Intelligent driving system reliability evaluation method based on EMC test

Abstract

The invention discloses an intelligent driving system reliability evaluation method based on an EMC test, and relates to the technical field of EMC tests, and the method comprises the steps: obtaining intelligent driving system software and hardware interaction request data under the condition that a target vehicle is subjected to the closed electromagnetic interference, evaluating software and hardware interaction states of the intelligent driving system of the target vehicle, performing prior distribution screening on the software and hardware interaction states and functional fault feedback data of the known intelligent driving system, and establishing a known fault chain development network of software and hardware components of the intelligent driving system of the target vehicle; according to the intelligent driving system soft and hard component fault chain development network of the target vehicle, analyzing posterior distribution of the target vehicle open road real-time intelligent driving system soft and hard component interaction request list, and establishing an intelligent driving system fault chain development trend evaluation model; and predicting an intelligent driving system soft and hard component function failure risk score of the target vehicle. The method has the advantage that the safety and robustness of the intelligent driving system of the EMC test are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of EMC testing, and in particular to a reliability assessment method for an intelligent driving system based on EMC testing. Background Art

[0002] Existing intelligent driving system reliability solutions based on EMC testing use static interaction data obtained in closed environments to accurately reflect the dynamic electromagnetic interference characteristics in real road environments, resulting in large deviations between test results and actual conditions. At the same time, existing methods lack the ability to accurately model multi-component coupled failures in complex electromagnetic environments. They can only identify single failure modes but have difficulty predicting the propagation path of the failure chain, resulting in serious lack of accuracy and reliability in the evaluation results. In addition, the disconnection between test data and functional safety assessments further reduces the effectiveness of risk warnings and fails to provide accurate reliability feedback for intelligent driving systems. Summary of the Invention

[0003] In order to solve the above technical problems, a reliability assessment method for intelligent driving systems based on EMC testing is provided. This technical solution solves the problem that the static interaction data obtained in the closed environment of the existing intelligent driving system reliability solution based on EMC testing cannot accurately reflect the dynamic electromagnetic interference characteristics in the real road environment, resulting in a large deviation between the test results and the actual situation; at the same time, the existing method lacks the ability to accurately model multi-component coupling failures in complex electromagnetic environments, and can only identify a single failure mode but has difficulty in predicting the fault chain propagation path, which makes the accuracy and reliability of the assessment results seriously insufficient; in addition, the disconnection between test data and functional safety assessment further reduces the effectiveness of risk warning and cannot provide accurate reliability feedback for the intelligent driving system.

[0004] In order to achieve the above objects, the technical solution adopted by the present invention is: A reliability assessment method for an intelligent driving system based on EMC testing, including: S1. Based on the closed-environment EMC test of the target vehicle, obtain the target vehicle's intelligent driving system hardware and software interaction request data under closed electromagnetic interference, and evaluate the target vehicle's intelligent driving system hardware and software interaction status; S2. Based on the internet and the improvement plans of automobile companies, functional fault feedback data of known intelligent driving systems is obtained and the interaction status of the target vehicle's intelligent driving system hardware and software components is screened by prior distribution to build a development network of known fault chains of the target vehicle's intelligent driving system hardware and software components. S3. Based on the open environment EMC test of the target vehicle, analyze the posterior distribution of the target vehicle's open road real-time intelligent driving system software and hardware component interaction request list according to the target vehicle's intelligent driving system software and hardware component failure chain development network, and determine the target vehicle's intelligent driving system software and hardware component failure chain development propagation path; S4. Based on the development and propagation path of the fault chain of the target vehicle's intelligent driving system's hardware and software components, establish an intelligent driving system fault chain development trend assessment model to predict the target vehicle's intelligent driving system's hardware and software components' functional failure risk score.

[0005] Preferably, the target vehicle is subjected to several rounds of EMC testing in a closed anechoic chamber to obtain software and hardware interaction request data of the intelligent driving system under closed electromagnetic interference of the target vehicle; According to hardware interaction synchronization and hardware interaction asynchrony, the software and hardware interaction request data of the intelligent driving system under closed electromagnetic interference of the target vehicle are divided to obtain the software and hardware interaction synchronous request data and the software and hardware interaction asynchronous request data of the intelligent driving system under closed electromagnetic interference of the target vehicle; Using linear interpolation, asynchronously request data completion for the intelligent driving system's hardware and software interactions under closed electromagnetic interference of the target vehicle; Using the intelligent driving system hardware and software interaction transmission protocol, the target vehicle's closed electromagnetic interference intelligent driving system hardware and software interaction synchronous request data and the intelligent driving system hardware and software interaction asynchronous request data are temporally and spatially aligned to obtain the target vehicle's closed electromagnetic interference intelligent driving system hardware and software interaction synchronous and asynchronous request spatiotemporal alignment data. Taking the time step of the synchronous-asynchronous request of the intelligent driving system's software and hardware interaction as the observation window, and the synchronous-asynchronous request interaction data of the intelligent driving system's software and hardware interaction as the observation object, the synchronous-asynchronous request timing data of the intelligent driving system's software and hardware interaction under closed electromagnetic interference of the target vehicle is obtained.

[0006] Preferably, based on ARIMA autoregressive integrated moving average, a synchronous-asynchronous request deviation analysis model for the intelligent driving system software and hardware interaction is established; Using partial autocorrelation functions and autocorrelation functions, the synchronous-asynchronous request timing data of the intelligent driving system's hardware and software interactions under closed electromagnetic interference of the target vehicle is verified, and the autoregressive order and moving average order are determined; Based on the intelligent driving system software and hardware interaction synchronous-asynchronous request deviation analysis model, the autoregressive order and moving average order are used as training constraints, the target vehicle's closed electromagnetic interference intelligent driving system software and hardware interaction synchronous-asynchronous request time series data observation value is used as input, and the target vehicle's closed electromagnetic interference intelligent driving system software and hardware interaction synchronous-asynchronous request time series data prediction value is used as output; Calculate the difference between the observed value and the predicted value of the synchronous-asynchronous request time series data of the intelligent driving system software and hardware interaction under the closed electromagnetic interference of the target vehicle, and establish the residual sequence of the synchronous-asynchronous request time series data of the intelligent driving system software and hardware interaction under the closed electromagnetic interference of the target vehicle; Determine the standardized timing data of the synchronous-asynchronous requests of the target vehicle intelligent driving system's software and hardware interaction and substitute it into the intelligent driving system's software and hardware interaction synchronous-asynchronous request deviation analysis model to obtain the residual sequence of the standardized timing data of the target vehicle intelligent driving system's software and hardware interaction synchronous-asynchronous requests.

[0007] Preferably, according to the three-standard deviation threshold method, it is verified whether the standard deviation of the residual sequence of the timing data of the synchronization-asynchronous request of the intelligent driving system software and hardware interaction under the closed electromagnetic interference of the target vehicle is within the range of three times the standard deviation of the residual sequence of the standardized timing data of the synchronization-asynchronous request of the intelligent driving system software and hardware interaction of the target vehicle; if so, it is marked as a normal driving software and hardware interaction synchronization-asynchronous request; if not, it is marked as an abnormal driving software and hardware interaction synchronization-asynchronous request; Determine the error coding types of synchronous-asynchronous requests for normal driving software-hardware interaction and synchronous-asynchronous requests for abnormal driving software-hardware interaction; Using one-hot encoding, we map and transform the error coding types of synchronous-asynchronous requests for normal driving software-hardware interaction and those for abnormal driving software-hardware interaction. This yields the error coding type vectors for synchronous-asynchronous requests for normal driving software-hardware interaction and those for abnormal driving software-hardware interaction of the intelligent driving system under closed electromagnetic interference. Normalize the error coding type vectors of the synchronization-asynchronous request for normal driving software and hardware interaction and the error coding type vectors of abnormal driving software and hardware interaction of the intelligent driving system under closed electromagnetic interference of the target vehicle; Based on DBSCAN density clustering, the Mahalanobis distance between the normal driving software and hardware interaction synchronous-asynchronous request error coding type vector and the abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector of the intelligent driving system under closed electromagnetic interference of the target vehicle is used as the domain radius. The abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector is divided to obtain the set of abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vectors. Using the Fourier transform function, the frequency components corresponding to each abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector set are calculated, and a weight is assigned to each abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector; Using the set of each abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector and the weight of each abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector, a weighted comprehensive evaluation formula is established to calculate the interaction status indicators of the target vehicle's intelligent driving system software and hardware components.

[0008] Preferably, based on the Internet and the improvement plans of car companies, obtain functional failure feedback data of known intelligent driving systems; Hierarchical clustering is used to combine the functional failure feedback parameters of known intelligent driving systems according to the Levenshtein edit distance based on each functional failure type, and generate a functional failure type label for the target vehicle's intelligent driving system. Using linear mapping, spatially map the functional fault type labels of the target vehicle's intelligent driving system to the target vehicle's intelligent driving system hardware and software component interaction state indicators, obtaining the target vehicle's intelligent driving system functional fault type label vector and the target vehicle's intelligent driving system hardware and software component interaction state vector. The correlation coefficient is used to calculate the degree of correlation between the functional fault type label vector of the target vehicle's intelligent driving system and the interaction state vector of the target vehicle's intelligent driving system's soft and hard components. Positive correlation screening is performed to obtain the target vehicle's intelligent driving system's soft and hard component interaction functional fault type label set.

[0009] Preferably, a sliding window is used to count the frequency of occurrence of functional failure type labels in a set of functional failure type labels of interaction functions of software and hardware components of the intelligent driving system of the target vehicle per unit time, and a priori probability of the functional failure type label of the intelligent driving system of the target vehicle is calculated; Based on the functional failure type label of the intelligent driving system of the target vehicle, an undirected graph of functional failures of the intelligent driving system of the target vehicle is established with each functional failure type label as a node; Using the chi-square test, the dependency relationship between adjacent nodes in the initialized completely undirected graph is verified, and non-positive dependency edges are deleted to obtain the functional fault directed graph of the intelligent driving system of the target vehicle. According to the PageRank algorithm, the transition probability between each node in the functional failure directed graph of the intelligent driving system of the target vehicle is calculated, and the associated edge weight of each node is assigned to obtain the functional failure directed weighted graph of the intelligent driving system of the target vehicle; Based on the directed weighted graph of functional faults of the target vehicle's intelligent driving system, each fault point is used as the starting node, and the CPT probabilistic random walk is used to generate the fault propagation path, thus obtaining the known fault chain development network of the target vehicle's intelligent driving system's software and hardware components.

[0010] Preferably, based on the open environment EMC test of the target vehicle, a list of interaction requests for software and hardware components of the open road real-time intelligent driving system of the target vehicle under various test scenarios is obtained; Using the wavelet transform method, the spectrum of the open road real-time intelligent driving system hardware and software component interaction request list corresponding to each test scenario of the target vehicle is extracted; Based on the spectrum of the list of software and hardware component interaction requests for the open road intelligent driving system in various test scenarios, the single software and hardware component interaction request tasks are divided according to the unit timestamp, and the spectrum feature matrix of the software and hardware component interaction requests for the open road real-time intelligent driving system in various test scenarios is established; Calculate the Euclidean distance of each element in the spectrum feature matrix of the interaction requests of the software and hardware components of the real-time intelligent driving system on open roads under various test scenarios to each node in the known fault chain development network of the software and hardware components of the intelligent driving system of the target vehicle, and substitute it into the matrix to verify the post-transfer verification probability of the spectrum feature of the interaction requests of the software and hardware components of the real-time intelligent driving system on open roads under various test scenarios for the corresponding network nodes, and generate the fault chain development propagation path of the software and hardware components of the intelligent driving system under various test scenarios of the target vehicle.

[0011] Preferably, based on the development and propagation paths of the fault chains of the soft and hard components of the intelligent driving system under various test scenarios of the target vehicle, multi-dimensional characteristic parameters of the development of the fault chains of the soft and hard components of the intelligent driving system are marked, and a characteristic vector matrix of the development of the fault chains of the soft and hard components of the intelligent driving system is constructed; the multi-dimensional characteristic parameters of the development of the fault chains of the soft and hard components of the intelligent driving system include: component fault propagation transfer probability, component fault propagation path length, and component fault propagation soft and hard component type; Based on the fault chain development feature vector matrix of the intelligent driving system's hardware and software components, an observation function for the fault chain development of the hardware and software components is established to generate a multi-dimensional vector of the fault chain development trend of the intelligent driving system's hardware and software components under various test scenarios of the target vehicle. The method is as follows: ; in, is the binary state vector of the hardware and software components of the intelligent driving system at the t-th unit time, Develop a state transition matrix for the fault chain of hardware and software components of an intelligent driving system; is the EMC interference value of the intelligent driving system’s hardware and software components under the t-th unit time, B is the input matrix of the fault chain development state transition of the intelligent driving system’s hardware and software components, is the development noise of the intelligent driving system’s hardware and software components under the t-th unit time, G is the observation matrix of the development state of the fault chain of the intelligent driving system’s hardware and software components, is the observation noise of the intelligent driving system’s hardware and software components under the t-th unit time, E is the development activity of the fault chain of the intelligent driving system’s hardware and software components, is the total number of times the fault chain development path of the intelligent driving system's hardware and software components is activated within the time window T, where T is the observation time length. is the cumulative value of the overall risk score of the intelligent driving system's hardware and software components under the t-th unit time, is the activation probability of the kth fault chain development path of the hardware and software components of the intelligent driving system, is the time interval between the tth unit time and the last activation of the fault chain development path, To control the forgetting speed coefficient of historical risk, The decay rate coefficient that determines the effect of time interval on risk; Based on logistic regression, a fault chain development trend assessment model for intelligent driving systems is established; Using the analytic hierarchy process, we assign weights to the vector dimensions of the fault chain development trends of the soft and hard components of the intelligent driving system under various test scenarios of the target vehicle. Based on the intelligent driving system failure chain development trend assessment model, the multi-dimensional vectors of the failure chain development trends of the intelligent driving system's software and hardware components under various test scenarios of the target vehicle and the weights of each dimensional vector of the failure chain development trends of the intelligent driving system's software and hardware components under various test scenarios of the target vehicle are used as influencing factors to predict the functional failure risk score of the target vehicle's intelligent driving system's software and hardware components.

[0012] Compared with the prior art, the present invention has the following beneficial effects: The present invention proposes a reliability assessment scheme for intelligent driving systems based on EMC testing. By integrating static test data and dynamic road data, combining a priori fault knowledge base with real-time interaction features, a multi-dimensional fault development network model is constructed. This scheme can accurately identify abnormal interaction patterns of software and hardware under electromagnetic interference, and dynamically predict functional failure risks. It provides quantitative assessment means and risk warning capabilities for intelligent driving systems in complex electromagnetic environments, significantly improving the safety and robustness of monitoring of intelligent driving systems undergoing EMC testing. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a flow chart of an intelligent driving system reliability assessment method based on EMC testing. DETAILED DESCRIPTION

[0014] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are merely examples, and those skilled in the art may conceive of other obvious variations.

[0015] Reference Figure 1As shown in FIG, a reliability assessment method for an intelligent driving system based on EMC testing includes: Step 1: Based on the closed-environment EMC test of the target vehicle, obtain the target vehicle's intelligent driving system software and hardware interaction request data under closed electromagnetic interference, and evaluate the target vehicle's intelligent driving system software and hardware interaction status; The step 1 includes the following: Based on several rounds of EMC testing in a closed anechoic chamber of the target vehicle, the software and hardware interaction request data of the intelligent driving system under the closed electromagnetic interference of the target vehicle is obtained; As a further example, since the target vehicle is stationary during EMC testing in a closed anechoic chamber, it is still possible to use the background API of the intelligent driving system to call function functions and perform interface call spoofing to simulate the software and hardware interaction of the vehicle in motion. According to hardware interaction synchronization and hardware interaction asynchrony, the software and hardware interaction request data of the intelligent driving system under closed electromagnetic interference of the target vehicle are divided to obtain the software and hardware interaction synchronous request data and the software and hardware interaction asynchronous request data of the intelligent driving system under closed electromagnetic interference of the target vehicle; Using linear interpolation, asynchronously request data completion for the intelligent driving system's hardware and software interactions under closed electromagnetic interference of the target vehicle; Using the intelligent driving system hardware and software interaction transmission protocol, the target vehicle's closed electromagnetic interference intelligent driving system hardware and software interaction synchronous request data and the intelligent driving system hardware and software interaction asynchronous request data are temporally and spatially aligned to obtain the target vehicle's closed electromagnetic interference intelligent driving system hardware and software interaction synchronous and asynchronous request spatiotemporal alignment data. The time step of the synchronous-asynchronous request of the intelligent driving system's software and hardware interaction is used as the observation window, and the synchronous-asynchronous request interaction data of the intelligent driving system's software and hardware interaction is used as the observation object. The timing data of the synchronous-asynchronous request of the intelligent driving system's software and hardware interaction under the target vehicle's closed electromagnetic interference is obtained. Based on ARIMA autoregressive integrated moving average, a synchronous-asynchronous request deviation analysis model for the intelligent driving system's software and hardware interactions is established; Using partial autocorrelation functions and autocorrelation functions, the synchronous-asynchronous request timing data of the intelligent driving system's hardware and software interactions under closed electromagnetic interference of the target vehicle is verified, and the autoregressive order and moving average order are determined; Based on the intelligent driving system software and hardware interaction synchronous-asynchronous request deviation analysis model, the autoregressive order and moving average order are used as training constraints, the target vehicle's closed electromagnetic interference intelligent driving system software and hardware interaction synchronous-asynchronous request time series data observation value is used as input, and the target vehicle's closed electromagnetic interference intelligent driving system software and hardware interaction synchronous-asynchronous request time series data prediction value is used as output; Calculate the difference between the observed value and the predicted value of the synchronous-asynchronous request time series data of the intelligent driving system software and hardware interaction under the closed electromagnetic interference of the target vehicle, and establish the residual sequence of the synchronous-asynchronous request time series data of the intelligent driving system software and hardware interaction under the closed electromagnetic interference of the target vehicle; Determine the target vehicle intelligent driving system software and hardware interaction synchronous-asynchronous request standardized time series data and substitute it into the intelligent driving system software and hardware interaction synchronous-asynchronous request deviation analysis model to obtain the target vehicle intelligent driving system software and hardware interaction synchronous-asynchronous request standardized time series data residual sequence; According to the three-times standard deviation threshold method, verify whether the standard deviation of the residual sequence of the target vehicle's intelligent driving system software and hardware interaction synchronous-asynchronous request timing data under closed electromagnetic interference is within the range of three times the standard deviation of the residual sequence of the target vehicle's intelligent driving system software and hardware interaction synchronous-asynchronous request standardized timing data. If so, mark it as a normal driving software and hardware interaction synchronous-asynchronous request; if not, mark it as an abnormal driving software and hardware interaction synchronous-asynchronous request; Determine the error coding types of synchronous-asynchronous requests for normal driving software-hardware interaction and synchronous-asynchronous requests for abnormal driving software-hardware interaction; Using one-hot encoding, we map and transform the error coding types of synchronous-asynchronous requests for normal driving software-hardware interaction and those for abnormal driving software-hardware interaction. This yields the error coding type vectors for synchronous-asynchronous requests for normal driving software-hardware interaction and those for abnormal driving software-hardware interaction of the intelligent driving system under closed electromagnetic interference. Normalize the error coding type vectors of the synchronization-asynchronous request for normal driving software and hardware interaction and the error coding type vectors of abnormal driving software and hardware interaction of the intelligent driving system under closed electromagnetic interference of the target vehicle; Based on DBSCAN density clustering, the Mahalanobis distance between the normal driving software and hardware interaction synchronous-asynchronous request error coding type vector and the abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector of the intelligent driving system under closed electromagnetic interference of the target vehicle is used as the domain radius. The abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector is divided to obtain the set of abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vectors. Using the Fourier transform function, the frequency components corresponding to each abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector set are calculated, and a weight is assigned to each abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector; Using the set of each abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector and the weight of each abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector, a weighted comprehensive evaluation formula is established to calculate the interaction status indicators of the target vehicle's intelligent driving system software and hardware components.

[0016] When using, combine the contents in the above steps. As a further point, traditional EMC testing is usually carried out in a static environment and cannot truly reflect the dynamic electromagnetic interference during vehicle driving, resulting in deviations between test results and actual scenarios. In addition, existing methods lack the ability to distinguish between synchronous / asynchronous requests for hardware interaction, resulting in insufficient accuracy in timing data alignment and difficulty in capturing microsecond-level signal delays or packet loss problems under electromagnetic interference. Secondly, statistical analysis cannot identify intermittent anomalies caused by electromagnetic interference (such as occasional error frames on the CAN bus), and does not consider the frequency domain characteristics of the error type. Most solutions do not combine the unique communication protocols of intelligent driving systems (such as the PDU routing mechanism under the AutoSAR architecture) for data analysis, resulting in a disconnect between interaction status evaluation and actual hardware behavior.

[0017] This solution combines synchronous / asynchronous classification, spatiotemporal alignment, and ARIMA residual analysis to significantly improve the anomaly detection accuracy and timing resolution of intelligent driving systems under electromagnetic interference. It uses DBSCAN clustering and frequency domain weighting to build an error feature library, establishes a full-stack evaluation from electromagnetic interference patterns to hardware components, enhances the diagnostic granularity of intelligent driving systems, and solves the problems of large static and dynamic differences and low detection rate of hidden errors in traditional EMC testing.

[0018] Step 2: Based on the internet and the improvement plans of car companies, we obtain the functional fault feedback data of known intelligent driving systems and conduct prior distribution screening on the interaction status of the target vehicle's intelligent driving system software and hardware components to build a known fault chain development network of the target vehicle's intelligent driving system software and hardware components; The second step includes the following: Obtaining functional failure feedback data of known intelligent driving systems based on the Internet and car companies' improvement plans; Hierarchical clustering is used to combine the functional failure feedback parameters of known intelligent driving systems according to the Levenshtein edit distance based on each functional failure type, and generate a functional failure type label for the target vehicle's intelligent driving system. Using linear mapping, spatially map the functional fault type labels of the target vehicle's intelligent driving system to the target vehicle's intelligent driving system hardware and software component interaction state indicators, obtaining the target vehicle's intelligent driving system functional fault type label vector and the target vehicle's intelligent driving system hardware and software component interaction state vector. Using the correlation coefficient, the correlation between the target vehicle's intelligent driving system's functional fault type label vector and the target vehicle's intelligent driving system's hardware and software component interaction state vector is calculated. Positive correlation screening is performed to obtain the target vehicle's intelligent driving system's hardware and software component interaction functional fault type label set. Using a sliding window, the frequency of functional failure type labels in the target vehicle's intelligent driving system software and hardware component interaction function failure type label set per unit time is counted, and the prior probability of the target vehicle's intelligent driving system functional failure type label is calculated; Based on the functional failure type label of the intelligent driving system of the target vehicle, an undirected graph of functional failures of the intelligent driving system of the target vehicle is established with each functional failure type label as a node; Using the chi-square test, the dependency relationship between adjacent nodes in the initialized completely undirected graph is verified, and non-positive dependency edges are deleted to obtain the functional fault directed graph of the intelligent driving system of the target vehicle. According to the PageRank algorithm, the transition probability between each node in the functional failure directed graph of the intelligent driving system of the target vehicle is calculated, and the associated edge weight of each node is assigned to obtain the functional failure directed weighted graph of the intelligent driving system of the target vehicle; Based on the directed weighted graph of functional faults of the target vehicle's intelligent driving system, each fault point is used as a starting node, and the CPT probabilistic random walk is used to generate the fault propagation path, thereby obtaining the known fault chain development network of the target vehicle's intelligent driving system's software and hardware components; When using, combine the contents in the above steps. As a further development, traditional intelligent driving system fault analysis relies on a single data source, resulting in incomplete coverage of fault samples. Simple methods such as keyword matching make it difficult to accurately classify semantically similar faults (such as "AEB false triggering" and "automatic emergency braking abnormality"). Static analysis based on expert experience or fixed rules cannot effectively explore the dynamic statistical dependencies between faults, making it difficult to identify key fault nodes. This overall restricts the comprehensiveness, accuracy, and timeliness of fault analysis.

[0019] By integrating multi-source data (Internet + automotive company data), a comprehensive fault sample library is constructed. Intelligent fault merging is achieved by combining Levenshtein edit distance and hierarchical clustering, significantly improving classification accuracy. The chi-square test and PageRank algorithm are used to dynamically mine fault associations and quantify key nodes, accurately identifying high-impact faults. Fault propagation paths are simulated based on CPT random walks, and a sliding window mechanism is used to adaptively update fault paths. This improves the fault analysis and recognition rate of intelligent driving systems, increases the efficiency of discovering fault development paths, and supports the real-time incorporation of new fault modes.

[0020] Step 3: Based on the open environment EMC test of the target vehicle, the posterior distribution of the target vehicle's open road real-time intelligent driving system software and hardware component interaction request list is analyzed according to the target vehicle's intelligent driving system software and hardware component failure chain development network, and the target vehicle's intelligent driving system software and hardware component failure chain development propagation path is determined; The step three includes the following: Based on the open environment EMC test of the target vehicle, a list of software and hardware interaction requests for the open road real-time intelligent driving system of the target vehicle is obtained under various test scenarios; the various test scenarios include: urban road scenarios, highway scenarios, and special scenarios of high-voltage substations; Using the wavelet transform method, the spectrum of the open road real-time intelligent driving system hardware and software component interaction request list corresponding to each test scenario of the target vehicle is extracted; Based on the spectrum of the list of software and hardware component interaction requests for the open road intelligent driving system in various test scenarios, the single software and hardware component interaction request tasks are divided according to the unit timestamp, and the spectrum feature matrix of the software and hardware component interaction requests for the open road real-time intelligent driving system in various test scenarios is established; Calculate the Euclidean distance of each element in the spectrum feature matrix of the interaction requests of the software and hardware components of the real-time intelligent driving system on open roads under various test scenarios to each node in the known fault chain development network of the software and hardware components of the intelligent driving system of the target vehicle, and substitute it into the matrix to verify the post-transfer verification probability of the spectrum feature of the interaction requests of the software and hardware components of the real-time intelligent driving system on open roads under various test scenarios for the corresponding network nodes, and generate the fault chain development propagation path of the software and hardware components of the intelligent driving system under various test scenarios of the target vehicle.

[0021] When using, combine the contents in the above steps. As a further aspect, the existing EMC testing scenarios for intelligent driving systems are limited to standard open road environments, lacking coverage of special electromagnetic scenarios such as high-voltage substations, making it difficult to capture instantaneous interference characteristics; and the fault propagation modeling uses static assumptions and cannot reflect the dynamic evolution characteristics under different electromagnetic environments; the analysis method relies on post-processing and lacks real-time monitoring capabilities; secondly, the test data of each scenario is analyzed in isolation, and interference coupling correlation across scenarios cannot be performed, which restricts the comprehensiveness and accuracy of the test.

[0022] This solution builds an EMC interference database covering real driving environments through multi-dimensional scenario testing. It uses wavelet transform to implement millisecond-level time-frequency analysis to accurately capture transient interference characteristics. Based on dynamic probability transfer analysis, it improves the accuracy of fault chain location and enhances the precision of EMC test intelligent driving systems.

[0023] Step 4: Based on the development and propagation paths of the fault chains of the target vehicle's intelligent driving system's hardware and software components, establish an intelligent driving system fault chain development trend assessment model to predict the target vehicle's intelligent driving system hardware and software component functional failure risk score; The step 4 includes the following contents: Based on the development and propagation paths of the fault chains of the soft and hard components of the intelligent driving system under various test scenarios of the target vehicle, multi-dimensional characteristic parameters of the fault chains of the soft and hard components of the intelligent driving system are marked, and a characteristic vector matrix of the fault chains of the soft and hard components of the intelligent driving system is constructed. The multi-dimensional characteristic parameters of the fault chains of the soft and hard components of the intelligent driving system include: component fault propagation transfer probability, component fault propagation path length, and component fault propagation soft and hard component type; Based on the fault chain development feature vector matrix of the intelligent driving system's hardware and software components, an observation function for the fault chain development of the hardware and software components is established to generate a multi-dimensional vector of the fault chain development trend of the intelligent driving system's hardware and software components under various test scenarios of the target vehicle. The method is as follows: ; in, is the state binary vector of the intelligent driving system's hardware and software components at the t-th unit time, and A is the fault chain development state transition matrix of the intelligent driving system's hardware and software components; is the EMC interference value of the intelligent driving system’s hardware and software components under the t-th unit time, B is the input matrix of the fault chain development state transition of the intelligent driving system’s hardware and software components, is the development noise of the intelligent driving system’s hardware and software components under the t-th unit time, G is the observation matrix of the development state of the fault chain of the intelligent driving system’s hardware and software components, is the observation noise of the intelligent driving system’s hardware and software components under the t-th unit time, E is the development activity of the fault chain of the intelligent driving system’s hardware and software components, is the total number of times the fault chain development path of the intelligent driving system's hardware and software components is activated within the time window T, where T is the observation time length. is the cumulative value of the overall risk score of the intelligent driving system's hardware and software components under the t-th unit time, is the activation probability of the kth fault chain development path of the hardware and software components of the intelligent driving system, is the time interval between the tth unit time and the last activation of the fault chain development path, To control the forgetting speed coefficient of historical risk, The decay rate coefficient that determines the effect of time interval on risk; Based on logistic regression, a fault chain development trend assessment model for intelligent driving systems is established; Using the analytic hierarchy process, we assign weights to the vector dimensions of the fault chain development trends of the soft and hard components of the intelligent driving system under various test scenarios of the target vehicle. Based on the intelligent driving system failure chain development trend assessment model, the multi-dimensional vectors of the failure chain development trends of the intelligent driving system's soft and hard components under various test scenarios of the target vehicle and the weights of the dimensional vectors of the failure chain development trends of the intelligent driving system's soft and hard components under various test scenarios of the target vehicle are used as influencing factors to predict the functional failure risk score of the target vehicle's intelligent driving system's soft and hard components; When using, combine the contents in the above steps. As a further example, traditional intelligent driving system risk assessment relies solely on a single failure probability dimension (such as RPN scoring), ignoring key features such as propagation path length and component type; the use of fixed weight distribution (such as ASIL level) cannot adapt to the dynamic needs of different scenarios; simple analysis based on linear models makes it difficult to capture the nonlinear synergistic effects between failures; the lack of unified assessment standards for each scenario makes it impossible to compare risk priorities horizontally; and the lack of real-time data processing capabilities means that only post-analysis can be performed, and risk warnings cannot be implemented, resulting in one-sided assessment results, poor adaptability, and delayed response.

[0024] This solution integrates key parameters such as fault propagation probability, path length, and component type, and combines dynamic weight optimization and nonlinear modeling techniques to achieve unified risk quantification assessment across scenarios. This enables this EMC testing method for intelligent driving systems to have real-time risk warning capabilities. With the help of visual analysis tools, it significantly improves the comprehensiveness and accuracy of risk assessments, providing support for the safe design and quantitative numerical decision-making of intelligent driving systems.

[0025] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions merely illustrate the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A reliability assessment method for an intelligent driving system based on EMC testing, characterized in that: include: S1. Based on the closed-environment EMC test of the target vehicle, obtain the target vehicle's intelligent driving system hardware and software interaction request data under closed electromagnetic interference, and evaluate the target vehicle's intelligent driving system hardware and software interaction status; S2. Based on the internet and the improvement plans of automobile companies, functional fault feedback data of known intelligent driving systems is obtained and the interaction status of the target vehicle's intelligent driving system hardware and software components is screened by prior distribution to build a development network of known fault chains of the target vehicle's intelligent driving system hardware and software components. S3. Based on the open environment EMC test of the target vehicle, analyze the posterior distribution of the target vehicle's open road real-time intelligent driving system software and hardware component interaction request list according to the target vehicle's intelligent driving system software and hardware component failure chain development network, and determine the target vehicle's intelligent driving system software and hardware component failure chain development propagation path; S4. Based on the development and propagation path of the fault chain of the target vehicle's intelligent driving system's hardware and software components, establish an intelligent driving system fault chain development trend assessment model to predict the target vehicle's intelligent driving system's hardware and software components' functional failure risk score.

2. The reliability assessment method of an intelligent driving system based on EMC testing according to claim 1, characterized in that: Said S1 comprises: Based on several rounds of EMC testing in a closed anechoic chamber of the target vehicle, the software and hardware interaction request data of the intelligent driving system under the closed electromagnetic interference of the target vehicle is obtained; According to hardware interaction synchronization and hardware interaction asynchrony, the software and hardware interaction request data of the intelligent driving system under closed electromagnetic interference of the target vehicle are divided to obtain the software and hardware interaction synchronous request data and the software and hardware interaction asynchronous request data of the intelligent driving system under closed electromagnetic interference of the target vehicle; Using linear interpolation, asynchronously request data completion for the intelligent driving system's hardware and software interactions under closed electromagnetic interference of the target vehicle; Using the intelligent driving system hardware and software interaction transmission protocol, the target vehicle's closed electromagnetic interference intelligent driving system hardware and software interaction synchronous request data and the intelligent driving system hardware and software interaction asynchronous request data are temporally and spatially aligned to obtain the target vehicle's closed electromagnetic interference intelligent driving system hardware and software interaction synchronous and asynchronous request spatiotemporal alignment data. Taking the time step of the synchronous-asynchronous request of the intelligent driving system's software and hardware interaction as the observation window, and the synchronous-asynchronous request interaction data of the intelligent driving system's software and hardware interaction as the observation object, the synchronous-asynchronous request timing data of the intelligent driving system's software and hardware interaction under closed electromagnetic interference of the target vehicle is obtained.

3. The reliability assessment method of an intelligent driving system based on EMC testing according to claim 2, characterized in that: Said S1 further comprises: Based on ARIMA autoregressive integrated moving average, a synchronous-asynchronous request deviation analysis model for the intelligent driving system's software and hardware interactions is established; Using partial autocorrelation functions and autocorrelation functions, the synchronous-asynchronous request timing data of the intelligent driving system's hardware and software interactions under closed electromagnetic interference of the target vehicle is verified, and the autoregressive order and moving average order are determined; Based on the intelligent driving system software and hardware interaction synchronous-asynchronous request deviation analysis model, the autoregressive order and moving average order are used as training constraints, the target vehicle's closed electromagnetic interference intelligent driving system software and hardware interaction synchronous-asynchronous request time series data observation value is used as input, and the target vehicle's closed electromagnetic interference intelligent driving system software and hardware interaction synchronous-asynchronous request time series data prediction value is used as output; Calculate the difference between the observed value and the predicted value of the synchronous-asynchronous request time series data of the intelligent driving system software and hardware interaction under the closed electromagnetic interference of the target vehicle, and establish the residual sequence of the synchronous-asynchronous request time series data of the intelligent driving system software and hardware interaction under the closed electromagnetic interference of the target vehicle; Determine the standardized timing data of the synchronous-asynchronous requests of the target vehicle intelligent driving system's software and hardware interaction and substitute it into the intelligent driving system's software and hardware interaction synchronous-asynchronous request deviation analysis model to obtain the residual sequence of the standardized timing data of the target vehicle intelligent driving system's software and hardware interaction synchronous-asynchronous requests.

4. The reliability assessment method of an intelligent driving system based on EMC testing according to claim 3, characterized in that: Said S1 further comprises: According to the three-times standard deviation threshold method, verify whether the standard deviation of the residual sequence of the target vehicle's intelligent driving system software and hardware interaction synchronous-asynchronous request timing data under closed electromagnetic interference is within the range of three times the standard deviation of the residual sequence of the target vehicle's intelligent driving system software and hardware interaction synchronous-asynchronous request standardized timing data. If so, mark it as a normal driving software and hardware interaction synchronous-asynchronous request; if not, mark it as an abnormal driving software and hardware interaction synchronous-asynchronous request; Determine the error coding types of synchronous-asynchronous requests for normal driving software-hardware interaction and synchronous-asynchronous requests for abnormal driving software-hardware interaction; Using one-hot encoding, we map and transform the error coding types of synchronous-asynchronous requests for normal driving software-hardware interaction and those for abnormal driving software-hardware interaction. This yields the error coding type vectors for synchronous-asynchronous requests for normal driving software-hardware interaction and those for abnormal driving software-hardware interaction of the intelligent driving system under closed electromagnetic interference. Normalize the error coding type vectors of the synchronization-asynchronous request for normal driving software and hardware interaction and the error coding type vectors of abnormal driving software and hardware interaction of the intelligent driving system under closed electromagnetic interference of the target vehicle; Based on DBSCAN density clustering, the Mahalanobis distance between the normal driving software and hardware interaction synchronous-asynchronous request error coding type vector and the abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector of the intelligent driving system under closed electromagnetic interference of the target vehicle is used as the domain radius. The abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector is divided to obtain the set of abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vectors. Using the Fourier transform function, the frequency components corresponding to each abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector set are calculated, and a weight is assigned to each abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector; Using the set of each abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector and the weight of each abnormal driving software and hardware interaction synchronous-asynchronous request error coding type vector, a weighted comprehensive evaluation formula is established to calculate the interaction status indicators of the target vehicle's intelligent driving system software and hardware components.

5. The reliability assessment method of an intelligent driving system based on EMC testing according to claim 4, characterized in that: The S2 includes: Obtaining functional failure feedback data of known intelligent driving systems based on the Internet and car companies' improvement plans; Hierarchical clustering is used to combine the functional failure feedback parameters of known intelligent driving systems according to the Levenshtein edit distance based on each functional failure type, and generate a functional failure type label for the target vehicle's intelligent driving system. Using linear mapping, spatially map the functional fault type labels of the target vehicle's intelligent driving system to the target vehicle's intelligent driving system hardware and software component interaction state indicators, obtaining the target vehicle's intelligent driving system functional fault type label vector and the target vehicle's intelligent driving system hardware and software component interaction state vector. The correlation coefficient is used to calculate the degree of correlation between the functional fault type label vector of the target vehicle's intelligent driving system and the interaction state vector of the target vehicle's intelligent driving system's soft and hard components. Positive correlation screening is performed to obtain the target vehicle's intelligent driving system's soft and hard component interaction functional fault type label set.

6. The reliability assessment method of an intelligent driving system based on EMC testing according to claim 5, characterized in that: Said S2 further comprises: Using a sliding window, the frequency of functional failure type labels in the target vehicle's intelligent driving system software and hardware component interaction function failure type label set per unit time is counted, and the prior probability of the target vehicle's intelligent driving system functional failure type label is calculated; Based on the functional failure type label of the intelligent driving system of the target vehicle, an undirected graph of functional failures of the intelligent driving system of the target vehicle is established with each functional failure type label as a node; Using the chi-square test, the dependency relationship between adjacent nodes in the initialized completely undirected graph is verified, and non-positive dependency edges are deleted to obtain the functional fault directed graph of the intelligent driving system of the target vehicle. According to the PageRank algorithm, the transition probability between each node in the functional failure directed graph of the intelligent driving system of the target vehicle is calculated, and the associated edge weight of each node is assigned to obtain the functional failure directed weighted graph of the intelligent driving system of the target vehicle; Based on the directed weighted graph of functional faults of the target vehicle's intelligent driving system, each fault point is used as the starting node, and the CPT probabilistic random walk is used to generate the fault propagation path, thus obtaining the known fault chain development network of the target vehicle's intelligent driving system's software and hardware components.

7. The reliability assessment method of an intelligent driving system based on EMC testing according to claim 6, characterized in that: The S3 includes: Based on the open environment EMC test of the target vehicle, obtain the software and hardware interaction request list of the open road real-time intelligent driving system under various test scenarios of the target vehicle; Using the wavelet transform method, the spectrum of the open road real-time intelligent driving system hardware and software component interaction request list corresponding to each test scenario of the target vehicle is extracted; Based on the spectrum of the list of software and hardware component interaction requests for the open road intelligent driving system in various test scenarios, the single software and hardware component interaction request tasks are divided according to the unit timestamp, and the spectrum feature matrix of the software and hardware component interaction requests for the open road real-time intelligent driving system in various test scenarios is established; Calculate the Euclidean distance of each element in the spectrum feature matrix of the interaction requests of the software and hardware components of the real-time intelligent driving system on open roads under various test scenarios to each node in the known fault chain development network of the software and hardware components of the intelligent driving system of the target vehicle, and substitute it into the matrix to verify the post-transfer verification probability of the spectrum feature of the interaction requests of the software and hardware components of the real-time intelligent driving system on open roads under various test scenarios for the corresponding network nodes, and generate the fault chain development propagation path of the software and hardware components of the intelligent driving system under various test scenarios of the target vehicle.

8. The reliability assessment method of an intelligent driving system based on EMC testing according to claim 7, characterized in that: The S4 includes: Based on the development and propagation paths of the fault chains of the soft and hard components of the intelligent driving system under various test scenarios of the target vehicle, multi-dimensional characteristic parameters of the fault chains of the soft and hard components of the intelligent driving system are marked, and a characteristic vector matrix of the fault chains of the soft and hard components of the intelligent driving system is constructed. The multi-dimensional characteristic parameters of the fault chains of the soft and hard components of the intelligent driving system include: component fault propagation transfer probability, component fault propagation path length, and component fault propagation soft and hard component type; Based on the fault chain development characteristic vector matrix of the intelligent driving system's hardware and software components, an observation function for the fault chain development of the hardware and software components is established to generate a multi-dimensional vector of the fault chain development trend of the intelligent driving system's hardware and software components under various test scenarios of the target vehicle. The method is as follows: ； in, is the state binary vector of the intelligent driving system's hardware and software components at the t-th unit time, and A is the fault chain development state transition matrix of the intelligent driving system's hardware and software components; is the EMC interference value of the intelligent driving system’s hardware and software components under the t-th unit time, B is the input matrix of the fault chain development state transition of the intelligent driving system’s hardware and software components, is the development noise of the intelligent driving system’s hardware and software components under the t-th unit time, G is the observation matrix of the development state of the fault chain of the intelligent driving system’s hardware and software components, is the observation noise of the intelligent driving system’s hardware and software components under the t-th unit time, E is the development activity of the fault chain of the intelligent driving system’s hardware and software components, is the total number of times the fault chain development path of the intelligent driving system's hardware and software components is activated within the time window T, where T is the observation time length. is the cumulative value of the overall risk score of the intelligent driving system's hardware and software components under the t-th unit time, is the activation probability of the kth fault chain development path of the hardware and software components of the intelligent driving system, is the time interval between the tth unit time and the last activation of the fault chain development path, To control the forgetting speed coefficient of historical risk, The decay rate coefficient that determines the effect of time interval on risk; Based on logistic regression, a fault chain development trend assessment model for intelligent driving systems is established; Using the analytic hierarchy process, we assign weights to the vector dimensions of the fault chain development trends of the soft and hard components of the intelligent driving system under various test scenarios of the target vehicle. Based on the intelligent driving system failure chain development trend assessment model, the multi-dimensional vectors of the failure chain development trends of the intelligent driving system's software and hardware components under various test scenarios of the target vehicle and the weights of each dimensional vector of the failure chain development trends of the intelligent driving system's software and hardware components under various test scenarios of the target vehicle are used as influencing factors to predict the functional failure risk score of the target vehicle's intelligent driving system's software and hardware components.

Images