Intelligent automobile automatic driving fault diagnosis method and system

Abstract

The invention provides an intelligent automobile automatic driving fault diagnosis method and system, and relates to the technical field of automobile fault diagnosis, and the technical scheme comprises the steps: obtaining the operation state data of a to-be-monitored part of an automobile, and determining a fault phenomenon; a diagnosis strategy library is constructed, the diagnosis strategy library comprises diagnosis strategies of multiple levels, and the diagnosis strategy of each level comprises a corresponding diagnosis process, a judgment rule and a processing measure; based on the fault phenomenon, a diagnosis strategy most conforming to the current fault phenomenon is matched from the diagnosis strategies of the multiple levels, and processing measures in the most conforming diagnosis strategy are executed; obtaining feedback data after processing is finished; and based on the feedback data, updating and optimizing the diagnosis strategy in the diagnosis strategy library. Through combined use of the cluster sentinel network and the preset diagnosis strategy library, the diagnosis efficiency is improved, the adaptive learning algorithm is introduced to continuously optimize and adjust the diagnosis strategy, the diagnosis accuracy is improved, and misdiagnosis and missed diagnosis phenomena are reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of automobile fault diagnosis, and in particular to a method and system for diagnosing faults of automatic driving of intelligent vehicles. Background Art

[0002] With the rapid development of autonomous driving technology, vehicle systems are becoming increasingly complex, and the interdependence and correlation between systems have increased significantly. This has made troubleshooting a very challenging task in the process of autonomous driving software development and vehicle maintenance.

[0003] The traditional troubleshooting process for automatic driving of smart cars usually requires the collaboration of multiple departments and professionals, which is time-consuming, labor-intensive and inefficient. Traditional automotive diagnostic tools and methods can only provide superficial or preliminary fault information, and cannot go deep into the system to accurately locate the root cause of the fault, resulting in poor diagnostic accuracy. Summary of the invention

[0004] In order to overcome the shortcomings of the above-mentioned prior art, the present invention provides a method and system for diagnosing faults in automatic driving of intelligent vehicles. By combining the use of a cluster sentinel network and a preset diagnostic strategy library, the diagnostic efficiency is improved, and an adaptive learning algorithm is introduced to continuously optimize and adjust the diagnostic strategy, thereby improving the diagnostic accuracy and reducing the occurrence of misdiagnosis and missed diagnosis.

[0005] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0006] The first aspect of the present invention provides a method for diagnosing faults in automatic driving of an intelligent vehicle.

[0007] Obtain the operating status data of the vehicle components to be monitored and determine the fault phenomenon;

[0008] Constructing a diagnostic strategy library, wherein the diagnostic strategy library includes multiple levels of diagnostic strategies, and each level of diagnostic strategy includes corresponding diagnostic processes, judgment rules and processing measures;

[0009] Based on the fault phenomenon and in combination with the diagnostic strategy library, the diagnostic strategy that best matches the current fault phenomenon is matched from multiple levels of diagnostic strategies, and the treatment measures in the most matching diagnostic strategy are executed;

[0010] Get feedback data after processing;

[0011] Based on the feedback data, the diagnostic strategies in the diagnostic strategy library are updated and optimized.

[0012] As an optional technical solution, the operating status data of the vehicle components to be monitored is obtained in the following specific ways:

[0013] Build a distributed monitoring network consisting of multiple sentinel nodes;

[0014] Arrange multiple sentinel nodes on key systems and components of the vehicle to collect vehicle operation status data in real time, each of the sentinel nodes includes multiple sensors and has data processing capabilities;

[0015] The multiple sentinel nodes are connected to each other via a communication network.

[0016] As an optional technical solution, the multiple levels of diagnostic strategies are formulated based on the principles of the vehicle system, historical fault data and expert experience. The multiple levels of diagnostic strategies specifically include primary diagnostic strategies, intermediate diagnostic strategies and advanced diagnostic strategies. The primary diagnostic strategies, intermediate diagnostic strategies and advanced diagnostic strategies gradually narrow the scope of faults.

[0017] As an optional technical solution, the primary diagnostic network, intermediate diagnostic network and advanced diagnostic network are constructed using preset rules. The primary diagnostic rules include the correspondence between fault phenomena and fault types; the intermediate diagnostic rules include the correspondence between rough fault types and detailed fault types; and the advanced diagnostic rules include detailed fault types and corresponding fault descriptions.

[0018] As an optional technical solution, determine the fault phenomenon, the specific method is as follows:

[0019] The current sentinel node aggregates the vehicle operation status data collected by itself and the vehicle operation status data transmitted by other sentinel nodes to construct an operation data set;

[0020] Determine the current fault phenomenon based on the constructed operation data set and fault diagnosis network;

[0021] Among them, for image data, the fault diagnosis network adopts convolutional neural network, and for time series data, the fault diagnosis network uses recurrent neural network and its variant long short-term memory network.

[0022] As an optional technical solution, obtaining feedback data after the processing is completed includes:

[0023] Confirm the cause of the fault phenomenon through further data collection, testing or simulation;

[0024] After processing according to the treatment measures in the diagnostic strategy that best matches the current fault phenomenon and is most effective, it is determined whether the fault phenomenon has been eliminated, thereby determining the effectiveness of the selected treatment measures.

[0025] As an optional technical solution, the diagnostic strategies in the diagnostic strategy library are updated and optimized based on the feedback data, including:

[0026] Based on the adaptive learning mechanism, it continuously learns from feedback data, new fault cases and diagnostic experience;

[0027] When encountering new fault types or the diagnostic strategy is not effective, the diagnostic parameters and processes are automatically adjusted to update the diagnostic strategy library.

[0028] A second aspect of the present invention provides a smart car automatic driving fault diagnosis system, comprising:

[0029] The data acquisition and fault determination module is configured to: acquire the operating status data of the vehicle components to be monitored and determine the fault phenomenon;

[0030] The diagnostic strategy library construction module is configured to: construct a diagnostic strategy library, wherein the diagnostic strategy library includes multiple levels of diagnostic strategies, and each level of diagnostic strategy includes corresponding diagnostic processes, judgment rules and processing measures;

[0031] The matching module is configured to: match the diagnosis strategy that best matches the current fault phenomenon from multiple levels of diagnosis strategies based on the fault phenomenon, and execute the processing measures in the most consistent diagnosis strategy;

[0032] The feedback module is configured to: obtain feedback data after the processing is completed;

[0033] The diagnosis strategy update module is configured to update and optimize the diagnosis strategy in the diagnosis strategy library based on the feedback data.

[0034] The third aspect of the present invention provides a computer-readable storage medium on which a program is stored. When the program is executed by a processor, the steps in the intelligent vehicle automatic driving fault diagnosis method as described in the first aspect of the present invention are implemented.

[0035] The fourth aspect of the present invention provides an electronic device, including a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps in the intelligent vehicle automatic driving fault diagnosis method as described in the first aspect of the present invention are implemented.

[0036] One or more of the above technical solutions have the following beneficial effects:

[0037] The present invention provides a method and system for diagnosing faults in autonomous driving of intelligent vehicles, which significantly improves the diagnostic efficiency. Through the combined use of a cluster sentinel network and a preset diagnostic strategy library, the system can accurately and quickly locate vehicle system-level faults, greatly shortening the troubleshooting time.

[0038] The present invention significantly improves the diagnostic accuracy. The present invention introduces an adaptive learning algorithm, which enables the system to continuously optimize and adjust the diagnostic strategy, improve the diagnostic accuracy, and reduce the occurrence of misdiagnosis and missed diagnosis.

[0039] The present invention has relatively low maintenance costs. Through rapid and accurate fault diagnosis, the system can promptly detect and handle potential fault hazards, reduce vehicle downtime and repair costs caused by faults, and reduce maintenance costs.

[0040] The present invention enhances system stability and safety. Through accurate diagnosis of vehicle system-level faults and system-level maintenance optimization, it effectively improves the stability and safety of autonomous driving vehicles, providing strong guarantees for the popularization and application of autonomous driving technology.

[0041] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0043] Figure 1 This is a flow chart of the intelligent vehicle automatic driving fault diagnosis method of embodiment 1.

[0044] Figure 2 This is a data transmission diagram of Example 1. DETAILED DESCRIPTION

[0045] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0046] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

[0047] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0048] Embodiment 1

[0049] This embodiment discloses a method for diagnosing faults in automatic driving of an intelligent vehicle.

[0050] like Figure 1 As shown, the intelligent vehicle automatic driving fault diagnosis method includes the following steps:

[0051] Obtain the operating status data of the vehicle components to be monitored and determine the fault phenomenon;

[0052] Constructing a diagnostic strategy library, wherein the diagnostic strategy library includes multiple levels of diagnostic strategies, and each level of diagnostic strategy includes corresponding diagnostic processes, judgment rules and processing measures;

[0053] Based on the fault phenomenon, the diagnostic strategy that best matches the current fault phenomenon is matched from multiple levels of diagnostic strategies, and the treatment measures in the most consistent diagnostic strategy are executed;

[0054] Get feedback data after processing;

[0055] Based on the feedback data, the diagnostic strategies in the diagnostic strategy library are updated and optimized.

[0056] Furthermore, the operating status data of the vehicle components to be monitored is obtained in the following manner:

[0057] Build a distributed monitoring network consisting of multiple sentinel nodes;

[0058] Arrange multiple sentinel nodes on key systems and components of the vehicle to collect vehicle operation status data in real time, each of the sentinel nodes includes multiple sensors and has data processing capabilities;

[0059] The multiple sentinel nodes are connected to each other via a communication network.

[0060] Further, determine the fault phenomenon, the specific method is as follows:

[0061] The current sentinel node aggregates the vehicle operation status data collected by itself and the vehicle operation status data transmitted by other sentinel nodes to construct an operation data set;

[0062] Based on the constructed operation data set, the pre-trained fault diagnosis network is used to determine the current fault phenomenon;

[0063] In this embodiment, a suitable existing fault diagnosis network model can be selected according to the specific requirements and data characteristics of vehicle fault diagnosis. For example, convolutional neural network (CNN) is suitable for processing image data, and recurrent neural network (RNN) and its variant long short-term memory network (LSTM) have a good effect on processing time series data (such as sensor readings that change over time).

[0064] Furthermore, the multiple levels of diagnostic strategies are formulated based on the principles of the vehicle system, historical fault data and expert experience, and the multiple levels of diagnostic strategies specifically include primary diagnostic strategies, intermediate diagnostic strategies and advanced diagnostic strategies, and the primary diagnostic strategies, intermediate diagnostic strategies and advanced diagnostic strategies gradually narrow the scope of faults, and each level of diagnostic strategy includes corresponding diagnostic processes, judgment rules and processing measures;

[0065] In this embodiment, the primary diagnostic network, the intermediate diagnostic network and the advanced diagnostic network are constructed using preset rules. The primary diagnostic rules include the correspondence between fault phenomena and fault types; the intermediate diagnostic rules include the correspondence between rough fault types and detailed fault types; and the advanced diagnostic rules include detailed fault types and corresponding fault descriptions.

[0066] Furthermore, based on the fault phenomenon, the diagnostic strategy that best matches the current fault phenomenon is matched from multiple levels of diagnostic strategies, and the treatment measures in the most consistent diagnostic strategy are executed, including:

[0067] Consider the type of current fault phenomenon, frequency of occurrence, historical data, and current environmental conditions, and evaluate multiple possible diagnostic strategies in combination with the established diagnostic strategy library;

[0068] Select the diagnostic process in the most effective diagnostic strategy that best matches the current fault phenomenon, gradually troubleshoot and analyze the fault, narrow the scope of the problem, and implement corresponding treatment measures.

[0069] Furthermore, feedback data is obtained after the processing is completed, including:

[0070] Confirm the cause of the fault phenomenon through further data collection, testing or simulation;

[0071] After processing according to the treatment measures in the diagnostic strategy that best matches the current fault phenomenon and is most effective, it is determined whether the fault phenomenon has been eliminated, thereby determining the effectiveness of the selected treatment measures.

[0072] Furthermore, based on the feedback data, the diagnostic strategies in the diagnostic strategy library are updated and optimized, including:

[0073] Based on the adaptive learning mechanism, it continuously learns from feedback data, new fault cases and diagnostic experience;

[0074] When encountering new fault types or when the diagnostic strategy is ineffective, the diagnostic parameters and processes are automatically adjusted to improve diagnostic accuracy and efficiency.

[0075] This embodiment has a significant improvement in diagnostic efficiency. Through the combined use of the cluster sentinel network and the preset diagnostic strategy library, the system can accurately and quickly locate vehicle system-level faults, greatly shortening the troubleshooting time; the diagnostic accuracy is significantly improved. The present invention introduces an adaptive learning algorithm, which enables the system to continuously optimize and adjust the diagnostic strategy, improve diagnostic accuracy, and reduce the occurrence of misdiagnosis and missed diagnosis; the maintenance cost is relatively low. Through fast and accurate fault diagnosis, the system can promptly discover and deal with potential fault hazards, reduce vehicle downtime and maintenance costs caused by faults, and reduce maintenance costs; the system stability and safety are enhanced. Through accurate diagnosis of vehicle system-level faults and system-level maintenance optimization, the stability and safety of autonomous driving vehicles are effectively improved, providing a strong guarantee for the popularization and application of autonomous driving technology.

[0076] This embodiment uses a cluster sentinel network model to obtain data when acquiring the operating status data of the vehicle's monitored components; and pre-builds a diagnostic strategy library for matching the current fault phenomenon, thereby finding an optimal diagnostic strategy with the highest matching degree in the diagnostic strategy library; and uses an adaptive learning mechanism to update and optimize the diagnostic strategy library. The details are as follows:

[0077] (I) Cluster Sentinel Network:

[0078] A distributed monitoring network consisting of multiple sentinel nodes is built. These nodes are distributed on key systems and components of the vehicle to collect and transmit vehicle status data in real time. Working in clusters, the sentinel nodes can achieve comprehensive and detailed monitoring of the vehicle status and detect and report potential faults in a timely manner.

[0079] In this embodiment, the key systems and components of the vehicle may include a power system, such as an engine and a motor, and a chassis system, such as a brake system;

[0080] For the design of sentinel nodes, each sentinel node is equipped with highly sensitive sensors and data processing capabilities, which can monitor and record the operating status data of key vehicle components in real time. The nodes are connected to each other through a high-speed communication network to form a cluster network to achieve data sharing and collaborative work.

[0081] (II) Preset diagnostic strategy library:

[0082] A database containing rich diagnostic strategies and rules has been established, which are based on the principles of vehicle systems, historical fault data and expert experience. The system can automatically match and call the corresponding diagnostic strategy according to the fault phenomenon to conduct gradually in-depth fault diagnosis.

[0083] Among them, the multiple levels of diagnostic strategies specifically include primary diagnostic strategies, intermediate diagnostic strategies and advanced diagnostic strategies, and the primary diagnostic strategies, intermediate diagnostic strategies and advanced diagnostic strategies gradually narrow the scope of faults.

[0084] In specific implementation:

[0085] 1) Construction of diagnostic strategy library:

[0086] The diagnostic strategy library contains multiple levels of diagnostic strategies, from primary troubleshooting to in-depth system analysis, gradually narrowing the scope of the fault. Each strategy contains detailed diagnostic processes, judgment rules and treatment measures. The system automatically matches the most appropriate diagnostic strategy for execution based on the fault phenomenon. During execution, when the system detects a certain fault phenomenon (such as abnormal data fed back by the sensor, abnormal status of the component, or the driver's alarm prompt), the system will automatically select the most appropriate strategy from the pre-built diagnostic strategy library to analyze and handle the fault.

[0087] Specifically, the multiple levels of diagnostic strategies specifically include primary diagnostic strategies, intermediate diagnostic strategies and advanced diagnostic strategies, wherein the primary diagnostic strategies, intermediate diagnostic strategies and advanced diagnostic strategies gradually narrow the scope of the fault, and each level of diagnostic strategy includes corresponding diagnostic processes, judgment rules and processing measures;

[0088] In this embodiment, the primary diagnostic network, the intermediate diagnostic network and the advanced diagnostic network are constructed using preset rules. The primary diagnostic rules include the correspondence between fault phenomena and fault types; the intermediate diagnostic rules include the correspondence between rough fault types and detailed fault types; and the advanced diagnostic rules include detailed fault types and corresponding fault descriptions.

[0089] Among them, the primary diagnosis strategy is: determine the primary fault type according to the determined fault phenomenon combined with the primary diagnosis network;

[0090] Each fault phenomenon includes multiple fault types, for example, the fault phenomenon includes deviation in driving direction, unstable speed, etc.;

[0091] Take the driving direction deviation as an example:

[0092] Driving direction deviation may correspond to sensor damage, control system abnormality, map data error, key execution component failure, and communication failure;

[0093] The intermediate diagnosis strategy is: combining the primary fault type and the intermediate diagnosis network to determine the intermediate fault diagnosis result;

[0094] Since each primary fault type contains multiple fault types, such as:

[0095] Sensor damage includes cameras, millimeter-wave radars, laser radars and other sensors, because obstructions will affect the camera's recognition of roads and lane lines, which will cause the vehicle's driving direction to deviate. If the surface of millimeter-wave radars and laser radars is damaged, it will affect the transmission and reception of signals. For example, if cracks appear on the antenna cover of a millimeter-wave radar, it may change the propagation direction of the radar wave, leading to incorrect judgment of the position of surrounding vehicles or obstacles, and indirectly causing direction deviation.

[0096] Control system anomalies include calculation errors in the autonomous driving control algorithm and chassis control system errors;

[0097] The autonomous driving control algorithm has calculation errors. For example, when the sensor detects the position of the lane line, the control algorithm deviates from the direction and angle that the vehicle needs to adjust.

[0098] Errors in chassis control systems, such as the Electronic Power Steering System (EPS) and the Electronic Stability Program (ESP), play a key role in executing steering, braking and other operations during autonomous driving. If they do not work properly, they may cause the vehicle to deviate from the direction.

[0099] Map data error: High-precision maps are an important reference for autonomous vehicles, and are used to understand the road's curvature, slope, number of lanes, and other information in advance. If the map data is inaccurate, such as if lane information is not updated in a timely manner, the vehicle may drive according to incorrect map data in autonomous driving mode, resulting in directional deviation.

[0100] Failure of key actuators: such as steering motors, brake actuators, etc. These components are directly related to the direction and speed control of the vehicle. If these actuators fail, the vehicle may not be able to correctly control the direction according to the instructions of the autonomous driving system.

[0101] Communication failure: During the autonomous driving process, a large amount of data needs to be exchanged between systems, including sensor data, control instructions, etc. If there is a problem with the communication, such as data transmission delay, data packet loss, etc., it may cause the vehicle to deviate from the direction.

[0102] The advanced diagnosis strategy is to combine the intermediate fault diagnosis results and the advanced diagnosis network to determine the final fault result.

[0103] In order to locate more accurately based on the intermediate fault diagnosis results, especially for complex systems, it is still difficult to determine the specific fault type, for example, "chassis control system error", which is not conducive to subsequent maintenance. The detailed fault types include "capacitor damage", "chip short circuit", "line break", etc., so as to further determine the scope of the fault. For example, the scope of "power system fault" is too broad. If it is refined into "engine fuel injection system fault" and "turbocharger system fault", maintenance personnel can more quickly determine the key parts of the inspection.

[0104] 2) For matching diagnosis strategy:

[0105] The diagnostic strategy library has preset multi-level diagnostic strategies. Each strategy targets a specific fault type or phenomenon and contains detailed diagnostic steps and processing methods. The system will compare the detected fault phenomenon with each diagnostic strategy in the strategy library to analyze the possible fault type and severity of the phenomenon. After matching, the most appropriate strategy is selected. The system evaluates multiple possible diagnostic strategies through preset rules or algorithms, and selects the most matching and effective strategy for the current fault phenomenon. This process may involve considering factors such as the type of fault, frequency of occurrence, historical data, and current environmental conditions. Once the most appropriate strategy is matched, the system will gradually troubleshoot and analyze the fault according to the diagnostic process in the strategy, thereby narrowing the scope of the problem and proposing corresponding solutions.

[0106] In this embodiment, each level of fault type corresponds to a corresponding processing measure, including:

[0107] For example, for cameras, use the vehicle's own diagnostic program or professional equipment to test its image acquisition and processing capabilities. Check whether the camera's resolution, contrast, color reproduction and other parameters are within the normal range. At the same time, check the camera's recognition accuracy of key elements such as lane lines and traffic signs. For example, use a special calibration tool to show the camera a standard lane line pattern to see if it can accurately recognize and output the correct lane position information.

[0108] Test the signal strength and accuracy of millimeter wave radar and lidar. By sending and receiving analog signals, the radar's effective detection distance, angular resolution and other performance indicators are measured. If the radar's signal strength weakens or its accuracy decreases, it may lead to deviations in the perception of surrounding vehicles and obstacles, causing the vehicle to deviate from the direction when planning its driving path.

[0109] Check whether the control algorithm of the autonomous driving system is running normally. Read the system operation log through the vehicle's diagnostic interface. Check the control algorithm's processing of sensor data in the log to see whether the path planning and vehicle control are performed according to the preset rules. For example, check whether the control algorithm correctly calculates the direction and angle that the vehicle needs to adjust after the sensor detects the lane line position in the lane keeping function.

[0110] Professional diagnostic tools can be used to check the fault codes and operating parameters of the EPS and ESP systems.

[0111] Software updates usually fix some known system vulnerabilities and algorithm defects. If the software version is too old, there may be software problems that cause direction deviation. You can query and update the software through the software update interface of the vehicle's central control system or the manufacturer's official website.

[0112] Update the map data via the vehicle's map update function or by contacting the manufacturer's customer service.

[0113] For sensors suspected of being faulty, disassemble and inspect them (if possible and within the warranty period). For example, check the circuit board and lens assembly inside the camera to see if there are any damaged components or loose solder joints. For millimeter-wave radar and lidar, check the working status of key components such as the internal transmitting and receiving modules and signal processing chips. This requires professional electronic equipment repair technology and tools, and care must be taken to avoid secondary damage to the sensor during operation.

[0114] Professional testing equipment is used to measure parameters such as the torque output of the steering motor and the pressure of the brake actuator to see if they meet the design requirements. If these actuators fail, the vehicle may not be able to control the direction correctly according to the instructions of the autonomous driving system.

[0115] Professional communication testing equipment can be used to monitor the communication links between systems and find possible communication fault points.

[0116] 3) Finally, the selected diagnostic strategy is executed and the system may confirm the cause of the fault through further data collection, testing or simulation. After the diagnosis is completed, the system will feedback the results and take appropriate maintenance or adjustment measures according to the recommendations in the strategy.

[0117] (III) Adaptive learning algorithm:

[0118] This embodiment introduces advanced machine learning algorithms, which enable the system to continuously optimize and adjust the diagnostic strategy based on the feedback data during the actual diagnosis process. Through continuous learning and accumulation, the diagnostic accuracy and efficiency of the system will gradually improve.

[0119] For example, it can be learned through reinforcement learning algorithms. For example, various sensor data of the vehicle are defined as state variables, a state space model is constructed, and actions that can be taken are defined, such as adjusting engine parameters, controlling the steering system, etc. The reward function is designed according to the goal of fault diagnosis to guide the agent to learn the best action sequence. When encountering new fault cases, the agent learns and adjusts the action strategy in the environment through continuous trial and error to find the best diagnosis and processing method.

[0120] The system has a built-in adaptive learning algorithm that can continuously learn new fault cases and diagnostic experience. When the system encounters a new fault type or the diagnostic strategy is ineffective, it will automatically adjust the diagnostic parameters and processes to update the diagnostic strategy library to improve diagnostic accuracy and efficiency.

[0121] 4. Real-time feedback and remote collaboration:

[0122] The system has a real-time feedback function, which can promptly transmit the diagnosis results to developers, maintenance personnel or remote support teams. At the same time, it supports remote collaboration, allowing professionals in different locations to jointly participate in troubleshooting, further improving work efficiency.

[0123] The system provides an intuitive and easy-to-use user interface and a remote collaboration platform, allowing developers, maintenance personnel or remote support teams to view diagnostic results in real time, exchange opinions and jointly develop solutions.

[0124] like Figure 2 As shown, in this embodiment, multiple sentinel nodes are respectively arranged on ECU1, ECU2, ECU3, and ECU4, so as to construct a distributed monitoring network; multiple sentinel nodes on ECU1, ECU2, ECU3, and ECU4 are mutually connected in communication, and each sentinel node is respectively connected in communication with VCC; the diagnostic instrument using the method of this embodiment is connected in communication with VCC. In addition, a cloud and a PC are also provided.

[0125] In the specific implementation, taking a self-driving vehicle as an example, when the vehicle has driving abnormalities (such as direction deviation, unstable speed, etc.), the system immediately starts the fault diagnosis process. The sentinel node network comprehensively monitors the operating status data of each system of the vehicle; then, the system automatically matches and calls the corresponding diagnostic strategy according to the fault phenomenon to conduct a step-by-step in-depth troubleshooting; finally, the diagnostic strategy is continuously optimized and adjusted through the adaptive learning algorithm until the root cause of the fault is accurately located. The diagnostic results and feedback information generated during the whole process will be transmitted to the relevant personnel in real time for timely processing.

[0126] In summary, this strategy emphasizes the collaborative work between different systems, fully considering the mutual influence and dependency between systems during the diagnosis process. Through cross-system data sharing and collaborative analysis, a comprehensive diagnosis of complex faults can be achieved.

[0127] On the advantage, this diagnostic strategy can better cope with complex fault conditions involving multiple systems and improve the comprehensiveness and accuracy of diagnosis.

[0128] Embodiment 2

[0129] This embodiment discloses a smart car automatic driving fault diagnosis system, including:

[0130] The data acquisition and fault determination module is configured to: acquire the operating status data of the vehicle components to be monitored and determine the fault phenomenon;

[0131] The diagnostic strategy library construction module is configured to: construct a diagnostic strategy library, wherein the diagnostic strategy library includes multiple levels of diagnostic strategies, and each level of diagnostic strategy includes corresponding diagnostic processes, judgment rules and processing measures;

[0132] The matching module is configured to: match the diagnosis strategy that best matches the current fault phenomenon from multiple levels of diagnosis strategies based on the fault phenomenon, and execute the processing measures in the most consistent diagnosis strategy;

[0133] The feedback module is configured to: obtain feedback data after the processing is completed;

[0134] The diagnosis strategy update module is configured to update and optimize the diagnosis strategy in the diagnosis strategy library based on the feedback data.

[0135] Embodiment 3

[0136] The purpose of this embodiment is to provide a computer-readable storage medium.

[0137] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the intelligent vehicle automatic driving fault diagnosis method as described in Example 1 of the present disclosure.

[0138] Embodiment 4

[0139] The purpose of this embodiment is to provide an electronic device.

[0140] An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps in the intelligent vehicle automatic driving fault diagnosis method as described in Example 1 of the present disclosure are implemented.

[0141] The steps involved in the apparatuses of the above embodiments 2, 3 and 4 correspond to the method embodiment 1, and the specific implementation methods can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.

[0142] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0143] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A method for diagnosing faults in automatic driving of an intelligent vehicle, characterized in that: The following steps are involved: Obtain the operating status data of the vehicle components to be monitored and determine the fault phenomenon; Constructing a diagnostic strategy library, wherein the diagnostic strategy library includes multiple levels of diagnostic strategies, and each level of diagnostic strategy includes corresponding diagnostic processes, judgment rules and processing measures; Based on the fault phenomenon and combined with the diagnostic strategy library, the diagnostic strategy that best matches the current fault phenomenon is matched from multiple levels of diagnostic strategies, and the treatment measures in the most consistent diagnostic strategy are executed; Get feedback data after processing; Based on the feedback data, the diagnostic strategies in the diagnostic strategy library are updated and optimized.

2. The intelligent vehicle automatic driving fault diagnosis method according to claim 1, characterized in that: Obtain the operating status data of the vehicle components to be monitored, specifically: Build a distributed monitoring network consisting of multiple sentinel nodes; Arrange multiple sentinel nodes on key systems and components of the vehicle to collect vehicle operation status data in real time, each of the sentinel nodes includes multiple sensors and has data processing capabilities; The multiple sentinel nodes are connected to each other via a communication network.

3. The intelligent vehicle automatic driving fault diagnosis method according to claim 1, characterized in that: The multiple levels of diagnostic strategies are formulated based on the principles of the vehicle system, historical fault data and expert experience. The multiple levels of diagnostic strategies specifically include primary diagnostic strategies, intermediate diagnostic strategies and advanced diagnostic strategies. The primary diagnostic strategies, intermediate diagnostic strategies and advanced diagnostic strategies gradually narrow the scope of faults.

4. The intelligent vehicle automatic driving fault diagnosis method according to claim 3, characterized in that: The primary diagnosis network, intermediate diagnosis network and advanced diagnosis network are constructed using preset rules. The primary diagnosis rules include the correspondence between fault phenomena and fault types; the intermediate diagnosis rules include the correspondence between rough fault types and detailed fault types; Advanced diagnostic rules include detailed fault types and corresponding fault descriptions.

5. The intelligent vehicle automatic driving fault diagnosis method according to claim 1, characterized in that: Determine the fault phenomenon by: The current sentinel node aggregates the vehicle operation status data collected by itself and the vehicle operation status data transmitted by other sentinel nodes to construct an operation data set; Determine the current fault phenomenon based on the constructed operation data set and fault diagnosis network; Among them, for image data, the fault diagnosis network adopts convolutional neural network, and for time series data, the fault diagnosis network uses recurrent neural network and its variant long short-term memory network.

6. The intelligent vehicle automatic driving fault diagnosis method according to claim 1, characterized in that: Get feedback data after processing, including: Confirm the cause of the fault phenomenon through further data collection, testing or simulation; After processing according to the treatment measures in the diagnostic strategy that best matches the current fault phenomenon and is most effective, it is determined whether the fault phenomenon has been eliminated, thereby determining the effectiveness of the selected treatment measures.

7. The intelligent vehicle automatic driving fault diagnosis method according to claim 1, characterized in that: Based on the feedback data, the diagnostic strategies in the diagnostic strategy library are updated and optimized, including: Based on the adaptive learning mechanism, it continuously learns from feedback data, new fault cases and diagnostic experience; When encountering new fault types or the diagnostic strategy is not effective, the diagnostic parameters and processes are automatically adjusted to update the diagnostic strategy library.

8. Intelligent car automatic driving fault diagnosis method system, characterized in that: include: The data acquisition and fault determination module is configured to: acquire the operating status data of the vehicle components to be monitored and determine the fault phenomenon; The diagnostic strategy library construction module is configured to: construct a diagnostic strategy library, wherein the diagnostic strategy library includes multiple levels of diagnostic strategies, and each level of diagnostic strategy includes corresponding diagnostic processes, judgment rules and processing measures; The matching module is configured to: match the diagnosis strategy that best matches the current fault phenomenon from multiple levels of diagnosis strategies based on the fault phenomenon, and execute the processing measures in the most consistent diagnosis strategy; The feedback module is configured to: obtain feedback data after the processing is completed; The diagnosis strategy update module is configured to update and optimize the diagnosis strategy in the diagnosis strategy library based on the feedback data.

9. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, the steps in the intelligent vehicle automatic driving fault diagnosis method as described in any one of claims 1 to 7 are implemented.

10. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the intelligent vehicle automatic driving fault diagnosis method as described in any one of claims 1-7 are implemented.

Images