Vehicle failure detection method, vehicle, storage medium, and program product

Abstract

The embodiment of the application provides a vehicle fault detection method, a vehicle, a storage medium and a program product. The method comprises the following steps: acquiring sensing parameters sensed by a plurality of sensors in a current driving environment of a vehicle; performing feature extraction on the sensing parameters to obtain sensing features of the vehicle; determining information entropy of the sensing features based on a probability distribution of the sensing features, wherein the information entropy is used to represent uncertainty of the sensing features; and performing fault detection on the vehicle based on the information entropy to obtain a fault detection result, wherein the fault detection result is used to represent whether the vehicle currently has a fault. The application solves the technical problem of low accuracy in detecting potential faults in the vehicle in the related art.

Description

Technical Field

[0001] This application relates to the field of fault detection, and more specifically, to a vehicle fault detection method, a vehicle, a storage medium, and a program product. Background Technology

[0002] In the era of intelligent connected vehicles, real-time monitoring of vehicle health status and early warning of potential faults are key to ensuring vehicle driving safety and reducing vehicle maintenance costs. Although current fault detection technology can monitor vehicle operating status to a certain extent and detect faults in a timely manner, the faults detected so far are all obvious ones. However, it cannot accurately detect potential faults or less obvious faults in the vehicle.

[0003] There is currently no good solution to the above problems. Summary of the Invention

[0004] This application provides a vehicle fault detection method, a vehicle, a storage medium, and a program product to at least solve the technical problem of low accuracy in detecting potential faults in vehicles in related technologies.

[0005] According to one aspect of the embodiments of this application, a vehicle fault detection method is provided, comprising: acquiring perception parameters sensed by multiple sensors in the current driving environment of the vehicle; extracting features from the perception parameters to obtain perception features of the vehicle; determining the information entropy of the perception features based on the probability distribution of the perception features, wherein the information entropy is used to characterize the uncertainty of the perception features; and performing fault detection on the vehicle based on the information entropy to obtain a fault detection result, wherein the fault detection result is used to characterize whether the vehicle currently has a fault.

[0006] Furthermore, vehicle fault detection is performed based on information entropy to obtain fault detection results, including: dividing the perceived features based on information entropy to obtain a first feature and a second feature, wherein the information entropy corresponding to the first feature is greater than a preset threshold, and the information entropy corresponding to the second feature is less than or equal to the preset threshold; inputting the first feature into a first fault detection model, and using the first fault detection model to detect faults in the vehicle to obtain a first detection result; inputting the second feature into a second fault detection model, and using the second fault detection model to detect faults in the vehicle to obtain a second detection result, wherein the model framework of the second fault detection model is different from that of the first fault detection model; and constructing a fault detection result based on the first and second detection results.

[0007] Further, the first feature is input into the first fault detection model, and the first fault detection model is used to detect faults in the vehicle to obtain a first detection result, including: extracting features from the first feature to obtain fault feature variables of the first feature, wherein the fault feature variables are used to reflect the fault state of the vehicle equipment corresponding to the first feature; performing correlation analysis on any two fault feature variables to determine the correlation analysis result, wherein the correlation analysis result is used to characterize the probability that any two fault feature variables occur simultaneously; generating a feature association strength matrix corresponding to any two fault feature variables based on the correlation analysis result; and generating the first detection result based on the fault feature variables corresponding to the first feature, the feature association strength matrix, and the external confidence score, wherein the external confidence score is used to characterize the confidence score of the target sensor corresponding to the first feature, and the target sensor is used to perceive the sensing parameters of the first feature.

[0008] Further, based on the first detection result and the second detection result, a fault detection result is constructed, including: evaluating the accuracy of the first detection result and the second detection result respectively based on a preset strategy to obtain a first evaluation result and a second evaluation result; generating a first weight corresponding to the first detection result and a second weight corresponding to the second detection result based on the first evaluation result and the second evaluation result; and performing weighted processing on the first detection result and the second detection result based on the first weight and the second weight to obtain the fault detection result.

[0009] Furthermore, the multiple sensors include: on-board sensors on the vehicle and roadside units on the road where the vehicle is currently traveling; the sensing parameters include: a first driving parameter sensed by the on-board sensors and a second driving parameter sensed by the roadside units; feature extraction is performed on the sensing parameters to obtain the vehicle's sensing features, including: feature extraction of the first driving parameter to obtain a first state feature of the vehicle, wherein the first state feature is used to reflect the operating state of the vehicle equipment that generates the first driving parameter; feature extraction of the second driving parameter to obtain a second state feature of the vehicle, wherein the second state feature is used to reflect the environmental state of the environment in which the vehicle is currently located and the vehicle's driving behavior; and the sensing features are constructed based on the first and second features.

[0010] Furthermore, the above method also includes: responding to the fault detection result indicating that a fault currently exists in the vehicle, inputting the fault detection result into a digital twin vehicle, simulating the fault occurring in the vehicle using the digital twin vehicle to obtain fault parameters, wherein the digital twin vehicle is used to simulate the vehicle's driving state in a virtual environment; generating fault prompt information and fault resolution measures based on the fault parameters, wherein the fault prompt information is used to remind the driver of the type of fault currently existing in the vehicle, and the fault resolution measures are used to handle the fault; outputting the fault prompt information, and controlling the vehicle operation based on the fault resolution measures.

[0011] Furthermore, the above method also includes: obtaining the vehicle's structural configuration parameters and equipment operation rules, as well as the vehicle's historical driving parameters, wherein the structural configuration parameters are used to characterize the equipment parameters of different devices on the vehicle, and the equipment operation rules are used to characterize the operation rules between different devices on the vehicle; determining the behavioral logic generated by the driver when driving the vehicle based on the historical driving parameters; constructing an initial twin vehicle based on the structural configuration parameters and equipment operation rules; and adjusting the initial twin vehicle based on the behavioral logic to obtain a digital twin vehicle.

[0012] According to one aspect of the embodiments of this application, a vehicle fault detection device is also provided, comprising: a perception parameter acquisition module, used to acquire perception parameters sensed by multiple sensors in the current driving environment of the vehicle; a perception feature extraction module, used to extract features from the perception parameters to obtain perception features of the vehicle; an information entropy determination module, used to determine the information entropy of the perception features based on the probability distribution of the perception features, wherein the information entropy is used to characterize the uncertainty of the perception features; and a fault detection module, used to perform fault detection on the vehicle based on the information entropy to obtain a fault detection result, wherein the fault detection result is used to characterize whether the vehicle currently has a fault.

[0013] Furthermore, the fault detection module is also used to: divide the perceived features based on information entropy to obtain a first feature and a second feature, wherein the information entropy corresponding to the first feature is greater than a preset threshold, and the information entropy corresponding to the second feature is less than or equal to the preset threshold; input the first feature into a first fault detection model, and use the first fault detection model to perform fault detection on the vehicle to obtain a first detection result; input the second feature into a second fault detection model, and use the second fault detection model to perform fault detection on the vehicle to obtain a second detection result, wherein the model framework of the second fault detection model is different from the model framework of the first fault detection model; and construct a fault detection result based on the first detection result and the second detection result.

[0014] Furthermore, the fault detection module is also used to: extract features from the first feature to obtain fault feature variables of the first feature, wherein the fault feature variables are used to reflect the fault state of the vehicle equipment corresponding to the first feature; perform correlation analysis on any two fault feature variables to determine the correlation analysis result, wherein the correlation analysis result is used to characterize the probability that any two fault feature variables occur simultaneously; generate a feature association strength matrix corresponding to any two fault feature variables based on the correlation analysis result; and generate a first detection result based on the fault feature variables corresponding to the first feature, the feature association strength matrix, and the external confidence score, wherein the external confidence score is used to characterize the confidence score of the target sensor corresponding to the first feature, and the target sensor is used to perceive the sensing parameters at which the first feature is located.

[0015] Furthermore, the fault detection module is also used to: evaluate the accuracy of the first detection result and the second detection result based on a preset strategy to obtain a first evaluation result and a second evaluation result; generate a first weight corresponding to the first detection result and a second weight corresponding to the second detection result based on the first evaluation result and the second evaluation result; and perform weighted processing on the first detection result and the second detection result based on the first weight and the second weight to obtain a fault detection result.

[0016] Furthermore, the multiple sensors include: on-board sensors on the vehicle and roadside units on the road where the vehicle is currently traveling; the sensing parameters include: a first driving parameter sensed by the on-board sensors and a second driving parameter sensed by the roadside units; the sensing feature extraction module is also used to: extract features from the first driving parameter to obtain a first state feature of the vehicle, wherein the first state feature is used to reflect the operating state of the vehicle equipment that generates the first driving parameter; extract features from the second driving parameter to obtain a second state feature of the vehicle, wherein the second state feature is used to reflect the environmental state of the environment in which the vehicle is currently located and the driving behavior of the vehicle; and construct sensing features based on the first and second features.

[0017] Furthermore, the aforementioned device also includes: a fault simulation module, used to input the fault detection result into a digital twin vehicle in response to the fault detection result indicating that a fault currently exists in the vehicle, and to simulate the fault occurring in the vehicle using the digital twin vehicle to obtain fault parameters, wherein the digital twin vehicle is used to simulate the driving state of the vehicle in a virtual environment; a measure generation module, used to generate fault prompt information and fault resolution measures based on the fault parameters, wherein the fault prompt information is used to remind the driver of the type of fault currently existing in the vehicle, and the fault resolution measures are used to handle the fault; and a measure execution module, used to output the fault prompt information and control the vehicle operation based on the fault resolution measures.

[0018] Furthermore, the aforementioned device also includes: a rule acquisition module, used to acquire the vehicle's structural configuration parameters and equipment operation rules, as well as the vehicle's historical driving parameters, wherein the structural configuration parameters are used to characterize the equipment parameters of different devices on the vehicle, and the equipment operation rules are used to characterize the operation rules between different devices on the vehicle; a behavior logic determination module, used to determine the behavior logic generated by the driver when driving the vehicle based on the historical driving parameters; a twin system construction module, used to construct an initial twin vehicle based on the structural configuration parameters and equipment operation rules; and a twin system adjustment module, used to adjust the initial twin vehicle based on the behavior logic to obtain a digital twin vehicle.

[0019] According to another aspect of the embodiments of this application, a vehicle is also provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of this application when it runs.

[0020] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0021] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0022] According to another aspect of the embodiments of this application, a computer program product is also provided, including a non-volatile computer-readable storage medium storing a computer program, which, when executed by a processor, implements the methods in various embodiments of this application.

[0023] According to another aspect of the embodiments of this application, a computer program is also provided, which, when executed by a processor, implements the methods of the various embodiments of this application.

[0024] In this embodiment, the system acquires perception parameters from multiple sensors in the vehicle's current driving environment; extracts features from the perception parameters to obtain the vehicle's perception features; determines the information entropy of the perception features based on their probability distribution; and performs fault detection on the vehicle based on the information entropy to obtain the fault detection result. By utilizing the information entropy of the perception features of the perception parameters to reflect the uncertainty of the perception features, the fault detection system can determine whether there are potential faults in the corresponding vehicle equipment that have not yet occurred, or faults that have occurred but are not obvious, based on the uncertainty of the perception features. This greatly improves the accuracy of detecting potential faults in the vehicle and solves the technical problem of low accuracy in detecting potential faults in vehicles in related technologies. Attached Figure Description

[0025] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0026] Figure 1 This is a flowchart illustrating a vehicle fault detection method according to an embodiment of this application;

[0027] Figure 2This is a structural block diagram of a vehicle fault detection device according to an embodiment of this application. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] According to an embodiment of this application, a method for vehicle fault detection is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0031] This embodiment provides a vehicle fault detection method. Figure 1 This is a flowchart illustrating a vehicle fault detection method according to an embodiment of this application, such as... Figure 1 As shown, the process includes the following steps:

[0032] Step S102: Obtain the perception parameters sensed by multiple sensors in the current driving environment of the vehicle.

[0033] The aforementioned sensors can refer to sensors deployed in the vehicle's current driving environment that sense the vehicle's driving status and obtain the aforementioned sensing parameters. These sensors can be deployed on the vehicle itself or on other external devices that communicate with the vehicle. For example, sensors deployed on the vehicle may include, but are not limited to, temperature sensors, pressure sensors, and spectral sensors, and the corresponding sensing parameters may include, but are not limited to, engine temperature, motor temperature, tire pressure, brake system pressure, and the spectrum of the exhaust system or coolant. Sensors deployed on other external devices may include, but are not limited to, lidar, spectral analyzers, and road condition detectors, and the corresponding sensing parameters may include, but are not limited to, the distance between the vehicle and its surroundings, the spectrum of vehicle emissions, road surface humidity, temperature, and degree of damage.

[0034] In one optional solution of this embodiment, considering that traditional methods of vehicle fault detection typically rely on preset data thresholds and historical fault data to monitor and evaluate the data currently generated by different devices on the vehicle to determine whether the corresponding devices have malfunctioned, the threshold settings are usually static and difficult to adapt to changing vehicle driving environments or device operating environments. During fault detection, inappropriate threshold settings may lead to misjudgments or missed detections, resulting in inaccurate fault detection. Furthermore, this method typically detects faults that have become apparent and reached a certain severity, but cannot effectively detect less obvious or potential faults on the vehicle. Considering that data entropy can typically be used to measure the degree of uncertainty in data to capture irregular, random, or complex changes, thereby reflecting whether data anomalies have occurred, using information entropy to detect data anomalies and corresponding device malfunctions can reduce the risk of misjudgments or missed detections.

[0035] Therefore, to improve the accuracy of detecting potential faults in vehicles, the fault detection system can introduce information entropy of sensing parameters related to the vehicle's driving state and use this information entropy to detect faults, determining whether a potential fault exists in the vehicle. This reduces the impact of changes in the vehicle's driving environment and equipment operating environment on the fault detection process. Based on this, during fault detection, the system can first acquire sensing parameters that reflect the vehicle's driving state. To ensure the accuracy and completeness of the acquired sensing parameters, the system can first identify multiple sensors deployed in the vehicle's current environment, and then use these sensors to perceive the vehicle's driving state to obtain the aforementioned sensing parameters.

[0036] Step S104: Extract features from the perception parameters to obtain the vehicle's perception features.

[0037] In one optional embodiment, considering that the sensing parameters detected by the sensors may contain a large amount of redundant information and noise, and the information truly useful for fault detection may only occupy a small portion of the sensing parameters, directly calculating the information entropy of the sensing parameters may result in this useful information being ignored or covered by redundant information and noise, leading to low accuracy or even invalidity of the calculated information entropy. Therefore, in order to accurately determine the information entropy of the sensing parameters, the fault detection system can first extract features from the sensing parameters to obtain the vehicle's sensing features before calculating the information entropy. Then, the information entropy is calculated based on these sensing features, thereby improving the targeting and effectiveness of the information entropy calculation and ensuring the accuracy and timeliness of fault warnings. For example, the fault detection system can use the extracted sensing features to filter out information useful for fault detection from the sensing parameters, and then use the information entropy of this information to detect vehicle faults.

[0038] Step S106: Determine the information entropy of the perceptual features based on the probability distribution of the perceptual features.

[0039] Information entropy is used to characterize the uncertainty of perceived features.

[0040] In one optional embodiment, considering that the information entropy of perception parameters reflects the uncertainty of vehicle state at the raw data level, while the information entropy of perception features reflects the uncertainty of vehicle health status and potential faults at a more granular data level, the information entropy of perception features focuses more on key changes in the perceived information, resulting in higher accuracy in vehicle fault detection based on the information entropy of perception features. Therefore, in addition to filtering out useful information for fault detection through perception features and then using the information entropy of this information for vehicle fault detection, the fault detection system can also determine the information entropy of the extracted perception features to use for vehicle fault detection. To accurately determine the information entropy of perception features, the fault detection system can first determine the probability distribution of the perception features, and then determine the information entropy of the perception features based on the probability distribution, using the information entropy to reflect the uncertainty of the perception features, thereby more accurately detecting vehicle faults.

[0041] Step S108: Perform fault detection on the vehicle based on information entropy to obtain fault detection results.

[0042] Among them, the fault detection results are used to characterize whether the vehicle currently has a fault.

[0043] In one optional embodiment, after determining the information entropy of the perceived features, the fault detection system can use the information entropy to perform fault detection on the vehicle to determine whether a fault exists in the vehicle, thus obtaining the aforementioned fault detection result. For example, an information entropy threshold based on historical fault data analysis can be set in the fault detection system. When the information entropy of a perceived feature exceeds this threshold, it can be considered that the uncertainty of the perceived feature has increased abnormally, and there may be a potential fault in the vehicle equipment corresponding to that perceived feature. Alternatively, the fault detection system can analyze the information entropy of multiple interrelated perceived features to construct a multidimensional information entropy matrix or network, and then identify potential faults in the vehicle by observing the correlation and change patterns of information entropy among different features.

[0044] In this embodiment, the system acquires perception parameters from multiple sensors in the vehicle's current driving environment; extracts features from the perception parameters to obtain the vehicle's perception features; determines the information entropy of the perception features based on their probability distribution; and performs fault detection on the vehicle based on the information entropy to obtain the fault detection result. By utilizing the information entropy of the perception features of the perception parameters to reflect the uncertainty of the perception features, the fault detection system can determine whether there are potential faults in the corresponding vehicle equipment that have not yet occurred, or faults that have occurred but are not obvious, based on the uncertainty of the perception features. This greatly improves the accuracy of detecting potential faults in the vehicle and solves the technical problem of low accuracy in detecting potential faults in vehicles in related technologies.

[0045] Furthermore, vehicle fault detection is performed based on information entropy to obtain fault detection results, including: dividing the perceived features based on information entropy to obtain a first feature and a second feature, wherein the information entropy corresponding to the first feature is greater than a preset threshold, and the information entropy corresponding to the second feature is less than or equal to the preset threshold; inputting the first feature into a first fault detection model, and using the first fault detection model to detect faults in the vehicle to obtain a first detection result; inputting the second feature into a second fault detection model, and using the second fault detection model to detect faults in the vehicle to obtain a second detection result, wherein the model framework of the second fault detection model is different from that of the first fault detection model; and constructing a fault detection result based on the first and second detection results.

[0046] In one optional embodiment, considering that the magnitude of the information entropy of the perceived features typically reflects the uncertainty of the perceived features, and that the uncertainty of the perceived features reflects their importance in the fault detection process, the greater the uncertainty of the perceived features, the higher the variability and complexity of the perceived features in the vehicle state, and the greater the importance of the perceived features in the fault detection process; conversely, the smaller the uncertainty of the perceived features, the higher the stability of the perceived features in the vehicle state, and the smaller the importance of the perceived features in the fault detection process. Based on this, to improve the accuracy and efficiency of vehicle fault detection using the information entropy of different perceived features, the fault detection system can employ different fault detection models to perform fault detection based on perceived features with different levels of information entropy. Then, the detection results output by different fault detection models are summarized to obtain the corresponding fault detection result. For example, a higher-precision fault detection model can be used to process perceived features with higher information entropy to improve the processing accuracy of these perceived features and the weight of faults detected using these perceived features in the entire fault detection process, thereby improving the accuracy of the summarized fault detection result. Meanwhile, by using a fault detection model with high processing efficiency, the perceptual features with low information entropy are processed to improve the efficiency of processing these perceptual features. This avoids the situation where the processing efficiency is low when using a single fault detection model to process a large number of perceptual features, thereby improving the efficiency of vehicle fault detection.

[0047] Based on this, in the process of using information entropy for vehicle fault detection, the fault detection system can first divide the perceived features using information entropy to obtain a first feature with higher information entropy and a second feature with lower information entropy. For example, the information entropy of the first feature can be greater than a preset threshold, and the information entropy of the second feature can be less than or equal to the preset threshold. After obtaining the first and second features, the fault detection system can input the first feature into a first fault detection model with higher detection accuracy to perform fault detection on the vehicle and obtain a first detection result. Simultaneously, the fault detection system can input the second feature into a second fault detection model with higher detection efficiency to perform fault detection on the vehicle and obtain a second detection result. Finally, the fault detection system can construct the aforementioned fault detection result based on the first and second detection results, thereby improving both the efficiency and accuracy of vehicle fault detection. Since the first fault detection model and the second fault detection model have different focuses, the model framework of the first fault detection model can be different from that of the second fault detection model. For example, the model framework of the first fault detection model can be a framework constructed using algorithms such as quantum annealing and quantum circuit simulation to ensure the accuracy of processing the first feature; the model framework of the second fault detection model can be a framework constructed using algorithms such as deep learning and transfer learning.

[0048] Further, the first feature is input into the first fault detection model, and the first fault detection model is used to detect faults in the vehicle to obtain a first detection result, including: extracting features from the first feature to obtain fault feature variables of the first feature, wherein the fault feature variables are used to reflect the fault state of the vehicle equipment corresponding to the first feature; performing correlation analysis on any two fault feature variables to determine the correlation analysis result, wherein the correlation analysis result is used to characterize the probability that any two fault feature variables occur simultaneously; generating a feature association strength matrix corresponding to any two fault feature variables based on the correlation analysis result; and generating the first detection result based on the fault feature variables corresponding to the first feature, the feature association strength matrix, and the external confidence score, wherein the external confidence score is used to characterize the confidence score of the target sensor corresponding to the first feature, and the target sensor is used to perceive the sensing parameters of the first feature.

[0049] In one optional embodiment, in order to accurately process the first feature, taking the quantum annealing algorithm to construct a model framework as an example, in the process of using the first fault detection model to process the first feature to detect vehicle faults, the fault detection system can first extract features from the first feature to obtain fault feature variables of the first feature, so as to use the fault feature variables to determine the fault state of the vehicle equipment corresponding to the first feature, such as whether the engine, braking system and other components can operate normally, and whether the temperature of the cooling system, the vibration frequency of the transmission system, the current intensity of the power system and other parameters are normal.

[0050] After extracting the fault feature variables, the fault detection system can perform correlation analysis on any two fault feature variables to determine the correlation analysis results, thereby determining the probability of any two fault feature variables occurring simultaneously. After determining the correlation analysis results, the fault detection system can use these results to generate a feature correlation strength matrix corresponding to any two fault feature variables. Finally, when performing fault detection on the vehicle based on the first feature, the fault detection system can generate the first fault detection result based on the fault feature variables corresponding to the first feature, the feature correlation strength matrix, and the external confidence score. The external confidence score can refer to the confidence score of the target sensor corresponding to the first feature, and the target sensor can be used to perceive the sensing parameters of the first feature.

[0051] For ease of understanding, the formula for the quantum annealing algorithm can be:

[0052] H(s)=-∑i <j,J ij ,S i ,S j -∑i,h i ,S i .

[0053] Where H(s) represents the fault characteristic variable S i The energy level of the combined states, h i J represents external confidence level. ij This represents the fault characteristic variable S. i and S j The corresponding feature correlation strength matrix.

[0054] Further, based on the first detection result and the second detection result, a fault detection result is constructed, including: evaluating the accuracy of the first detection result and the second detection result respectively based on a preset strategy to obtain a first evaluation result and a second evaluation result; generating a first weight corresponding to the first detection result and a second weight corresponding to the second detection result based on the first evaluation result and the second evaluation result; and performing weighted processing on the first detection result and the second detection result based on the first weight and the second weight to obtain the fault detection result.

[0055] In one optional embodiment, to accurately construct fault detection results, the fault detection system can weight the first detection result and the second detection result according to preset weights. To improve the matching degree between the determined fault detection results and the current driving environment of the vehicle and the current operating environment of the equipment, when weighting the first and second detection results, the fault detection system can first evaluate the accuracy of the first and second detection results according to a preset strategy, obtaining corresponding first and second evaluation results. Then, using the first and second evaluation results, it generates a first weight corresponding to the first detection result and a second weight corresponding to the second detection result. This allows the fault detection system to use the real-time evaluated first and second weights to weight the first and second detection results, obtaining a more reasonable fault detection result.

[0056] The aforementioned preset strategy for evaluating the accuracy of the first and second detection results may include, but is not limited to: using statistical models to analyze the probability distribution of the detection results to determine whether they meet the expected failure mode, or comparing the current detection results with the results of similar cases in historical data to evaluate their consistency and trend of change, thereby determining the accuracy of the results.

[0057] Furthermore, the multiple sensors include: on-board sensors on the vehicle and roadside units on the road where the vehicle is currently traveling; the sensing parameters include: a first driving parameter sensed by the on-board sensors and a second driving parameter sensed by the roadside units; feature extraction is performed on the sensing parameters to obtain the vehicle's sensing features, including: feature extraction of the first driving parameter to obtain a first state feature of the vehicle, wherein the first state feature is used to reflect the operating state of the vehicle equipment that generates the first driving parameter; feature extraction of the second driving parameter to obtain a second state feature of the vehicle, wherein the second state feature is used to reflect the environmental state of the environment in which the vehicle is currently located and the vehicle's driving behavior; and the sensing features are constructed based on the first and second features.

[0058] In one optional embodiment, to improve the comprehensiveness of the acquired perception parameters and thus ensure the comprehensiveness of the extracted perception features, the multiple sensors deployed in the vehicle's current driving environment may include: onboard sensors on the vehicle and roadside units on the road where the vehicle is currently traveling. The perception parameters may include: a first driving parameter perceived by the onboard sensors and a second driving parameter perceived by the roadside units. Correspondingly, when extracting features from the perception parameters, the fault detection system can first extract features from the first driving parameter to obtain the vehicle's first state features, which reflect the operating state of the vehicle equipment that generates the first driving parameter, such as engine operation, electrical system operation, and mechanical wear of the equipment. Simultaneously, features can be extracted from the second driving parameter to obtain the vehicle's second state features, which reflect the environmental state of the vehicle's current environment and the vehicle's driving behavior, such as the vehicle's driving route, speed changes, and relative position to surrounding obstacles. Finally, the fault detection system can construct the aforementioned perception features based on the first and second features, thereby ensuring the accuracy and rationality of the perception features.

[0059] Furthermore, the above method also includes: responding to the fault detection result indicating that a fault currently exists in the vehicle, inputting the fault detection result into a digital twin vehicle, simulating the fault occurring in the vehicle using the digital twin vehicle to obtain fault parameters, wherein the digital twin vehicle is used to simulate the vehicle's driving state in a virtual environment; generating fault prompt information and fault resolution measures based on the fault parameters, wherein the fault prompt information is used to remind the driver of the type of fault currently existing in the vehicle, and the fault resolution measures are used to handle the fault; outputting the fault prompt information, and controlling the vehicle operation based on the fault resolution measures.

[0060] In one optional embodiment, after obtaining the fault detection result, in order to more accurately locate the fault, the fault detection system can also input the fault detection result into the aforementioned digital twin vehicle if the fault detection result indicates that a fault exists in the vehicle. This allows the digital twin vehicle to simulate the fault in the vehicle, determining the vehicle's current driving state in a virtual environment, thereby obtaining fault parameters related to the fault, such as fault type and fault location. Based on these fault parameters, the fault detection system can also generate corresponding fault prompts and fault resolution measures, and control the vehicle's operation according to the fault resolution measures to handle the fault. Simultaneously, it outputs fault prompts to remind the user of the fault in the vehicle and the current fault resolution measures, thereby improving the user experience.

[0061] Furthermore, the above method also includes: obtaining the vehicle's structural configuration parameters and equipment operation rules, as well as the vehicle's historical driving parameters, wherein the structural configuration parameters are used to characterize the equipment parameters of different devices on the vehicle, and the equipment operation rules are used to characterize the operation rules between different devices on the vehicle; determining the behavioral logic generated by the driver when driving the vehicle based on the historical driving parameters; constructing an initial twin vehicle based on the structural configuration parameters and equipment operation rules; and adjusting the initial twin vehicle based on the behavioral logic to obtain a digital twin vehicle.

[0062] In one optional embodiment, to construct a digital twin vehicle that matches the current driving state of the vehicle, the fault detection system can first acquire the vehicle's structural configuration parameters and equipment operating rules. The structural configuration parameters are used to determine the equipment parameters of different devices on the vehicle, while the equipment's operating trajectory is used to determine the operating rules between these different devices. Then, the fault detection system can construct an initial digital twin vehicle based on the structural configuration parameters and equipment operating rules. Considering that the driver's historical driving parameters can reveal information such as the vehicle's usage habits and operating style, the fault detection system can also acquire the vehicle's historical driving parameters to determine the driver's behavioral logic while driving. Finally, the initial digital twin vehicle is adjusted based on this behavioral logic to obtain the aforementioned digital twin vehicle. This allows the fault detection system to better model the actual usage of the vehicle using the digital twin vehicle, thereby ensuring the accuracy of the fault parameters determined using the digital twin vehicle.

[0063] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0064] According to an embodiment of this application, a device for vehicle fault detection is provided. It should be noted that the device can be used to perform the above-described vehicle fault detection method. Figure 2 This is a structural block diagram of a vehicle fault detection device according to an embodiment of this application, such as... Figure 2 As shown, the device may include: a sensing parameter acquisition module 202, a sensing feature extraction module 204, an information entropy determination module 206, and a fault detection module 208.

[0065] The system includes: a perception parameter acquisition module 202 for acquiring perception parameters from multiple sensors in the current driving environment of the vehicle; a perception feature extraction module 204 for extracting features from the perception parameters to obtain the perception features of the vehicle; an information entropy determination module 206 for determining the information entropy of the perception features based on the probability distribution of the perception features, wherein the information entropy is used to characterize the uncertainty of the perception features; and a fault detection module 208 for performing fault detection on the vehicle based on the information entropy to obtain fault detection results, wherein the fault detection results are used to characterize whether the vehicle currently has a fault.

[0066] Furthermore, the fault detection module is also used to: divide the perceived features based on information entropy to obtain a first feature and a second feature, wherein the information entropy corresponding to the first feature is greater than a preset threshold, and the information entropy corresponding to the second feature is less than or equal to the preset threshold; input the first feature into a first fault detection model, and use the first fault detection model to perform fault detection on the vehicle to obtain a first detection result; input the second feature into a second fault detection model, and use the second fault detection model to perform fault detection on the vehicle to obtain a second detection result, wherein the model framework of the second fault detection model is different from the model framework of the first fault detection model; and construct a fault detection result based on the first detection result and the second detection result.

[0067] Furthermore, the fault detection module is also used to: extract features from the first feature to obtain fault feature variables of the first feature, wherein the fault feature variables are used to reflect the fault state of the vehicle equipment corresponding to the first feature; perform correlation analysis on any two fault feature variables to determine the correlation analysis result, wherein the correlation analysis result is used to characterize the probability that any two fault feature variables occur simultaneously; generate a feature association strength matrix corresponding to any two fault feature variables based on the correlation analysis result; and generate a first detection result based on the fault feature variables corresponding to the first feature, the feature association strength matrix, and the external confidence score, wherein the external confidence score is used to characterize the confidence score of the target sensor corresponding to the first feature, and the target sensor is used to perceive the sensing parameters at which the first feature is located.

[0068] Furthermore, the fault detection module is also used to: evaluate the accuracy of the first detection result and the second detection result based on a preset strategy to obtain a first evaluation result and a second evaluation result; generate a first weight corresponding to the first detection result and a second weight corresponding to the second detection result based on the first evaluation result and the second evaluation result; and perform weighted processing on the first detection result and the second detection result based on the first weight and the second weight to obtain a fault detection result.

[0069] Furthermore, the multiple sensors include: on-board sensors on the vehicle and roadside units on the road where the vehicle is currently traveling; the sensing parameters include: a first driving parameter sensed by the on-board sensors and a second driving parameter sensed by the roadside units; the sensing feature extraction module is also used to: extract features from the first driving parameter to obtain a first state feature of the vehicle, wherein the first state feature is used to reflect the operating state of the vehicle equipment that generates the first driving parameter; extract features from the second driving parameter to obtain a second state feature of the vehicle, wherein the second state feature is used to reflect the environmental state of the environment in which the vehicle is currently located and the driving behavior of the vehicle; and construct sensing features based on the first and second features.

[0070] Furthermore, the aforementioned device also includes: a fault simulation module, used to input the fault detection result into a digital twin vehicle in response to the fault detection result indicating that a fault currently exists in the vehicle, and to simulate the fault occurring in the vehicle using the digital twin vehicle to obtain fault parameters, wherein the digital twin vehicle is used to simulate the driving state of the vehicle in a virtual environment; a measure generation module, used to generate fault prompt information and fault resolution measures based on the fault parameters, wherein the fault prompt information is used to remind the driver of the type of fault currently existing in the vehicle, and the fault resolution measures are used to handle the fault; and a measure execution module, used to output the fault prompt information and control the vehicle operation based on the fault resolution measures.

[0071] Furthermore, the aforementioned device also includes: a rule acquisition module, used to acquire the vehicle's structural configuration parameters and equipment operation rules, as well as the vehicle's historical driving parameters, wherein the structural configuration parameters are used to characterize the equipment parameters of different devices on the vehicle, and the equipment operation rules are used to characterize the operation rules between different devices on the vehicle; a behavior logic determination module, used to determine the behavior logic generated by the driver when driving the vehicle based on the historical driving parameters; a twin system construction module, used to construct an initial twin vehicle based on the structural configuration parameters and equipment operation rules; and a twin system adjustment module, used to adjust the initial twin vehicle based on the behavior logic to obtain a digital twin vehicle.

[0072] Embodiments of this application also provide a vehicle, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods described in various embodiments of this application when it runs.

[0073] Embodiments of this application also provide a computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0074] Embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0075] Embodiments of this application also provide a computer program product, including a non-volatile computer-readable storage medium for storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0076] Embodiments of this application also provide a computer program that, when executed by a processor, implements the methods described in the various embodiments of this application.

[0077] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0078] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0079] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0080] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0081] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0082] The above are merely preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A vehicle fault detection method, characterized in that, include: Acquire the perception parameters sensed by multiple sensors in the vehicle's current driving environment; The perception parameters are used to extract features to obtain the perception features of the vehicle; Based on the probability distribution of the perceived features, the information entropy of the perceived features is determined, wherein the information entropy is used to characterize the uncertainty of the perceived features; Based on the information entropy, the vehicle is subjected to fault detection to obtain a fault detection result, wherein the fault detection result is used to characterize whether the vehicle currently has a fault; The perceptual features are divided based on the information entropy to obtain a first feature and a second feature, wherein the information entropy corresponding to the first feature is greater than a preset threshold, and the information entropy corresponding to the second feature is less than or equal to the preset threshold. The first feature is input into the first fault detection model, and the first fault detection model is used to detect faults in the vehicle to obtain a first detection result. The second feature is input into the second fault detection model, and the second fault detection model is used to detect faults in the vehicle to obtain a second detection result. The model framework of the second fault detection model is different from that of the first fault detection model. The fault detection result is constructed based on the first detection result and the second detection result.

2. The method according to claim 1, characterized in that, The first feature is input into a first fault detection model, and the first fault detection model is used to detect faults in the vehicle to obtain a first detection result, including: Feature extraction is performed on the first feature to obtain fault feature variables of the first feature, wherein the fault feature variables are used to reflect the fault state of the vehicle equipment corresponding to the first feature. A correlation analysis is performed on any two fault characteristic variables to determine the correlation analysis results, wherein the correlation analysis results are used to characterize the probability that the two fault characteristic variables occur simultaneously. Based on the correlation analysis results, a feature correlation strength matrix is ​​generated corresponding to any two fault feature variables. Based on the fault feature variable corresponding to the first feature, the feature association strength matrix and the external confidence score, the first detection result is generated, wherein the external confidence score is used to characterize the confidence score of the target sensor corresponding to the first feature, and the target sensor is used to perceive the sensing parameters at which the first feature is located.

3. The method according to claim 1, characterized in that, Based on the first detection result and the second detection result, the fault detection result is constructed, including: The accuracy of the first detection result and the second detection result are evaluated based on a preset strategy to obtain a first evaluation result and a second evaluation result. Based on the first evaluation result and the second evaluation result, a first weight corresponding to the first detection result and a second weight corresponding to the second detection result are generated; Based on the first weight and the second weight, the first detection result and the second detection result are weighted to obtain the fault detection result.

4. The method according to any one of claims 1 to 3, characterized in that, The plurality of sensors include: on-board sensors on the vehicle, and roadside units on the road where the vehicle is currently traveling; the sensing parameters include: a first driving parameter sensed by the on-board sensors, and a second driving parameter sensed by the roadside units; feature extraction is performed on the sensing parameters to obtain the vehicle's sensing features, including: Feature extraction is performed on the first driving parameter to obtain the first state feature of the vehicle, wherein the first state feature is used to reflect the operating state of the vehicle equipment that generates the first driving parameter. Feature extraction is performed on the second driving parameters to obtain the second state features of the vehicle, wherein the second state features are used to reflect the environmental state of the current environment in which the vehicle is located, as well as the driving behavior of the vehicle. The perception features are constructed based on the first state features and the second state features.

5. The method according to any one of claims 1 to 3, characterized in that, The method further includes: In response to the fault detection result indicating that the vehicle currently has a fault, the fault detection result is input into the digital twin vehicle, and the fault occurring in the vehicle is simulated using the digital twin vehicle to obtain fault parameters. The digital twin vehicle is used to simulate the driving state of the vehicle in a virtual environment. Based on the fault parameters, fault prompt information and fault resolution measures are generated, wherein the fault prompt information is used to remind the driver of the type of fault currently existing in the vehicle, and the fault resolution measures are used to handle the fault; The fault message is output, and the vehicle operation is controlled based on the fault resolution measures.

6. The method according to claim 5, characterized in that, The method further includes: The structural configuration parameters and equipment operation rules of the vehicle, as well as the historical driving parameters of the vehicle, are obtained. The structural configuration parameters are used to characterize the equipment parameters of different devices on the vehicle, and the equipment operation rules are used to characterize the operation rules between different devices on the vehicle. Based on the historical driving parameters, determine the behavioral logic generated by the driver when driving the vehicle; Based on the structural configuration parameters and the equipment operation rules, an initial twin vehicle is constructed; The initial twin vehicle is adjusted based on the behavioral logic to obtain the digital twin vehicle.

7. A vehicle, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the storage medium is located to perform the method according to any one of claims 1 to 6.

9. A computer program product, characterized in that, It includes a computer program, which, when executed, is used to implement the method described in any one of claims 1 to 6.

Images