End-cloud collaborative diagnosis and trust evaluation of drive-by-wire chassis error degradation suppression method and related equipment

Abstract

The application discloses a drive-by-wire chassis mis-degradation suppression method and related equipment for end-cloud cooperative diagnosis and credible evaluation, which can be applied to the technical field of intelligent networked vehicles. The application inputs an observation vector constructed by multiple source heterogeneous signals into a lightweight fault diagnosis model on the vehicle side to perform fault prediction and obtain a candidate fault event set. Then, the application performs end-side credibility evaluation on candidate fault events in the candidate fault event set based on multiple dimensions on the vehicle side to obtain an end-side comprehensive credibility. The application uploads a compressed feature package to the cloud side to enable the cloud side to perform cloud-side credibility evaluation on candidate fault events in the candidate fault event set to obtain a cloud-side comprehensive credibility. Then, when the end-side comprehensive credibility is located in a fuzzy interval, the application generates a target reduction operation signal in combination with the end-side comprehensive credibility, so that the degradation response speed and the diagnosis accuracy can be improved. After the target reduction operation signal is executed, the application updates and optimizes evaluation weights and fusion parameters, so that the diagnosis accuracy can be further improved.

Description

Technical Field

[0001] This application relates to the field of intelligent connected vehicle technology, and in particular to a method and related equipment for suppressing erroneous degradation of drive-by-wire chassis in a reliable assessment of end-to-cloud collaborative diagnostics. Background Technology

[0002] In related technologies, with the large-scale deployment of L2+ and L3 level autonomous driving, the functional safety requirements for drive-by-wire chassis, as a key vehicle execution system, are becoming increasingly stringent. Current safety strategies generally adopt the principle of "better to report a false alarm than to miss one." Once a suspected fault is detected, such as abnormal brake master cylinder pressure or interruption of steering torque, degraded measures are immediately triggered, such as speed limiting or switching to mechanical backup mode, to avoid the risk of loss of control. This strategy is prone to misjudging operating conditions, leading to unnecessary degrades and affecting the driving experience. Furthermore, limitations in vehicle-side computing power and communication latency result in low accuracy and slow response times for current degrade diagnostics.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0004] The main objective of this application is to propose a method and related equipment for suppressing the false degradation of a drive-by-wire chassis in a reliable assessment of end-to-end cloud collaborative diagnosis, which can effectively improve diagnostic accuracy and degradation response speed.

[0005] To achieve the above objectives, one aspect of this application proposes a method for suppressing erroneous degradation of a drive-by-wire chassis based on end-to-cloud collaborative diagnostic reliability assessment. The method includes the following steps: Acquire multi-source heterogeneous signals from the target vehicle; An observation vector is constructed based on the multi-source heterogeneous signals; The observation vector is input into the lightweight fault diagnosis model on the vehicle to predict the fault and obtain a candidate fault event set. The elements in the candidate fault event set include the fault type, the preliminary confidence probability corresponding to each fault, and the fault occurrence timestamp. On the vehicle side, the candidate fault events in the candidate fault event set are evaluated based on multi-dimensional indicators to obtain the end-side comprehensive credibility. The compressed feature package is then uploaded to the cloud so that the cloud can evaluate the candidate fault events in the candidate fault event set to obtain the cloud comprehensive credibility. The information in the compressed feature package includes fault type, observation vector, steering wheel angle, and fault occurrence timestamp. When the edge-side comprehensive credibility is in the fuzzy range, the fusion credibility is calculated based on the edge-side comprehensive credibility and the cloud-side comprehensive credibility. A target reduction operation signal is generated based on the fusion confidence level and fusion threshold. Based on the actual maintenance tag and actual downgrade result corresponding to the target reduction operation signal, optimize the evaluation weight of the end-side comprehensive credibility and the fusion parameters of the cloud-based comprehensive credibility, and send the optimized evaluation weight to the vehicle end to update the vehicle end's evaluation weight.

[0006] In some embodiments, constructing the observation vector based on the multi-source heterogeneous signals includes: Time synchronization and alignment are performed on the multi-source heterogeneous signals; The observation vector is composed of multi-source heterogeneous signals that have been time-synchronized and aligned.

[0007] In some embodiments, the step of evaluating the edge-side credibility of candidate fault events within the candidate fault event set based on multi-dimensional indicators to obtain the edge-side comprehensive credibility includes: Calculate the confidence scores of multiple components for each candidate fault event within the candidate fault event set. The confidence scores of each component include accuracy confidence score, logical confidence score, relevance confidence score, completeness confidence score, and consistency confidence score. The accuracy confidence level, logical confidence level, relevance confidence level, integrity confidence level, and consistency confidence level are weighted and summed to obtain the end-side comprehensive confidence level corresponding to each candidate fault event.

[0008] In some embodiments, the formula for calculating the end-side integrated reliability is as follows: ; In the formula, This represents the edge-side comprehensive confidence level of the i-th candidate fault event; This represents the confidence level of the k-th component of the i-th candidate fault event; Let represent the evaluation weight of the credibility of the k-th component, and .

[0009] In some embodiments, the step of performing a cloud-based credibility assessment on the candidate fault events within the candidate fault event set to obtain a comprehensive cloud-based credibility includes: The digital twin simulation matching degree of each candidate fault event is calculated based on the digital twin model corresponding to the target vehicle in the cloud. The credibility of the graph neural network output for each candidate fault event is calculated based on a cloud-based structured causal graph knowledge base. The historical case similarity score for each candidate fault event is calculated based on a cloud-based historical fault case database. The cloud-based comprehensive credibility is obtained by weighting and summing the matching degree of the digital twin simulation, the credibility of the graph neural network output, and the similarity score of the historical cases.

[0010] In some embodiments, the formula for calculating the overall trustworthiness of the cloud is as follows: ; In the formula, This represents the overall cloud-based credibility of the i-th candidate fault event; Indicates the matching degree of the digital twin simulation; Fusion parameters representing the matching degree of digital twin simulation; This indicates the credibility of the graph neural network output. The fusion parameters represent the confidence level of the graph neural network output; This indicates the similarity score of historical cases; The fusion parameter represents the similarity score of historical cases.

[0011] In some embodiments, the formula for calculating the fusion credibility is as follows: ; In the formula, This represents the fusion confidence level of the i-th candidate fault event; This represents the edge-side comprehensive confidence level of the i-th candidate fault event; This represents the overall cloud-based credibility of the i-th candidate fault event; ; Indicates the current system time. Indicates the time-degradation coefficient; This represents the timestamp of the occurrence of the i-th candidate fault event.

[0012] To achieve the above objectives, another aspect of this application proposes a device for suppressing false degradation of a drive-by-wire chassis in a cloud-edge collaborative diagnostic reliability assessment, the device comprising: The first module is used to acquire multi-source heterogeneous signals from the target vehicle; The second module is used to construct an observation vector based on the multi-source heterogeneous signals; The third module is used to input the observation vector into the lightweight fault diagnosis model of the vehicle terminal to perform fault prediction and obtain a candidate fault event set. The elements in the candidate fault event set include the fault type, the preliminary confidence probability corresponding to each fault, and the fault occurrence timestamp. The fourth module is used to perform end-side credibility assessment on the candidate fault events in the candidate fault event set based on multi-dimensional indicators at the vehicle end to obtain the end-side comprehensive credibility; and to upload the compressed feature package to the cloud so that the cloud can perform cloud credibility assessment on the candidate fault events in the candidate fault event set to obtain the cloud comprehensive credibility. The information in the compressed feature package includes fault type, observation vector, steering wheel angle and fault occurrence timestamp. The fifth module is used to calculate the fusion credibility based on the terminal-side comprehensive credibility and the cloud-based comprehensive credibility when the terminal-side comprehensive credibility is in a fuzzy range. The sixth module is used to generate a target reduction operation signal based on the fusion confidence level and the fusion threshold. The seventh module is used to optimize the evaluation weight of the end-side comprehensive credibility and the fusion parameters of the cloud-based comprehensive credibility based on the real maintenance tag and actual downgrade result corresponding to the target reduction operation signal, and to send the optimized evaluation weight to the vehicle end to update the evaluation weight of the vehicle end.

[0013] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0014] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods described above.

[0015] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer program product, including a computer program that, when executed by a processor, implements the aforementioned method.

[0016] The embodiments of this application include at least the following beneficial effects: This application provides a method and related equipment for suppressing the false degradation of a drive-by-wire chassis in a cloud-edge collaborative diagnostic reliability assessment. This scheme acquires multi-source heterogeneous signals from the target vehicle, constructs observation vectors based on these signals, inputs the observation vectors into a lightweight fault diagnosis model on the vehicle side for fault prediction to obtain a candidate fault event set, and then performs end-side reliability assessment on the candidate fault events within the candidate fault event set based on multi-dimensional indicators on the vehicle side to obtain an end-side comprehensive reliability. The compressed feature package is then uploaded to the cloud, enabling the cloud to perform cloud-based reliability assessment on the candidate fault events within the candidate fault event set to obtain a cloud-based comprehensive reliability. Then... When the end-side comprehensive confidence level is in the fuzzy range, the fusion confidence level is calculated based on the end-side comprehensive confidence level and the end comprehensive confidence level. Then, a target reduction operation signal is generated based on the fusion confidence level and the fusion threshold. This allows the vehicle-side lightweight fault diagnosis model to be used for fault prediction to improve the degradation response speed. The combination of vehicle-side and cloud-side approaches improves diagnostic accuracy. After executing the target reduction operation signal, the evaluation weight of the end-side comprehensive confidence level and the fusion parameters of the cloud-side comprehensive confidence level are optimized based on the real maintenance label and actual degradation result corresponding to the execution of the target reduction operation signal. The optimized evaluation weight is then sent to the vehicle-side to update the vehicle-side evaluation weight, thereby further improving diagnostic accuracy. Attached Figure Description

[0017] Figure 1 This is a flowchart of the wire-controlled chassis erroneous degradation suppression method for end-to-cloud collaborative diagnostic reliability assessment provided in the embodiments of this application; Figure 2 This is a schematic diagram of the structure of the wire-controlled chassis false degradation suppression device for end-to-cloud collaborative diagnostic reliability assessment provided in the embodiments of this application; Figure 3 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0019] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0020] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0022] In related technologies, with the large-scale deployment of L2+ and L3 level autonomous driving, the functional safety requirements for drive-by-wire chassis, as a key vehicle execution system, are becoming increasingly stringent. Current safety strategies generally adopt the principle of "better to report a false alarm than to miss one." Once a suspected fault is detected, such as abnormal brake master cylinder pressure or steering torque interruption, degrade measures are immediately triggered, such as speed limiting or switching to mechanical backup mode, to avoid the risk of loss of control. However, this strategy has the following drawbacks in practical applications: First, sensor noise, communication jitter, or software flaws can easily misjudge normal dynamic conditions as serious faults, leading to frequent unnecessary degrades, severely impacting the driving experience and user trust. Second, existing diagnostic mechanisms lack the ability to quantitatively assess the reliability of diagnostic results and are limited by vehicle-side computing power, only able to run lightweight diagnostic logic. While the cloud possesses high-precision inference capabilities, communication latency makes it difficult to participate in millisecond-level real-time decision-making, resulting in low accuracy and slow response of existing degrade diagnostic results.

[0023] In view of this, this application provides a method and related equipment for suppressing the false degradation of a drive-by-wire chassis in a cloud-edge collaborative diagnostic reliability assessment, which can effectively improve diagnostic accuracy and degradation response speed.

[0024] The method for suppressing the false degradation of a drive-by-wire chassis in end-to-end collaborative diagnostic and trusted assessment provided in this application relates to the field of intelligent connected vehicle technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or in-vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the method for suppressing the false degradation of a drive-by-wire chassis in end-to-end collaborative diagnostic and trusted assessment, but is not limited to the above forms.

[0025] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0026] It is understood that the method in this embodiment is deployed in an end-to-cloud collaborative system consisting of an on-board domain controller and a cloud-based diagnostic platform. The on-board unit is responsible for real-time acquisition of multi-source heterogeneous signals, including wheel speed, brake master cylinder pressure, steering angle, motor current, and CAN bus communication status. It also has a built-in lightweight fault detection module and a five-dimensional confidence preliminary assessment engine, which can complete the preliminary confidence quantification of diagnostic results within 10 milliseconds. It can also maintain low-latency, high-reliability data interaction with the cloud through 5G or V2X communication links.

[0027] The cloud platform builds a high-concurrency evaluation task pool based on distributed computing frameworks (such as Ray or Spark), integrates vehicle digital twin models, historical fault case libraries, and structured causal graph knowledge bases, and can perform high-precision credibility-enhanced inference on concurrent diagnostic requests from multiple vehicles, and efficiently transmit the verification results back to the vehicle, supporting collaborative decision-making and dynamic inhibition control.

[0028] The embodiments of this application will be described in detail below with reference to the accompanying drawings: Figure 1 This is an optional flowchart of the wire-controlled chassis erroneous degradation suppression method for end-to-cloud collaborative diagnostic reliability assessment provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S110 to S170: Step S110: Acquire the multi-source heterogeneous signal of the target vehicle; Step S120: Construct observation vectors based on multi-source heterogeneous signals; Step S130: Input the observation vector into the lightweight fault diagnosis model on the vehicle to perform fault prediction and obtain a candidate fault event set. The elements in the candidate fault event set include the fault type, the preliminary confidence probability corresponding to each fault, and the fault occurrence timestamp. Step S140: On the vehicle terminal, perform end-side credibility assessment on the candidate fault events in the candidate fault event set based on multi-dimensional indicators to obtain end-side comprehensive credibility; and upload the compressed feature package to the cloud so that the cloud can perform cloud credibility assessment on the candidate fault events in the candidate fault event set to obtain cloud comprehensive credibility. The information in the compressed feature package includes fault type, observation vector, steering wheel angle and fault occurrence timestamp. Step S150: When the edge-side comprehensive credibility is in the fuzzy range, calculate the fusion credibility based on the edge-side comprehensive credibility and the cloud-side comprehensive credibility. Step S160: Generate a target reduction operation signal based on the fusion confidence level and fusion threshold; Step S170: Based on the execution target, reduce the real maintenance label corresponding to the operation signal and the actual downgrade result, optimize the evaluation weight of the end-side comprehensive credibility and the fusion parameter of the cloud-side comprehensive credibility, and send the optimized evaluation weight to the vehicle end to update the evaluation weight of the vehicle end.

[0029] It is understood that the multi-source heterogeneous signals in this embodiment can originate from EHB (Electro-hydraulic Braking System), SBW (Steering-by-Wire System), motor controller, and vehicle network. After obtaining the multi-source heterogeneous signals, this embodiment performs high-level time synchronization on them to form an observation vector. The vehicle domain controller controls... (Typical value 10 ms) Acquire sensor signals to form the observation vector shown below: ; In the formula, This indicates the wheel rotation speed (or wheel speed), usually measured in rpm or m / s, and reflects the vehicle's motion state. This indicates the braking pressure (or pedal pressure), a pressure signal from the EHB (electro-hydraulic brake) system, used to determine if the brakes are malfunctioning. Indicates the steering angle (steering wheel angle), derived from the EPS (Electric Power Steering) system, reflecting the driver's intention; It represents the motor current, reflects the working status of the drive motor, and can be used to determine faults such as motor overload and short circuit. It indicates the CAN bus communication status, including node online / offline status, message loss rate, number of error frames, etc., and is used to determine whether network communication is normal.

[0030] It is understandable that, after obtaining the observation vector, this embodiment uses a lightweight fault diagnosis model on the vehicle side to predict and output a set of candidate fault events, including the fault type, preliminary confidence probability, and fault occurrence timestamp. This provides structured input for subsequent reliability assessment. The candidate fault event set predicted and output by the vehicle-side lightweight fault diagnosis model is as follows: ; In the formula, Indicate the fault type (e.g., "EHB master cylinder pressure abnormality"), This indicates the initial confidence probability. Indicates the timestamp of the fault occurrence.

[0031] Understandably, this embodiment initiates a dual-channel evaluation simultaneously on both the vehicle and cloud platforms after generating the candidate fault event set. On the vehicle side, a millisecond-level initial credibility assessment is performed on non-high-risk faults based on five dimensions: accuracy, logic, relevance, completeness, and consistency. Simultaneously, a compressed feature package is uploaded to the cloud for high-precision credibility enhancement using digital twin simulation, causal graph neural networks, and a historical case library, resulting in a parallel evaluation result that balances real-time performance and accuracy.

[0032] It is understandable that when performing credibility assessment on the vehicle side, this embodiment can calculate multiple component credibility of each candidate fault event in the candidate fault event set, including accuracy credibility, logical credibility, relevance credibility, integrity credibility and consistency credibility, and then perform a weighted summation of the accuracy credibility, logical credibility, relevance credibility, integrity credibility and consistency credibility to obtain the end-side comprehensive credibility of each candidate fault event.

[0033] Specifically, for each non-high-risk fault type The five-dimensional credibility score vector is calculated using the following formula: ; The confidence level of each component in each dimension is defined as follows: The formula for calculating accuracy and reliability is as follows: ; In the formula, This represents the predicted value from the dynamic model; Indicates the attenuation coefficient (calibration); .

[0034] The formula for calculating logical credibility is as follows: ; In the formula, Precompiled causal graph (JSON / hash table); The formula for calculating the reliability of relevance is as follows: ; In the formula, This indicates that highways / slopes have a high correlation (1.0), while parking has a low correlation (0.1). The formula for calculating integrity credibility is as follows: ; In the formula, This indicates that the root factors have been checked. This represents the total number of possible root factors.

[0035] The formula for calculating consistency reliability is as follows: ; In the formula, Indicates a redundant sensor group. This represents the variance sensitivity coefficient.

[0036] After calculating the accuracy credibility, logical credibility, relevance credibility, completeness credibility, and consistency credibility, a weighted sum is performed to obtain the overall end-side credibility. The calculation formula is as follows: ; In the formula, This represents the edge-side comprehensive confidence level of the i-th candidate fault event; This represents the confidence level of the k-th component of the i-th candidate fault event; Let represent the evaluation weight of the credibility of the k-th component, and .

[0037] Understandably, the vehicle-side compresses the feature packets. After the data (<100MB) is uploaded to the cloud, the cloud can perform parallel computation of the cloud-based comprehensive credibility based on the Ray distributed framework. Specifically, in this embodiment, the digital twin simulation matching degree of each candidate fault event can be calculated based on the digital twin model corresponding to the target vehicle in the cloud, the graph neural network output credibility of each candidate fault event can be calculated based on the structured causal graph knowledge base in the cloud, and the historical case similarity score of each candidate fault event can be calculated based on the historical fault case library in the cloud. Finally, the digital twin simulation matching degree, graph neural network output credibility, and historical case similarity score are weighted and summed to obtain the cloud-based comprehensive credibility.

[0038] Specifically, the formula for calculating the matching degree of digital twin simulation is as follows: ; In the formula, Indicates the matching degree of the digital twin simulation; Represents the physical system's first The actual measured values ​​at each moment (such as vehicle speed, motor current); This represents the simulation output of the digital twin model under the same conditions; The denominator represents the true mean of the data; the denominator is the total variance, and the numerator is the sum of squared residuals.

[0039] The formula for calculating the confidence level of a graph neural network output is as follows: ; In the formula, This indicates the credibility of the graph neural network output. This represents the output of the graph neural network.

[0040] The calculation process for historical case similarity scores is as follows: Let the current fault feature vector be A case in the historical case database is ,but Then, the score is taken from the most similar historical cases. ,and .

[0041] This embodiment calculates the digital twin simulation matching degree, graph neural network output credibility, and historical case similarity score separately, and then performs a weighted sum to obtain the overall cloud-based credibility. The calculation formula is as follows: ; In the formula, This represents the overall cloud-based credibility of the i-th candidate fault event; Indicates the matching degree of the digital twin simulation; Fusion parameters representing the matching degree of digital twin simulation; This indicates the credibility of the graph neural network output. The fusion parameters represent the confidence level of the graph neural network output; This indicates the similarity score of historical cases; The fusion parameter represents the similarity score of historical cases.

[0042] It is understood that this embodiment can implement a real-time three-level dynamic decision-making strategy based on the interval of the edge-side comprehensive credibility. The strategy is expressed by the following formula: ; In the formula, This represents the edge-side comprehensive confidence level of the i-th candidate fault event. Indicates a high threshold; This indicates a low threshold. Typical threshold values ​​are: (High confidence threshold); (Below the confidence threshold).

[0043] Specifically, when the overall trust level at the edge exceeds a high threshold, a security degradation will be triggered immediately. ) as a goal decision ( When the threshold is below a certain level, degradation will be forcibly suppressed and the alarm will be cleared. ) as a goal decision ( When the result is in a fuzzy range, a time-weighted mechanism will be introduced to integrate the results from the cloud. ) as a goal decision ( This embodiment only allows downgrading when the overall confidence level exceeds the final decision threshold, effectively avoiding unnecessary functional interruptions caused by diagnostic misjudgments.

[0044] It is understandable that this embodiment introduces a time-weighted fusion mechanism to fuse the cloud-based comprehensive trustworthiness when the overall trustworthiness on the edge is in an ambiguous range. Specifically, the formula for calculating the fused trustworthiness is as follows: ; In the formula, This represents the fusion confidence level of the i-th candidate fault event; This represents the edge-side comprehensive confidence level of the i-th candidate fault event; This represents the overall cloud-based credibility of the i-th candidate fault event; ; Indicates the current system time. Indicates the aging decay coefficient (typical value 0.01 ms). - ¹); This represents the timestamp of the occurrence of the i-th candidate fault event.

[0045] In the embodiments of this application, if (like If the alarm is triggered, a downgrade will be triggered; otherwise, the alarm will be cleared and the original control mode will be maintained.

[0046] Understandably, this embodiment also uses the actual maintenance tags and actual degradation results corresponding to the target reduction operation signals executed during multi-vehicle operation as supervision signals. It continuously optimizes the end-side credibility assessment weights and cloud-based fusion parameters through an online gradient descent algorithm, and sends them to the vehicle via a secure OTA channel, achieving adaptive evolution and robustness improvement of the credibility model throughout the vehicle's entire lifecycle. The cloud aggregates multi-vehicle decision results and actual maintenance tags. Update weights via online gradient descent: ; In the formula, Represents the loss function (such as F1-score guided loss). This represents the learning rate.

[0047] The updated parameters are sent to the vehicle via OTA to update the vehicle's evaluation weights, thereby effectively improving the accuracy of vehicle fault prediction.

[0048] As can be seen from the above, the method of this application embodiment, by constructing a parallel end-cloud trusted evaluation architecture and a multi-dimensional dynamic degradation suppression mechanism, effectively meets the stringent requirements of drive-by-wire chassis for millisecond-level real-time response and high-level functional safety without increasing hardware redundancy. It significantly reduces the false degradation rate caused by sensor noise, communication jitter, or model bias. The vehicle end adopts a lightweight five-dimensional trust model, which can complete self-check of the preliminary diagnostic results within 10 milliseconds, avoiding blind reliance on a single confidence output. The cloud relies on a distributed computing framework to execute digital twin simulation, causal graph reasoning, and historical case comparison in parallel, providing high-precision trust enhancement. The results from both ends achieve collaborative decision-making of "fast response" and "high confidence" through a time-weighted fusion strategy. This can fundamentally solve the problem of unnecessary safety degradation caused by the lack of self-reflection ability in existing drive-by-wire chassis, significantly improve the robustness, rationality, and interpretability of fault diagnosis, and effectively ensure the continuity of operation, control stability, and user trust in high-level autonomous driving scenarios.

[0049] Please see Figure 2 This application also provides a device for suppressing the false degradation of a drive-by-wire chassis for end-to-end collaborative diagnostic reliability assessment. The device includes: The first module is used to acquire multi-source heterogeneous signals from the target vehicle; The second module is used to construct observation vectors based on multi-source heterogeneous signals; The third module is used to input the observation vector into the lightweight fault diagnosis model on the vehicle end for fault prediction and obtain a candidate fault event set. The elements in the candidate fault event set include the fault type, the preliminary confidence probability corresponding to each fault, and the fault occurrence timestamp. The fourth module is used to perform end-side credibility assessment on the candidate fault events in the candidate fault event set based on multi-dimensional indicators at the vehicle end to obtain the end-side comprehensive credibility; and to upload the compressed feature package to the cloud so that the cloud can perform cloud credibility assessment on the candidate fault events in the candidate fault event set to obtain the cloud comprehensive credibility. The information in the compressed feature package includes fault type, observation vector, steering wheel angle and fault occurrence timestamp. The fifth module is used to calculate the fusion credibility based on the edge-side comprehensive credibility and the cloud-based comprehensive credibility when the edge-side comprehensive credibility is in the fuzzy range. The sixth module is used to generate a target reduction operation signal based on the fusion confidence level and the fusion threshold. The seventh module is used to optimize the evaluation weight of the end-side comprehensive credibility and the fusion parameters of the cloud-side comprehensive credibility based on the actual maintenance label and actual degradation result corresponding to the operation signal, and to send the optimized evaluation weight to the vehicle end to update the evaluation weight of the vehicle end.

[0050] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0051] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0052] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0053] Please see Figure 3 , Figure 3 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 310 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 320 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 320 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 320 and is called and executed by the processor 310 using the methods described in the embodiments of this application. Input / output interface 330 is used to realize information input and output; The communication interface 340 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 350 transmits information between various components of the device (e.g., processor 310, memory 320, input / output interface 330, and communication interface 340); The processor 310, memory 320, input / output interface 330 and communication interface 340 are connected to each other within the device via bus 350.

[0054] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0055] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0056] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0057] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0058] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0059] This application provides a method and related equipment for suppressing false degradation of a drive-by-wire chassis in a cloud-edge collaborative diagnostic reliability assessment. The scheme acquires multi-source heterogeneous signals from the target vehicle, constructs observation vectors based on these signals, and inputs these vectors into a lightweight fault diagnosis model on the vehicle to predict faults and obtain a candidate fault event set. Then, on the vehicle, the candidate fault events within the candidate fault event set are evaluated for end-side reliability based on multi-dimensional indicators to obtain an end-side comprehensive reliability. The compressed feature package is then uploaded to the cloud, allowing the cloud to perform a cloud-based reliability evaluation of the candidate fault events within the candidate fault event set to obtain a cloud-based comprehensive reliability. Finally, the end-side comprehensive reliability is evaluated... When the target is in a fuzzy region, the fusion confidence is calculated based on the end-side comprehensive confidence and the end-side comprehensive confidence. The target reduction operation signal is then generated based on the fusion confidence and the fusion threshold. This allows for fault prediction using a lightweight fault diagnosis model on the vehicle side to improve the speed of the degradation response. The combination of vehicle-side and cloud-side approaches improves diagnostic accuracy. After executing the target reduction operation signal, the evaluation weight of the end-side comprehensive confidence and the fusion parameters of the cloud-side comprehensive confidence are optimized based on the real maintenance label and actual degradation result corresponding to the execution target reduction operation signal. The optimized evaluation weight is then sent to the vehicle side to update the vehicle side's evaluation weight, thereby further improving diagnostic accuracy.

[0060] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0061] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0062] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0063] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0064] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification 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.

[0065] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0066] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only 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 coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0067] The units described above 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 network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0068] 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.

[0069] 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 multiple 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 programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0070] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for suppressing erroneous degradation of a drive-by-wire chassis based on end-to-cloud collaborative diagnostic reliability assessment, characterized in that, The method includes the following steps: Acquire multi-source heterogeneous signals from the target vehicle; An observation vector is constructed based on the multi-source heterogeneous signals; The observation vector is input into the lightweight fault diagnosis model on the vehicle to predict the fault and obtain a candidate fault event set. The elements in the candidate fault event set include the fault type, the preliminary confidence probability corresponding to each fault, and the fault occurrence timestamp. On the vehicle side, the candidate fault events in the candidate fault event set are evaluated based on multi-dimensional indicators to obtain the end-side comprehensive credibility. The compressed feature package is then uploaded to the cloud so that the cloud can evaluate the candidate fault events in the candidate fault event set to obtain the cloud comprehensive credibility. The information in the compressed feature package includes fault type, observation vector, steering wheel angle, and fault occurrence timestamp. When the edge-side comprehensive credibility is in the fuzzy range, the fusion credibility is calculated based on the edge-side comprehensive credibility and the cloud-side comprehensive credibility. A target reduction operation signal is generated based on the fusion confidence level and fusion threshold. Based on the actual maintenance tag and actual downgrade result corresponding to the target reduction operation signal, optimize the evaluation weight of the end-side comprehensive credibility and the fusion parameters of the cloud-based comprehensive credibility, and send the optimized evaluation weight to the vehicle end to update the vehicle end's evaluation weight.

2. The method according to claim 1, characterized in that, The step of constructing the observation vector based on the multi-source heterogeneous signals includes: Time synchronization and alignment are performed on the multi-source heterogeneous signals; The observation vector is composed of multi-source heterogeneous signals that have been time-synchronized and aligned.

3. The method according to claim 1, characterized in that, The endpoint-side credibility assessment of candidate fault events within the candidate fault event set based on multi-dimensional indicators yields an endpoint-side comprehensive credibility score, including: Calculate the confidence scores of multiple components for each candidate fault event within the candidate fault event set. The confidence scores of each component include accuracy confidence score, logical confidence score, relevance confidence score, completeness confidence score, and consistency confidence score. The accuracy confidence level, logical confidence level, relevance confidence level, integrity confidence level, and consistency confidence level are weighted and summed to obtain the end-side comprehensive confidence level corresponding to each candidate fault event.

4. The method according to claim 3, characterized in that, The formula for calculating the end-side integrated reliability is as follows: ； In the formula, This represents the edge-side comprehensive confidence level of the i-th candidate fault event; This represents the confidence level of the k-th component of the i-th candidate fault event; Let represent the evaluation weight of the credibility of the k-th component, and .

5. The method according to claim 1, characterized in that, The cloud-based credibility assessment of candidate fault events within the candidate fault event set, to obtain a comprehensive cloud-based credibility, includes: The digital twin simulation matching degree of each candidate fault event is calculated based on the digital twin model corresponding to the target vehicle in the cloud. The credibility of the graph neural network output for each candidate fault event is calculated based on a cloud-based structured causal graph knowledge base. The historical case similarity score for each candidate fault event is calculated based on a cloud-based historical fault case database. The cloud-based comprehensive credibility is obtained by weighting and summing the matching degree of the digital twin simulation, the credibility of the graph neural network output, and the similarity score of the historical cases.

6. The method according to claim 4, characterized in that, The formula for calculating the overall credibility of the cloud platform is as follows: ； In the formula, This represents the overall cloud-based credibility of the i-th candidate fault event; Indicates the matching degree of the digital twin simulation; Fusion parameters representing the matching degree of digital twin simulation; This indicates the credibility of the graph neural network output. The fusion parameters represent the confidence level of the graph neural network output; This indicates the similarity score of historical cases; The fusion parameter represents the similarity score of historical cases.

7. The method according to claim 1, characterized in that, The formula for calculating the fusion credibility is as follows: ； In the formula, This represents the fusion confidence level of the i-th candidate fault event; This represents the edge-side comprehensive confidence level of the i-th candidate fault event; This represents the overall cloud-based credibility of the i-th candidate fault event; ; Indicates the current system time. Indicates the time-degradation coefficient; This represents the timestamp of the occurrence of the i-th candidate fault event.

8. A device for suppressing false degradation of a drive-by-wire chassis in a reliable assessment of end-to-end cloud collaborative diagnostics, characterized in that, The device includes: The first module is used to acquire multi-source heterogeneous signals from the target vehicle; The second module is used to construct an observation vector based on the multi-source heterogeneous signals; The third module is used to input the observation vector into the lightweight fault diagnosis model of the vehicle terminal to perform fault prediction and obtain a candidate fault event set. The elements in the candidate fault event set include the fault type, the preliminary confidence probability corresponding to each fault, and the fault occurrence timestamp. The fourth module is used to perform end-side credibility assessment on the candidate fault events in the candidate fault event set based on multi-dimensional indicators at the vehicle end to obtain the end-side comprehensive credibility; and to upload the compressed feature package to the cloud so that the cloud can perform cloud credibility assessment on the candidate fault events in the candidate fault event set to obtain the cloud comprehensive credibility. The information in the compressed feature package includes fault type, observation vector, steering wheel angle and fault occurrence timestamp. The fifth module is used to calculate the fusion credibility based on the terminal-side comprehensive credibility and the cloud-based comprehensive credibility when the terminal-side comprehensive credibility is in a fuzzy range. The sixth module is used to generate a target reduction operation signal based on the fusion confidence level and the fusion threshold. The seventh module is used to optimize the evaluation weight of the end-side comprehensive credibility and the fusion parameters of the cloud-based comprehensive credibility based on the real maintenance tag and actual downgrade result corresponding to the target reduction operation signal, and to send the optimized evaluation weight to the vehicle end to update the evaluation weight of the vehicle end.

9. An electronic device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method as described in any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

Images