Fault detection method and device, electronic equipment and storage medium

Abstract

The invention provides a fault detection method and device, electronic equipment and a storage medium, which are applied to an automatic driving system. According to the fault detection method, the diagnosis unit deployed according to the structure of the automatic driving system is used to obtain the first diagnosis result of each module or assembly in the automatic driving system, and then the dependency relationship data of the automatic driving system is used to obtain the dependency relationship analysis result of each first diagnosis result; and finally obtaining a second diagnosis result indicating the overall state of the automatic driving system in combination with each first diagnosis result and the dependency analysis result. According to the invention, comprehensive fault monitoring and accurate fault source positioning of the complex automatic driving system can be realized.

Description

Technical Field

[0001] The present disclosure relates to the field of autonomous driving technology, and in particular to a fault detection method, device, electronic device and storage medium. Background Art

[0002] Autonomous driving technology is widely used in logistics, public transportation, ports, urban roads, low-altitude economy and other fields. Fault detection of autonomous driving systems is a key technology to ensure the safe operation of vehicles. Fault detection can not only identify potential faults in advance, improve the reliability and user experience of autonomous driving systems, but also reduce maintenance costs and support the application of autonomous driving technology in complex scenarios. Fault detection is crucial to the safety and reliability of autonomous driving systems.

[0003] However, since the autonomous driving system is composed of multiple heterogeneous components, which have diversity in functions, technologies and data processing, the fault diagnosis complexity of the autonomous driving system is relatively high. The fault diagnosis methods of related technologies are difficult to fully cover the complex autonomous driving system and cannot accurately identify and locate the faults of the autonomous driving system. Summary of the invention

[0004] In view of this, the present disclosure provides a fault detection method, device, electronic device and storage medium.

[0005] According to a first aspect of the present disclosure, there is provided a fault detection method, which is applied to an autonomous driving system, wherein the autonomous driving system comprises a plurality of modules, each of which comprises one or more components, and each module in the autonomous driving system is respectively configured with a diagnostic unit, the method comprising:

[0006] respectively obtaining, by each of the diagnostic units, a first diagnostic result of a corresponding module or a corresponding component in the automatic driving system, wherein the first diagnostic result is used to indicate a state of the corresponding module or the corresponding component;

[0007] Obtaining dependency analysis results of each of the first diagnostic results using dependency data of the autonomous driving system, wherein the dependency data includes logical dependencies and / or physical dependencies between modules, between modules and components, and between components in the autonomous driving system;

[0008] A second diagnostic result is generated based on each of the first diagnostic results and the dependency analysis results thereof, and the second diagnostic result is used to indicate an overall state of the automatic driving system.

[0009] In some embodiments of the first aspect of the present disclosure, the first diagnostic result of the corresponding module or corresponding component in the autonomous driving system is obtained by each of the diagnostic units, including: controlling the designated one or more diagnostic units to perform the following processing respectively: monitoring the status and data flow of the corresponding module or corresponding component to obtain status information, and using a preconfigured mapping table to map the status information into a first diagnostic result, wherein the first diagnostic result includes an identifier of the corresponding module or corresponding component and a status mark thereof.

[0010] In some embodiments of the first aspect of the present disclosure, the autonomous driving system includes a perception module, a positioning module, a path planning module, a behavior decision module and a control module; the perception module is configured with a first diagnostic unit, the first diagnostic unit is used to monitor the state and data flow of the perception module or one or more components in the perception module to obtain a first diagnostic result of the perception module or its component; the positioning module is configured with a second diagnostic unit, the second diagnostic unit is used to monitor the state and data flow of the positioning module or one or more components in the positioning module to obtain a first diagnostic result of the positioning module or its component; the path planning module is configured with a third diagnostic unit, the third diagnostic unit is used to monitor the state and data flow of the path planning module or one or more components in the path planning module to obtain a first diagnostic result of the path planning module or its component; the behavior decision module is configured with a fourth diagnostic unit, the fourth diagnostic unit is used to monitor the state and data flow of the behavior decision module or one or more components in the behavior decision module to obtain a first diagnostic result of the behavior decision module or its component; the control module is configured with one or more fifth diagnostic units, each of the fifth diagnostic units is used to monitor the state and data flow of the control module or one or more components in the control module to obtain a first diagnostic result of the control module or its component.

[0011] In some implementations of the first aspect of the present disclosure, the using the dependency data of the autonomous driving system to obtain the dependency analysis results of each of the first diagnostic results includes one or more of the following:

[0012] Traversing dependency data of the autonomous driving system to determine a propagation path of each first diagnostic result, wherein the propagation path of the first diagnostic result includes a component or module that has a direct dependency relationship with a corresponding component or module of the first diagnostic result;

[0013] The dependency data of the autonomous driving system is used to determine the influence scope of each of the first diagnostic results, and the influence scope of the first diagnostic result includes components or modules that have direct and indirect dependencies with the corresponding components or modules of the first diagnostic results.

[0014] In some implementations of the first aspect of the present disclosure, generating a second diagnosis result according to each of the first diagnosis results and the dependency analysis results thereof includes one or more of the following:

[0015] When the module or component corresponding to the first diagnostic result belongs to a preset key subject and its status is marked as an alarm, it is detected that each module or component indicated by the dependency analysis result of the first diagnostic result has a first diagnostic result with a status marked as an alarm, then the status mark of the first diagnostic result is changed to an error;

[0016] marking the overall state of the autonomous driving system as an error when the state of any one or more of the first diagnostic results is marked as an error;

[0017] When the status marks of all the first diagnostic results are normal, the overall status of the automatic driving system is marked as normal.

[0018] In some implementations of the first aspect of the present disclosure, the method further includes: determining a current state of the vehicle; loading a fault detection strategy corresponding to the current state of the vehicle so as to perform fault detection on the automatic driving system according to the fault detection strategy.

[0019] In some implementations of the first aspect of the present disclosure, the method further includes: feeding back the first diagnostic result and the second diagnostic result to an operator or an external monitoring system.

[0020] According to a second aspect of the present disclosure, there is provided a fault detection device, the fault detection device being applied to an autonomous driving system, the autonomous driving system comprising a plurality of modules, each of the modules comprising one or more components, the fault detection device comprising: a dependency analysis unit, a generation unit, and a plurality of diagnosis units deployed according to the structure of the autonomous driving system;

[0021] Each of the diagnostic units is used to obtain a first diagnostic result of a corresponding module or a corresponding component in the automatic driving system, where the first diagnostic result is used to indicate a state of the corresponding module or the corresponding component;

[0022] The dependency analysis unit is configured to obtain a dependency analysis result of each of the first diagnostic results using dependency data of the autonomous driving system, wherein the dependency data includes logical dependencies and / or physical dependencies between modules, between modules and components, and between components in the autonomous driving system;

[0023] The generating unit is used to generate a second diagnostic result based on each of the first diagnostic results and the dependency analysis results thereof, wherein the second diagnostic result is used to indicate an overall state of the automatic driving system.

[0024] According to a third aspect of the present disclosure, an electronic device is provided, comprising: one or more processors and a memory storing a program, wherein the program comprises instructions, and when the instructions are executed by the processor, the processor executes the above method.

[0025] According to a fourth aspect of the present disclosure, a computer-readable storage medium storing a program is provided, wherein the program includes instructions, which, when executed by one or more processors of a computing device, cause the computing device to execute the above method.

[0026] It can be seen from the above technical solutions that the disclosed embodiment monitors the status of each part of the autonomous driving system through a modularly designed diagnostic unit, and then accurately determines the overall status of the autonomous driving system in combination with the dependency data of the autonomous driving system. While improving the accuracy of fault identification and positioning of the autonomous driving system, it can accurately locate the source of the fault, effectively reduce misjudgment, missed detection, etc., has good scalability, and can flexibly respond to the complex architecture of the autonomous driving system and its dynamic changes. It has been verified by experiments that the disclosed embodiment can complete the diagnosis within a few milliseconds after the autonomous driving system fails, with fast response and good real-time performance, and can meet the real-time requirements of the autonomous driving system. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0028] Figure 1 An example diagram of the autonomous driving system and the dependencies between its modules involved in the present disclosure;

[0029] Figure 2 A flowchart of a fault detection method provided by an embodiment of the present disclosure;

[0030] Figure 3 A schematic diagram of the deployment of the diagnostic unit involved in the embodiment of the present disclosure;

[0031] Figure 4 A schematic diagram of the deployment of a diagnostic unit of a perception module in an autonomous driving system according to the present disclosure;

[0032] Figure 5 A schematic diagram of the structure of a fault detection device provided in an embodiment of the present disclosure;

[0033] Figure 6 A schematic structural block diagram of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0035] The terms used in the embodiments of the present disclosure are only for the purpose of describing specific embodiments, and are not intended to limit the present disclosure. The singular forms "a", "said" and "the" used in the embodiments of the present disclosure and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0036] As used herein, the words "if," "if," and the like may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)," depending on the context.

[0037] As mentioned above, the current fault detection scheme for autonomous driving systems is difficult to fully cover the entire autonomous driving system, and is prone to misjudgment and missed detection. It is unable to accurately identify faults and locate the source of faults. There are also problems such as insufficient real-time, flexibility and adaptability. Not only can it not meet the real-time requirements of the autonomous driving system, but it can also not adapt to the reality of the continuous changes in the autonomous driving system.

[0038] In view of this, the embodiments of the present disclosure provide the following fault detection method, device, electronic device and storage medium, which configure the diagnostic units in a modular design in advance according to the architecture of the autonomous driving system, use these diagnostic units to monitor and obtain the first diagnostic results of each part of the autonomous driving system respectively, use the dependency data of the autonomous driving system to obtain the dependency analysis results of each first diagnostic result, and combine all the first diagnostic results and their dependency analysis results to generate a second diagnostic result indicating the state of the entire autonomous driving system. The embodiments of the present disclosure monitor the state of each part of the autonomous driving system through the modularly designed diagnostic unit, and then accurately judge the overall state of the autonomous driving system in combination with the dependency data of the autonomous driving system. While improving the accuracy of fault identification and positioning of the autonomous driving system, it can effectively reduce misjudgment, missed detection, etc., has better scalability and customizability, and is easy to maintain, and can flexibly adapt to autonomous driving systems of various architectures. In addition, it has been verified by experiments that the embodiments of the present disclosure can complete the diagnosis and trigger corresponding measures within a few milliseconds after the autonomous driving system fails, with fast response and good real-time performance, which can meet the real-time requirements of the autonomous driving system.

[0039] The embodiments of the present disclosure can be applied to various scenarios such as ports, highways, mines, farms, closed parks, urban transportation, and the wild, and can be applied to logistics distribution, unmanned transportation, terminal distribution, car travel, automated agricultural operations, automated sanitation, and many other aspects. Of course, the embodiments of the present disclosure can also be applied to any other autonomous driving scenarios involving equipment such as vehicles. The present disclosure does not limit the application scenarios and applicable fields of the embodiments of the present disclosure.

[0040] The disclosed embodiments can be applied to various types of equipment such as multiple wheeled mobile robots, wheeled mobile robots, mobile robots, vehicles, aircraft, ships, and intelligent rail rapid transit systems (ART, Autonomous rail Rapid Transit). Vehicles can be, but are not limited to, passenger cars, commercial vehicles (e.g., trucks, buses, vans, etc.), special-purpose vehicles (e.g., ambulances, fire trucks, engineering vehicles, rescue vehicles, etc.), agricultural and industrial vehicles (e.g., harvesters, forklifts, etc.), transportation and logistics vehicles (e.g., container trucks, refrigerated trucks, etc.), new energy vehicles (e.g., electric vehicles, hybrid electric vehicles), and special vehicles (e.g., garbage trucks, sprinkler trucks, etc.). In other words, the "vehicle" in the disclosed embodiments is equivalent to the aforementioned various equipment.

[0041] To facilitate understanding, the autonomous driving system involved in the embodiments of the present disclosure is first exemplarily described below.

[0042] Figure 1 shows an exemplary structure of an autonomous driving system and the dependencies between its modules. Figure 1The single arrow connecting line indicates the dependency relationship between modules.

[0043] See also Figure 1 , the autonomous driving system may include: perception module, positioning module, path planning module, behavior decision module, and control module. The positioning module, path planning module, and behavior decision module all depend on the perception module, the path planning module also depends on the positioning module, the behavior decision module also depends on the path planning module, and the control module depends on the behavior decision module.

[0044] Specifically, the perception module is responsible for collecting environmental information around the vehicle, which may include but is not limited to relevant information on road conditions, traffic signs, obstacles, pedestrians, traffic rules, etc. The perception module can provide environmental information for the positioning module, path planning module, and behavior decision module. The positioning module determines the real-time position of the vehicle through the environmental information provided by the perception module. The path planning module plans the driving path through the real-time position provided by the positioning module and the environmental information provided by the perception module. The behavior decision module is used to analyze the environmental information provided by the perception module and the driving path provided by the path planning module to generate a driving strategy. The control module performs specific driving operations according to the driving strategy provided by the behavior decision module.

[0045] Each module in the autonomous driving system may include one or more components. That is, the perception module, positioning module, path planning module, behavior decision module, and control module may each include one or more components. In specific applications, the component structure of each module of the autonomous driving system may be divided according to its function, application scenario, physical architecture, data flow, and / or logical processing flow, etc. These components may be, but are not limited to, software, hardware, or a combination thereof.

[0046] The component structure of each module of the autonomous driving system is illustrated below by way of example.

[0047] The perception module may include, but is not limited to, one or more of the following components: sensor component, data preprocessing component, target detection component, environment modeling component, multi-sensor fusion component, etc. Among them, the sensor component may include components corresponding to various types of sensors, for example, it may include: camera component, laser radar component, millimeter wave radar component, ultrasonic radar component, etc., these components can be used to obtain the raw data collected by the corresponding sensor; the data preprocessing component can be used to perform processing such as cleaning and calibration on the raw data collected by each sensor; the target detection component can be used to identify and classify objects in the environment using the raw data collected by each sensor; the environment modeling component can be used to construct a digital representation of the environment; the multi-sensor fusion component can be used to fuse the data of multiple sensors to improve the accuracy and reliability of perception.

[0048] In specific applications, the multi-sensor fusion component in the perception module can also be set in the positioning module or set as an independent module as needed.

[0049] The positioning module is responsible for determining the real-time position of the vehicle through technologies such as high-precision maps, GPS and / or IMU. Exemplarily, the positioning module may include one or more of the following components: satellite positioning system (GNSS), real-time kinematic measurement (RTK), inertial measurement unit (IMU), map matching and visual positioning.

[0050] The path planning module is responsible for planning the driving path according to the target location and environmental information. Exemplarily, the path planning module may include, but is not limited to, one or more of the following components: global path planning, local path planning, path smoothing and optimization, map management, path verification and safety check, and interface components. Through the close collaboration of various components, the path planning module can generate a safe, efficient, and feasible driving path.

[0051] The driving behavior decision module may include but is not limited to one or more of the following components: behavior prediction, rule engine, scenario analysis, behavior selection, behavior optimization, interaction management, dynamic adjustment, behavior feedback, etc. Through the collaborative work of multiple components, the behavior decision module can achieve rapid response and decision-making in complex driving environments.

[0052] The control module may include but is not limited to one or more of the following components: longitudinal control, lateral control. Alternatively, the control module may include but is not limited to one or more of the following components: throttle control, brake control, dynamics model, steering module, path tracking.

[0053] It should be noted that the components of the aforementioned autonomous driving system and their dependencies are only examples. Those skilled in the art should understand that the components of the autonomous driving system and the dependencies between its internal components can be flexibly adjusted as needed for different specific application scenarios and application requirements, and are not limited to the above implementation methods.

[0054] In the disclosed embodiment, each module in the autonomous driving system is respectively configured with a diagnostic unit, and each diagnostic unit can be used to monitor the state and / or data flow of one or more modules or one or more components in a module in the autonomous driving system to obtain a first diagnostic result. For example, a diagnostic unit can be set for each module in the autonomous driving system, and the diagnostic unit can be used to monitor the state and data flow of the corresponding module. For another example, the diagnostic unit can be deployed for one or more modules in the autonomous driving system according to their component structure, for example, each component is configured with a diagnostic unit, and the diagnostic unit can be used to monitor the state and data flow of one or more components.

[0055] Figure 2 Shows Figure 1An exemplary deployment diagram of a diagnostic unit in an autonomous driving system is shown, Figure 3 An example diagram showing the deployment of diagnostic units for components within a perception module in an autonomous driving system.

[0056] See also Figure 2 The perception module is configured with a first diagnostic unit, which is used to monitor the state and data flow of the perception module or one or more components in the perception module to obtain a first diagnostic result of the perception module or its components; the positioning module is configured with a second diagnostic unit, which is used to monitor the state and data flow of the positioning module or one or more components in the positioning module to obtain a first diagnostic result of the positioning module or its components; the path planning module is configured with a third diagnostic unit, which is used to monitor the state and data flow of the path planning module or one or more components in the path planning module to obtain a first diagnostic result of the path planning module or its components; the behavior decision module is configured with a fourth diagnostic unit, which is used to monitor the state and data flow of the behavior decision module or one or more components in the behavior decision module to obtain a first diagnostic result of the behavior decision module or its components; the control module is configured with a fifth diagnostic unit, which is used to monitor the state and data flow of the control module or one or more components in the control module to obtain a first diagnostic result of the control module or its components.

[0057] See also Figure 3 All sensor components in the perception module share a first diagnostic unit, while the data preprocessing component, target detection component, environment modeling component, multi-sensor fusion component, and data output interface are respectively configured with independent first diagnostic units.

[0058] Other modules in the autonomous driving system (e.g., positioning module, path planning module, behavior decision module, control module) can also deploy diagnostic units according to their component architectures. The specific implementation method is the same as the above. Figure 3 Similar, no further elaboration.

[0059] Each diagnosis unit (ie, each first diagnosis unit, each second diagnosis unit, each third diagnosis unit, each fourth diagnosis unit, etc.) is independent of each other and does not affect each other.

[0060] Different diagnostic units can communicate through standardized interfaces to improve the overall coordination and consistency of fault detection.

[0061] Therefore, independently deployed diagnostic units can be used to monitor the status and diagnose faults for specific modules or components. With independent configuration and modular design, diagnostic units can be flexibly configured and expanded according to the specific needs and structure of the autonomous driving system, which improves the scalability, customizability and maintainability of the system and can flexibly adapt to the complexity, dynamic changes and system upgrades of the autonomous driving system.

[0062] The following will give a detailed description of the specific implementation manners of the embodiments of the present disclosure.

[0063] Figure 4 The flowchart of the fault detection method provided by the embodiments of the present disclosure is shown. The fault detection method of the embodiments of the present disclosure can be executed by the following electronic device, and the electronic device can be implemented as but not limited to a domain controller installed in a vehicle. Refer to Figure 4 , the fault detection method of the embodiments of the present disclosure may include:

[0064] Step 401, respectively obtain the first diagnosis results of the corresponding modules or corresponding components in the autonomous driving system through each diagnosis unit, and the first diagnosis results are used to indicate the states of the corresponding modules or corresponding components;

[0065] Step 402, obtain the dependency analysis results of the first diagnosis results by using the dependency data of the autonomous driving system, and the dependency data includes the logical dependencies and / or physical dependencies between modules, between modules and components, and between components in the autonomous driving system;

[0066] Step 403, generate the second diagnosis results according to the first diagnosis results and their dependency analysis results, and the second diagnosis results are used to indicate the overall state of the autonomous driving system.

[0067] The following will give an exemplary description of the specific implementation manners of each step.

[0068] In step 401, one or more specified diagnosis units can be controlled to respectively perform the following processing: monitor the states and data streams of the corresponding modules or corresponding components to obtain state information, and map the state information to the first diagnosis results by using a pre-configured mapping table. The first diagnosis results include the identifiers of the corresponding modules or corresponding components and their state tags. By representing the state information of different modules or different components with unified state tags, it is convenient to evaluate and monitor the overall state of the autonomous driving system, and at the same time, it can also provide a clear overview of the overall situation of the autonomous driving system, which is convenient for operators or monitoring systems to perform real-time monitoring and fault handling.

[0069] The state tags can be one of the following: normal (OK), warning (WARNING), error (ERROR), ignore (IGNORE), unknown (UNKNOWN). Among them, OK indicates normal and no intervention is required; WARNING indicates that there are potential problems and does not affect the system operation; ERROR indicates that there are serious problems and need to be processed immediately; IGNORE indicates that the current problem can be ignored and does not affect the system operation; UNKNOWN indicates a situation that the system cannot recognize or process and may require further detection.

[0070] In step 401, a mapping table applicable to each module or each component in the module in the autonomous driving system may be configured, and the mapping table defines the status mark of the module or the component under different conditions. For example, when the sensor data frequency monitored by the diagnostic unit is determined to be lower than the sensor data frequency threshold (for example, 10Hz) through the mapping table of the sensor, the sensor data frequency is marked as WARNING. For another example, when the positioning deviation monitored by the diagnostic unit is determined to exceed the safety range (for example, 0.5 meters) through the mapping table of the positioning module, the positioning deviation of the positioning module is marked as ERROR.

[0071] In some examples, the mapping table can be implemented as a key-value pair table or a rule set, and the mapping table can define conditions such as sensor data frequency threshold, safety range, etc. In other implementations, the aforementioned mapping table can also be replaced by a rule set, a rule engine, etc. This is not limited in the embodiments of the present disclosure.

[0072] The specific implementation of each diagnosis unit is exemplified below.

[0073] In step 401, Figure 3 For example, the first diagnostic unit of the corresponding sensor assembly can obtain the first diagnostic result of the sensor assembly in the following manner: monitor the output data of each sensor (e.g., camera, laser radar, ultrasonic radar, etc.) (e.g., the point cloud data frequency of the laser radar, the image clarity of the camera) to obtain the output data parameter of the sensor (e.g., output data frequency), compare the output data parameter (e.g., the data frequency of the laser radar) with the corresponding preset threshold (e.g., the data frequency threshold of the laser radar) to determine whether the sensor can work normally, thereby obtaining the first diagnostic result indicating the working state of the sensor. Therefore, the first diagnostic unit of the corresponding sensor assembly can timely discover sensor abnormalities such as laser radar data loss and camera image blur, thereby avoiding positioning errors or path planning errors caused by sensor abnormalities.

[0074] Specifically, the first diagnostic unit can be used to monitor the status and data flow of different types of sensors using their applicable diagnostic algorithms. For example, the first diagnostic unit can monitor the status and data flow using algorithms such as image quality assessment, feature matching, positioning accuracy assessment, and data consistency check based on the characteristics of the sensor. For example, in the fault diagnosis of a laser radar, the first diagnostic unit can use a Kalman filter algorithm to perform real-time filtering and state estimation on the point cloud data output by the laser radar. When there is a large deviation between the filtered data and the expected normal data, it is determined that the laser radar state is abnormal (for example, fault, data loss, excessive noise, etc.), and a first diagnostic result indicating that the laser radar state is abnormal is generated.

[0075] In step 401, Figure 2 For example, the second diagnostic unit can be used to uniformly perform status monitoring and fault diagnosis on the output data of the positioning module, the positioning algorithm, the working status of each component, and other aspects. Exemplarily, the second diagnostic unit can be used to perform the following multiple tasks: 1) The second diagnostic unit detects whether there is a positioning deviation by comparing the positioning results of different positioning components (for example, GPS, IMU), and determines whether the positioning deviation is within a safe range to obtain a first diagnostic result indicating the current status of the positioning function; 2) The second diagnostic unit can be used to analyze the output results of the positioning algorithm (for example, positioning accuracy and data consistency) to determine whether the positioning module can work normally to obtain a first diagnostic result indicating the working status of the positioning module; 3) In the vehicle's GPS positioning fault diagnosis, the second diagnostic unit can be used to process the GPS signal using a particle filtering algorithm, and update and resample the particle weights. , estimate the accurate position and speed of the vehicle. When there is a large deviation between the estimated result and the actual position or the weight distribution of the particles is abnormal, it is judged that the GPS positioning system may have a fault and generate a corresponding first diagnostic result; 4) The second diagnostic unit evaluates the positioning accuracy by comparing the deviation between the GPS positioning result and the known reference point, thereby detecting the multipath effect of the GPS signal, evaluating the signal quality of the GPS component, and generating a first diagnostic result indicating the current state of the GPS component; 5) The second diagnostic unit can also evaluate the health status of the IMU by checking the consistency of the accelerometer and gyroscope data, analyze the noise level of the IMU data to detect abnormal noise, and generate a first diagnostic result indicating the current state of the IMU component.

[0076] In step 401, Figure 2 For example, the third diagnostic unit can be used to uniformly monitor the status and diagnose faults in multiple aspects such as the output data of the path planning module, the path planning algorithm, and the working status of each component. For example, the third diagnostic unit can be used to monitor the planned path output by the path planning algorithm, and judge whether the status of the path planning algorithm is normal by evaluating whether the planned path is reasonable (for example, checking whether the planned path complies with traffic rules and safety requirements, etc.), and generate a first diagnostic result indicating the current status of the path planning algorithm. As a result, the third diagnostic module can timely discover problems that may be generated by the path planning module, such as unreasonable planning of the planned path passing through obstacles, thereby improving the safety of autonomous driving and further avoiding the occurrence of traffic accidents.

[0077] Specifically, the third diagnostic unit can be used to: use a rapidly exploring random tree (RRT) algorithm to generate a feasible path from the starting point to the end point, and judge whether the path planning algorithm is reasonable by evaluating the smoothness, safety and compliance with traffic regulations of the path; if the generated path has unreasonable situations such as frequent turns, proximity to obstacles or violation of traffic regulations, it is determined that the path planning algorithm may have a fault and generate a first diagnostic result indicating that the path planning algorithm may have a fault.

[0078] Similarly, other diagnostic units can respectively use algorithms applicable to their corresponding modules or components to monitor their status, output data, internal algorithms, etc. and perform fault diagnosis. The specific implementation process of each diagnostic unit obtaining the first diagnostic result is not limited in the disclosed embodiment.

[0079] Before step 402, the disclosed embodiment may further include: constructing a data structure for describing the logical dependencies and / or physical dependencies between modules, between modules and components, and between components in the autonomous driving system. The data structure is the dependency data of the autonomous driving system.

[0080] Logical dependency means that the normal operation of one subject depends on the output or state of another subject, and physical dependency means that the physical connection or data flow of one subject depends on another subject. The subject here refers to a module or a component in the autonomous driving system. For example, the path planning module depends on the environmental information provided by the perception module, that is, there is a dependency relationship between the path planning module and the perception module.

[0081] In the disclosed embodiments, the data structure of the dependency data may be, but is not limited to, a tree structure, a directed graph, or other types of relationship data. Considering that the composition architecture of the autonomous driving system is relatively complex, for example, the perception module and the positioning module both rely on sensor data, while the path planning module relies on both the output of the perception module and the output of the positioning module. Therefore, in some embodiments, the dependency data may use a directed graph, in which each node in the directed graph represents a subject (i.e., a module in the autonomous driving system or a component in a module), and each directed edge of the directed graph represents a dependency relationship from one subject to another. Since directed graphs are suitable for complex dependencies, they can represent the mutual dependence between multiple subjects. Therefore, using a directed graph as dependency data can accurately and comprehensively reflect the relationship between the various parts of the autonomous driving system.

[0082] In some examples, dependency data can be represented as an adjacency matrix. An adjacency matrix is ​​a two-dimensional array that can be used to represent the connection relationship between nodes in a directed graph. The rows and columns of the adjacency matrix represent the nodes in the directed graph, and the elements in the adjacency matrix represent the dependency relationship between the nodes. For example, an element of the adjacency matrix can be represented as adj[i][j]. The value of this element can be used to indicate whether there is a dependency relationship between node i and node j. If there is a dependency relationship, the value of adj[i][j] is 1, otherwise the value of adj[i][j] is 0. For example, the directed edge from the perception module to the positioning module in the directed graph represents that the positioning module depends on the data of the perception module, and the value of the element corresponding to the directed edge from the perception module to the positioning module in the adjacency matrix is ​​1.

[0083] Exemplarily, when only the dependencies between modules in the autonomous driving system are considered, the dependency data of the autonomous driving system can be expressed as the following adjacency matrix adj_matrix.

[0084] adj_matrix = {

[0085] "Sensor":{"Sensor":0,"Localization":1,"Planning":1,"Control":0},

[0086] "Localization":{"Sensor":0,"Localization":0,"Planning":1,"Control":0},

[0087] "Planning":{"Sensor":0,"Localization":0,"Planning":0,"Control":1},

[0088] "Control":{"Sensor":0,"Localization":0,"Planning":0,"Control":0}}

[0089] Among them, Sensor represents the perception module, Localization represents the positioning module, Planning represents the planning decision module (that is, the module formed by merging the aforementioned path planning module and behavior decision module), and Control represents the control module.

[0090] Furthermore, the method of the embodiment of the present disclosure may also include: updating the dependency data of the autonomous driving system in real time by acquiring topological structure information of the autonomous driving system when the architecture of the autonomous driving system changes.

[0091] In step 402, the dependency analysis result of the first diagnostic result may include information about modules or components in the autonomous driving system that have a dependency relationship with the module or component corresponding to the first diagnostic result. In addition, the dependency analysis result may also include attribute information of the dependency, and the attribute information of the dependency is used to indicate whether the dependency is a direct dependency or an indirect dependency.

[0092] Specifically, step 402 may include one or two of the following: 1) traversing the dependency data of the autonomous driving system to determine the propagation path of each first diagnostic result, the propagation path of the first diagnostic result includes components or modules that have direct dependencies with the corresponding components or modules of the first diagnostic result; 2) using the dependency data of the autonomous driving system to determine the impact range of each first diagnostic result, the impact range of the first diagnostic result includes components or modules that have direct dependencies and indirect dependencies with the corresponding components or modules of the first diagnostic result. It can be seen that in the fault detection process of the autonomous driving system, the use of dependency analysis can quickly identify the propagation path and impact range of the fault, thereby accurately locating the root cause of the fault, avoiding misdiagnosis and missed diagnosis, and improving the accuracy and efficiency of fault detection.

[0093] In a specific application, in step 402, the aforementioned propagation path implementation method may be preferentially adopted. If the propagation path of the corresponding module or corresponding component of the first diagnostic result is short or its propagation path is not found, the aforementioned influence range implementation method may be adopted. Of course, both the propagation path and the influence range may be adopted at the same time as required. This is not limited in the embodiments of the present disclosure.

[0094] In some embodiments, the specific implementation process of traversing the dependency data of the autonomous driving system to determine the propagation path of each first diagnostic result may include: if the status in the first diagnostic result is marked as WARNING or ERROR, use a stack to implement depth-first search (DFS), starting from the corresponding module or corresponding component of the first diagnostic result, gradually find all other subjects (i.e., modules or components) that directly depend on the module or component in the dependency data and update their status to "WARNING". If a subject has been marked as affected, skip the subject to avoid repeated processing. In this way, a set can be obtained, which contains the names of all modules or components that have a direct dependency relationship with the corresponding module or corresponding component of the first diagnostic result.

[0095] In some embodiments, the specific implementation process of using the dependency data of the autonomous driving system to determine the impact scope of each first diagnostic result is similar to the specific implementation process of the aforementioned propagation path. The difference is that all other entities that depend on (including direct and indirect dependencies) the module or component are gradually found in the dependency data. Thus, a set can be obtained, which contains the names of all modules or components that may be affected by the corresponding module or corresponding component of the first diagnostic result.

[0096] As described above, when a module or a component in a module in the autonomous driving system fails, the dependencies can be used to quickly and accurately infer other modules or components that may be affected, thereby quickly locating the source of the failure.

[0097] In step 403, all first diagnosis results may be logically aggregated according to the dependency analysis results of the first diagnosis results to obtain a second diagnosis result. Here, logical aggregation may be implemented using a preconfigured aggregation strategy and methods such as logical operations (such as OR, AND operations), weighted summation, etc.

[0098] In some implementations, a fault detection strategy may be preconfigured, and the fault detection strategy may be used to implement logical aggregation. In step 403, logical aggregation may be performed on each first diagnosis result according to the fault detection strategy to obtain a second diagnosis result.

[0099] In some implementations, step 403 may include, but is not limited to, one or more of the following:

[0100] 1) When the corresponding module or component of the first diagnostic result belongs to the preset key subject and the status is marked as warning, it is detected that each module or component indicated by the dependency analysis result of the first diagnostic result has the first diagnostic result marked as warning, then the status mark of the first diagnostic result is changed to error; key subjects refer to those modules or components that have a greater impact on the overall function of the autonomous driving system and may immediately cause the autonomous driving system to be unable to operate safely once a failure occurs. The key subject can be set in advance or flexibly determined according to the current state of the vehicle. Therefore, it is possible to determine whether the first diagnostic result of WARNING needs to be upgraded to ERROR in combination with the dependency analysis, thereby avoiding false alarms and missed diagnoses and improving the accuracy of fault diagnosis.

[0101] 2) When the status of any one or more first diagnostic results is marked as ERROR, the overall status of the autonomous driving system is marked as ERROR; that is, once the status of any first diagnostic result is marked as ERROR, an OR operation (i.e., an OR operation) is performed on each first diagnostic result to mark the overall status of the autonomous driving system as ERROR. Thus, the OR operation can be used to aggregate each first diagnostic result to determine that the overall status of the autonomous driving system is ERROR when any module reports ERROR.

[0102] 3) When the status marks of all the first diagnostic results are OK, the overall status of the autonomous driving system is marked as normal. That is, when the status marks of all the first diagnostic results are OK, an AND operation is performed on each first diagnostic result to mark the overall status of the autonomous driving system as OK. Thus, it can be determined that the overall status of the autonomous driving system is OK when all modules report OK.

[0103] In some examples, the specific implementation process of item 1) can be: traverse each first diagnostic result to diagnose each first diagnostic result and perform the following processing: if the status mark of the corresponding module or corresponding component of the first diagnostic result is "WARNING", then the dependency analysis result of the first diagnostic result obtains all the subjects that the module or the component depends on (that is, other modules or other components), and uses the all() function to check whether all the subjects that the module or the component depends on are "WARNING". If the status marks of all the subjects that the module or the component depends on are "WARNING", then the status mark of the current first diagnostic result is updated to "ERROR"; otherwise, the status mark of the current first diagnostic result is kept as "WARNING". Therefore, when some modules or some components report WARNING, the AND operation combined with the dependency analysis results of each first diagnostic result can be used to determine whether WARNING needs to be upgraded to ERROR, thereby further improving the overall safety and reliability of the autonomous driving system.

[0104] Furthermore, step 403 may also include: ignoring the first diagnosis result that the corresponding module or the corresponding component belongs to a preset non-critical subject and the status is marked as a warning. Non-critical subjects refer to those modules or components that have little impact on the overall function of the autonomous driving system and will not immediately cause the autonomous driving system to be unable to operate safely even if a fault occurs. Therefore, by ignoring the WARNING status of non-critical subjects, we can better focus on critical subjects and further improve the fault diagnosis efficiency of the autonomous driving system.

[0105] In some examples, non-critical entities can be flexibly determined based on the current state of the vehicle. For example, a path planning module can be considered a non-critical entity when the vehicle is parked.

[0106] In some examples, non-critical subjects can be pre-set. For example, one or more of the following can be set as non-critical subjects: auxiliary function components such as automatic wipers and automatic headlights, non-real-time data processing components such as log recording and data backup, and user interface components such as central control screen display and voice assistant.

[0107] In step 403, the second diagnostic result may include, but is not limited to, a status mark (OK or ERROR) for indicating the overall status of the autonomous driving system, and the name of the module or component with the status mark of ERROR. Thus, the second diagnostic result can not only indicate the overall status of the autonomous driving system, but also be used to accurately locate the fault source of the autonomous driving system.

[0108] As can be seen from the above, step 403 can comprehensively consider the first diagnostic results of each diagnostic unit and the dependency analysis results to accurately determine the fault at the autonomous driving system level.

[0109] Furthermore, step 403 may also include: generating a fault diagnosis report, the fault diagnosis report including the first diagnosis result and the second diagnosis result. Thus, the fault source of the autonomous driving system can be accurately located by combining the second diagnosis result with the first diagnosis result, and the fault status of the autonomous driving system and its modules and components can be fully and accurately reflected.

[0110] Furthermore, the fault detection method of the disclosed embodiment may also include: determining the current state of the vehicle, loading a fault detection strategy corresponding to the current state of the vehicle to perform fault detection of the autonomous driving system according to the fault detection strategy. Thus, the fault diagnosis strategy can be dynamically adjusted by sensing the current state of the vehicle to perform fault detection for different working states of the autonomous driving system, and the focus of fault diagnosis can be flexibly adjusted according to different driving states, the allocation of diagnostic resources can be optimized, and the pertinence and effectiveness of fault diagnosis can be improved.

[0111] The current state of the vehicle may be, but is not limited to, starting, waiting, driving, parking, etc.

[0112] In some examples, the current driving state of the vehicle can be identified and determined based on the sensor data and control signals of the vehicle. Specifically, a finite state machine algorithm can be used to predefine different driving states of the vehicle and their transition conditions, and the driving state of the vehicle can be updated in real time according to the transition conditions of the state machine, and the state information indicating the current state of the vehicle can be transmitted to the fault diagnosis system of the vehicle (for example, the fault diagnosis device 500 described below), so that the fault diagnosis system of the vehicle can adjust the fault diagnosis strategy according to the driving state of the vehicle. For example, the transition conditions may include but are not limited to: when the speed of the vehicle is greater than zero and the gear is in the driving gear, it is judged to be in the driving state, when the speed of the vehicle is zero and the gear is in the neutral gear, it is judged to be in the parking state, etc.

[0113] In some examples, the fault detection strategy can be used to generate the second diagnosis result. For example, the fault detection strategy in the parking state can include a strategy of "ignoring the first diagnosis result of the path planning module".

[0114] Furthermore, the fault detection method of the disclosed embodiment may further include: feeding back the first diagnosis result and the second diagnosis result to the operator or the external monitoring system. Thus, a real-time monitoring and feedback mechanism of the autonomous driving system is implemented, so that the operator or the monitoring system can timely understand the operating status and fault conditions of the autonomous driving system, facilitate rapid response and handling of faults, and further shorten the fault response time.

[0115] Specifically, a feedback interface can be provided, through which the second diagnostic result is transmitted to a human-computer interaction interface of the control end or an external monitoring system. The second diagnostic result can be displayed through the human-computer interaction interface of the control end so that the operator can clearly see the overall status of the vehicle's automatic driving system and the status of its various modules or components, or it can facilitate the external monitoring system to promptly understand the status of each vehicle's automatic driving system, facilitate rapid response and handling of faults, thereby further improving the reliability and safety of the automatic driving system.

[0116] For example, when a sensor component of a perception module in an autonomous driving system enters a WARNING state, the operator can be prompted to check by highlighting the sensor component on the human-computer interaction interface.

[0117] It can be seen from the above that the method of the embodiment of the present disclosure realizes comprehensive monitoring, accurate diagnosis and real-time feedback of complex autonomous driving system faults through modular design and multi-level analysis and aggregation, while reducing the possibility of misdiagnosis and missed diagnosis, and improving the accuracy of fault diagnosis, thereby effectively improving the safety and reliability of the autonomous driving system.

[0118] It has been found through experiments that when a sensor component in the perception module of an autonomous driving system fails, the diagnostic unit modularly designed according to the disclosed embodiment can quickly identify the fault and accurately determine through dependency analysis that the positioning module and path planning module may be affected, thereby avoiding misjudgment of other irrelevant modules (e.g., control modules).

[0119] Experiments have shown that the method of the disclosed embodiment can complete the diagnosis and trigger the corresponding countermeasures within a few milliseconds after the failure of the autonomous driving system occurs, and the real-time performance and response speed are significantly improved, which can meet the needs of the autonomous driving system. For example, in the test of the autonomous driving shuttle truck, when the path planning module appears in the ERROR state, the method of the disclosed embodiment can complete the diagnosis of the fault within 10 milliseconds, and immediately take deceleration measures to ensure driving safety.

[0120] Figure 5 FIG. 1 is a schematic diagram showing the structure of a fault detection device provided by an embodiment of the present disclosure. Figure 5 The fault detection device 500 of the embodiment of the present disclosure may include: a plurality of diagnosis units 501, a dependency analysis unit 502 and a generation unit 503 deployed according to the structure of the autonomous driving system.

[0121] Each diagnostic unit 501 is used to obtain a first diagnostic result of a corresponding module or a corresponding component in the automatic driving system, where the first diagnostic result is used to indicate a state of the corresponding module or the corresponding component;

[0122] A dependency analysis unit 502, configured to determine a dependency analysis result of each first diagnosis result using dependency data of the autonomous driving system, wherein the dependency data includes logical dependencies and / or physical dependencies between modules, between modules and components, and between components in the autonomous driving system;

[0123] The generating unit 503 is used to generate a second diagnostic result according to each first diagnostic result and the dependency analysis result thereof, wherein the second diagnostic result is used to indicate the overall state of the automatic driving system.

[0124] Further, regarding the deployment method and number of the diagnostic unit 501, see the above Figure 1 , Figure 2 and Figure 3 The description is not repeated here.

[0125] Furthermore, each diagnostic unit 501 can be specifically used to monitor the status and data flow of the corresponding module or corresponding component to obtain status information, and use a preconfigured mapping table to map the status information into a first diagnostic result, which includes the identification of the corresponding module or corresponding component and its status mark.

[0126] Furthermore, the dependency analysis unit 502 can be specifically used to: traverse the dependency data of the autonomous driving system to determine the propagation path of each first diagnostic result, the propagation path of the first diagnostic result includes components or modules that have a direct dependency relationship with the corresponding component or corresponding module of the first diagnostic result; and / or, use the dependency data of the autonomous driving system to determine the influence scope of each first diagnostic result, the influence scope of the first diagnostic result includes components or modules that have a direct dependency relationship and an indirect dependency relationship with the corresponding component or corresponding module of the first diagnostic result.

[0127] Furthermore, the generation unit 503 can be specifically used for: when the corresponding module or corresponding component of the first diagnostic result belongs to a preset key subject and its status is marked as a warning, it is detected that each module or each component indicated by the dependency analysis result of the first diagnostic result has a first diagnostic result with a status marked as a warning, then the status mark of the first diagnostic result is changed to error; and / or, when the status mark of any one or more first diagnostic results is error, the overall status of the autonomous driving system is marked as error; and / or, when the status marks of all first diagnostic results are normal, the overall status of the autonomous driving system is marked as normal.

[0128] Furthermore, the fault detection device 500 of the embodiment of the present disclosure may also include: a vehicle state determination unit 504 and a strategy loading unit 505, the vehicle state unit 504 is used to determine the current state of the vehicle, and the strategy loading unit 505 is used to load a fault detection strategy corresponding to the current state of the vehicle so that the dependency analysis unit 502, the generation unit 503 and each diagnosis unit 501 perform fault detection on the automatic driving system according to the fault detection strategy.

[0129] Furthermore, the fault detection device 500 of the embodiment of the present disclosure may further include: a feedback unit 506, which is used to feed back the first diagnosis result and the second diagnosis result to an operator or an external monitoring system.

[0130] For other technical details about the fault detection device 500, please refer to the above description about the fault detection method, which will not be repeated here.

[0131] In a specific application, the fault detection device 500 can be implemented by software, hardware or a combination of the two. For example, the fault detection device 500 can be implemented as software running in the electronic device 600 described below.

[0132] In addition, an embodiment of the present disclosure further provides a computer-readable storage medium on which a computer program is stored. The program includes instructions. When the instructions are executed by one or more processors of a computing device, the steps of the aforementioned fault detection method are executed.

[0133] Figure 6 FIG. 1 is a schematic diagram showing the structure of an electronic device provided by an embodiment of the present disclosure. Figure 6 The electronic device 600 may include: one or more processors 601, and also includes a memory 602 storing one or more programs, which are executed by the one or more processors 601 to implement the method flow shown in the above embodiments of the present disclosure and / or program units corresponding to each unit in the device.

[0134] The various components are interconnected using different buses and can be mounted on a common motherboard or otherwise as needed. The processor 601 can process instructions executed within the electronic device, including instructions stored in or on the memory to display graphical information of a user interface on an external input / output device (such as a display device coupled to the interface). In other embodiments, multiple processors and / or multiple buses can be used with multiple memories and multiple memories if desired.

[0135] The processor 601 may include one or more single-core processors or multi-core processors. The processor 601 may include any combination of general-purpose processors or dedicated processors (such as image processors, application processors, baseband processors, etc.).

[0136] The memory 602 is a computer-readable storage medium provided by the present disclosure, which can be used to store non-transient software programs, non-transient computer executable programs and units, such as the following in the embodiments of the present disclosure: Figure 1 The processor 601 executes the non-transient software programs, instructions and units stored in the memory 602, thereby executing the above-mentioned method embodiment. Figure 1 The programs, instructions and units corresponding to the fault detection method shown.

[0137] The electronic device 600 may further include: an input device 603 and an output device 606. The processor 601, the memory 602, the input device 603 and the output device 604 may be connected via a bus or other means. Figure 6 The example of connecting through bus is taken in the following.

[0138] The input device 603 can receive input digital or character information, and generate signal input related to user settings and function control, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick and other input devices. The output device 604 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The display device may include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display and a plasma display. In some embodiments, the display device may be a touch screen.

[0139] The above-mentioned programs (also referred to as software, software applications, or codes) include machine instructions for programmable processors, and these computer programs can be implemented using object-oriented programming languages, assembly or machine languages.

[0140] With the development of time and technology, the meaning of medium is becoming more and more extensive, and the propagation path of computer programs is no longer limited to tangible media, and can also be downloaded directly from the network, etc. Any combination of one or more computer-readable storage media can be used. Computer-readable storage media can be used but not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or devices, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this document, computer-readable storage media can be any tangible medium containing or storing programs, which can be used by or in combination with instruction execution systems, devices or devices.

[0141] In a specific application, the electronic device 600 may be implemented as, but not limited to, a domain controller or other similar devices.

[0142] The technical solution provided by the present disclosure is described in detail above. The principles and implementation methods of the present disclosure are described in detail using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present disclosure. At the same time, for those skilled in the art, according to the idea of ​​the present disclosure, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present disclosure.

[0143] The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure. Any modifications, equivalent substitutions, etc. made within the spirit and principles of the present disclosure should be included in the protection scope of the present disclosure.

Claims

1. A fault detection method, characterized in that: Applied to an autonomous driving system, the autonomous driving system includes a plurality of modules, each of the modules includes one or more components, each module in the autonomous driving system is respectively configured with a diagnostic unit, and the method includes: respectively obtaining, by each of the diagnostic units, a first diagnostic result of a corresponding module or a corresponding component in the automatic driving system, wherein the first diagnostic result is used to indicate a state of the corresponding module or the corresponding component; Obtaining dependency analysis results of each of the first diagnostic results using dependency data of the autonomous driving system, wherein the dependency data includes logical dependencies and / or physical dependencies between modules, between modules and components, and between components in the autonomous driving system; A second diagnostic result is generated based on each of the first diagnostic results and the dependency analysis results thereof, and the second diagnostic result is used to indicate an overall state of the automatic driving system.

2. The method according to claim 1, characterized in that The obtaining, by each of the diagnostic units, first diagnostic results of corresponding modules or components in the automatic driving system respectively includes: Control the designated one or more diagnostic units to respectively perform the following processing: monitor the status and data flow of the corresponding module or the corresponding component to obtain status information, and use a preconfigured mapping table to map the status information into a first diagnostic result, wherein the first diagnostic result includes the identification of the corresponding module or the corresponding component and its status mark.

3. The method according to claim 1 or 2, characterized in that: The autonomous driving system includes a perception module, a positioning module, a path planning module, a behavior decision module and a control module; The perception module is configured with a first diagnostic unit, the first diagnostic unit being used to monitor the state and data flow of the perception module or one or more components in the perception module to obtain a first diagnostic result of the perception module or its components; The positioning module is configured with a second diagnostic unit, the second diagnostic unit being used to monitor the status and data flow of the positioning module or one or more components in the positioning module to obtain a first diagnostic result of the positioning module or its components; The path planning module is configured with a third diagnostic unit, the third diagnostic unit being used to monitor the status and data flow of the path planning module or one or more components in the path planning module to obtain a first diagnostic result of the path planning module or its components; The behavior decision module is configured with a fourth diagnostic unit, the fourth diagnostic unit being used to monitor the state and data flow of the behavior decision module or one or more components in the behavior decision module to obtain a first diagnostic result of the behavior decision module or its components; The control module is configured with one or more fifth diagnostic units, each of which is used to monitor the state and data flow of the control module or one or more components in the control module to obtain a first diagnostic result of the control module or its components.

4. The method according to claim 1, characterized in that The using the dependency data of the autonomous driving system to obtain the dependency analysis results of each of the first diagnostic results includes one or more of the following: Traversing dependency data of the autonomous driving system to determine a propagation path of each first diagnostic result, wherein the propagation path of the first diagnostic result includes a component or module that has a direct dependency relationship with a corresponding component or module of the first diagnostic result; The dependency data of the autonomous driving system is used to determine the influence scope of each of the first diagnostic results, and the influence scope of the first diagnostic result includes components or modules that have direct and indirect dependencies with the corresponding components or modules of the first diagnostic results.

5. The method according to claim 1, characterized in that The generating of the second diagnosis result according to the first diagnosis results and the dependency analysis results thereof includes one or more of the following: When the module or component corresponding to the first diagnostic result belongs to a preset key subject and its status is marked as an alarm, it is detected that each module or component indicated by the dependency analysis result of the first diagnostic result has a first diagnostic result with a status marked as an alarm, then the status mark of the first diagnostic result is changed to an error; marking the overall state of the autonomous driving system as an error when the state of any one or more of the first diagnostic results is marked as an error; When the status marks of all the first diagnostic results are normal, the overall status of the automatic driving system is marked as normal.

6. The method according to claim 1, characterized in that The method further comprises: Determine the current state of the vehicle; A fault detection strategy corresponding to the current state of the vehicle is loaded so as to perform fault detection on the automatic driving system according to the fault detection strategy.

7. The method according to claim 1, characterized in that The method further includes: feeding back the first diagnosis result and the second diagnosis result to an operator or an external monitoring system.

8. A fault detection device, characterized in that: The fault detection device is applied to an autonomous driving system, the autonomous driving system includes a plurality of modules, each of the modules includes one or more components, the fault detection device includes: a dependency analysis unit, a generation unit, and a plurality of diagnosis units deployed according to the structure of the autonomous driving system; Each of the diagnostic units is used to obtain a first diagnostic result of a corresponding module or a corresponding component in the automatic driving system, where the first diagnostic result is used to indicate a state of the corresponding module or the corresponding component; The dependency analysis unit is configured to obtain a dependency analysis result of each of the first diagnostic results using dependency data of the autonomous driving system, wherein the dependency data includes logical dependencies and / or physical dependencies between modules, between modules and components, and between components in the autonomous driving system; The generating unit is used to generate a second diagnostic result based on each of the first diagnostic results and the dependency analysis results thereof, wherein the second diagnostic result is used to indicate an overall state of the automatic driving system.

9. An electronic device, characterized in that: include: One or more processors and a memory storing a program, wherein the program comprises instructions, and when the instructions are executed by the processor, the processor executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a program, wherein the program comprises instructions, and when the instructions are executed by one or more processors of a computing device, the instructions cause the computing device to execute the method according to any one of claims 1 to 7.

Images