read-only memory, flash memory, hard disk, or solid-state drive; and it may also include combinations of the above types of memory.
[0112] When one or more programs in the storage medium can be executed by one or more processors to implement the above-described method of execution on the device.
[0113] The processor is used to execute the device control program stored in the memory to implement the above-described method of execution on the device.
[0114] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0115] It should be noted that the terms "one implementation," "embodiment," "exemplary embodiment," and "some embodiments" used in the specification indicate that the described embodiment may include a specific feature, structure, or characteristic, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, such phrases do not necessarily refer to the same embodiment. Moreover, when a specific feature, structure, or characteristic is described in connection with an embodiment, implementing such a feature, structure, or characteristic in conjunction with other embodiments, whether explicitly described or not, is within the knowledge scope of those skilled in the art.
[0116] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0117] The above embodiments are merely preferred embodiments provided to fully illustrate this application, and the scope of protection of this application is not limited thereto. Equivalent substitutions or modifications made by those skilled in the art based on this application are all within the scope of protection of this application.
Claims
1. A fault monitoring method for intelligent driving, characterized in that, The fault monitoring method includes: Receive vehicle data sent by the vehicle; Based on the fault rules satisfied by the vehicle data and the fault rule configuration information, the target fault coarse classification type corresponding to the vehicle data is determined; wherein, the fault rule configuration information includes the correspondence between multiple fault rules and fault coarse classification types; Select fault data corresponding to the target fault coarse classification from the vehicle data; The fault data is processed to determine the target feature data; The target feature data is input into the fault diagnosis model to predict the fault diagnosis information corresponding to the target feature data; wherein, the fault diagnosis information includes fault sub-types and their corresponding fault degrees, and the fault coarse-classification includes at least one fault sub-type; the fault diagnosis model uses a support vector machine or artificial neural network model to predict the fault type, and predicts the fault degree based on the analysis of the trend, periodicity, and seasonality of time series data; The process of processing the fault data to determine target feature data includes: The fault data is subjected to a setting process to determine initial feature data; wherein, the setting process includes data cleaning, feature extraction, and data clustering; Anomaly detection is performed on the initial feature data, and the data containing anomalies in the initial feature data are identified as target feature data.
2. The fault monitoring method according to claim 1, characterized in that, The vehicle data is time-series data, and the vehicle data includes at least one of the following: The sensor data collected by the vehicle's sensors, the vehicle's driving data, and the camera data collected by the vehicle's camera unit.
3. The fault monitoring method according to claim 1, characterized in that, The fault diagnosis model is determined in the following way: Obtain a training sample set; wherein the training sample set includes multiple training sample pairs, each training sample pair includes an input sample and an output sample, the output sample includes a fault diagnosis information sample, the input sample includes a target feature data sample, and the target feature data sample is obtained by processing the fault data sample corresponding to the fault diagnosis information sample. The original model is trained based on the training sample set to determine the fault diagnosis model.
4. The fault monitoring method according to claim 3, characterized in that, The step of training the original model based on the training sample set to determine the fault diagnosis model includes: Based on the training sample set, the original model is trained using an adaptive learning rate scheduling strategy to determine the fault diagnosis model.
5. The fault monitoring method according to claim 3, characterized in that, The original model includes an attention mechanism.
6. The fault monitoring method according to any one of claims 1-5, characterized in that, The fault diagnosis information also includes the fault severity corresponding to the fault sub-type.
7. A fault monitoring device, characterized in that, The fault monitoring device includes: The receiving module is used to receive the vehicle data sent by the vehicle. The data processing module is used to determine the target coarse fault classification type corresponding to the vehicle data based on the fault rules satisfied by the vehicle data and the fault rule configuration information; wherein, the fault rule configuration information includes the correspondence between multiple fault rules and fault coarse classification types; It is also used to select fault data from the vehicle data that corresponds to the target fault coarse classification type; It is also used to process the fault data to determine target feature data; It is also used to input the target feature data into a fault diagnosis model to predict the fault diagnosis information corresponding to the target feature data; wherein, the fault diagnosis information includes fault sub-types and their corresponding fault degrees, and the fault coarse-classification includes at least one fault sub-type; the fault diagnosis model uses a support vector machine or artificial neural network model to predict the fault type, and predicts the fault degree based on the analysis of the trend, periodicity, and seasonality of time series data; The process of processing the fault data to determine target feature data includes: The fault data is subjected to a setting process to determine initial feature data; wherein, the setting process includes data cleaning, feature extraction, and data clustering; Anomaly detection is performed on the initial feature data, and the data containing anomalies in the initial feature data are identified as target feature data.
8. A device, characterized in that, include: A processor and a memory, the processor being configured to execute a control program stored in the memory to implement the fault monitoring method according to any one of claims 1-6.
9. A storage medium, characterized in that, The storage medium stores one or more programs, which can be executed by one or more processors to implement the fault monitoring method according to any one of claims 1-6.