Abnormal positioning method and device, storage medium and program product
By using an abnormal positioning system and training model in the digital key system, the log data is automatically analyzed to locate faults, and the problem of inefficient troubleshooting of digital key system is solved, achieving more efficient abnormal positioning.
Patent Information
- Application Number
- CN202510128730.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-27
- Publication Date
- 2025-04-29
AI Technical Summary
Due to its complex multi-end composition, the digital key system is inefficient in troubleshooting and it is difficult to quickly locate the causes of abnormalities.
By obtaining the log data of the car key system and inputting it into the abnormal positioning system, using the trained model for abnormal positioning, the log data is automatically analyzed to locate the fault location and cause.
It improves the efficiency of troubleshooting of digital key systems, reduces the collaboration costs of multi-end R&D personnel, and achieves faster and more accurate abnormal positioning.
Smart Images
Figure CN119904944A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of anomaly location technology, and in particular to anomaly location methods, devices, storage media, and program products. Background Art
[0002] At present, digital keys have been widely used in various car models. Usually, a digital key system may include multiple parts such as car manufacturer cloud services, vehicles, and equipment, which makes it difficult to troubleshoot digital key faults. Summary of the invention
[0003] To overcome the problems existing in the related art, the present disclosure provides an abnormality locating method, device, storage medium and program product.
[0004] According to a first aspect of an embodiment of the present disclosure, there is provided an abnormality locating method, comprising: In response to an abnormal positioning operation by a user, acquiring log data of at least one object in the vehicle key system to obtain first log data; The first log data is input into an abnormality locating system to obtain an abnormality locating result of the vehicle key system output by the abnormality locating system.
[0005] Optionally, the inputting the first log data into an abnormality locating system to obtain an abnormality locating result of the vehicle key system output by the abnormality locating system includes: Inputting the first log data into a first model to obtain an abnormality locating result of the vehicle key system output by the first model; Among them, the anomaly positioning system includes the first model, which is trained by taking sample log data of at least one object of the sample car key system as input and processing results as output, and the processing results include anomaly positioning results of the sample car key system.
[0006] Optionally, the step of inputting the first log data into a first model to obtain an abnormality locating result of the vehicle key system output by the first model includes: Inputting the first log data into the first model, obtaining an abnormality locating result of the vehicle key system output by the first model, and a function performed by the vehicle key system determined according to the first log data; Among them, the first model is trained by taking sample log data of at least one object of the sample car key system as input and processing results as output, and the processing results include the abnormality positioning results of the sample car key system and the functions performed by the sample car key system determined according to the sample log data.
[0007] Optionally, the method comprises: Acquire log data of at least one object of the sample car key system when performing a car key function to obtain second log data; Determine a reference abnormal positioning result, wherein the reference abnormal positioning result includes an abnormal cause and / or an execution result of the sample car key system when executing the car key function, and the execution result is configured to describe whether the car key function is successfully executed; First training data is generated, where the first training data includes the second log data and the reference anomaly location result, and the sample log data includes the first training data.
[0008] Optionally, generating the first training data includes: When the processing result includes the abnormality positioning result of the sample car key system and the function performed by the sample car key system determined according to the sample log data, the first training data is generated, and the first training data includes the second log data, the reference abnormality positioning result, and the car key function.
[0009] Optionally, the acquiring log data of at least one object of the sample car key system when performing a car key function to obtain second log data includes: For each object in the at least one object, obtaining log data of the thread of the object during the period when the object executes the car key function, to obtain third log data; Log data irrelevant to the execution of the vehicle key function is removed from the third log data to obtain the second log data.
[0010] Optionally, in response to the abnormal positioning operation of the user, acquiring log data of at least one object in the vehicle key system to obtain first log data includes: In response to the user's abnormal location operation, determine a time interval for log acquisition; The log data of each object in the vehicle key system within the time interval is obtained to obtain the first log data.
[0011] According to a second aspect of an embodiment of the present disclosure, there is provided an abnormality locating device, comprising: A first module is configured to obtain log data of at least one object in the vehicle key system in response to an abnormal positioning operation of a user, and obtain first log data; The processing module is configured to input the first log data into an abnormality locating system to obtain an abnormality locating result of the vehicle key system output by the abnormality locating system.
[0012] Optionally, the processing module includes: A second module is configured to input the first log data into a first model to obtain an abnormality positioning result of the vehicle key system output by the first model; Among them, the anomaly positioning system includes the first model, which is trained by taking sample log data of at least one object of the sample car key system as input and processing results as output, and the processing results include anomaly positioning results of the sample car key system.
[0013] Optionally, the second module includes: a first submodule, configured to input the first log data into the first model, obtain an abnormality locating result of the vehicle key system output by the first model, and a function performed by the vehicle key system determined according to the first log data; Among them, the first model is trained by taking sample log data of at least one object of the sample car key system as input and processing results as output, and the processing results include the abnormality positioning results of the sample car key system and the functions performed by the sample car key system determined according to the sample log data.
[0014] Optionally, the device comprises: A third module is configured to obtain log data of at least one object of the sample car key system when performing a car key function to obtain second log data; a fourth module, configured to determine a reference abnormal positioning result, wherein the reference abnormal positioning result includes an abnormal cause and / or an execution result of the sample car key system when executing the car key function, and the execution result is configured to describe whether the car key function is successfully executed; The fifth module is configured to generate first training data, wherein the first training data includes the second log data and the reference anomaly location result, and the sample log data includes the first training data.
[0015] Optionally, the fifth module includes: The second submodule is configured to generate the first training data when the processing result includes the abnormality positioning result of the sample car key system and the function performed by the sample car key system determined according to the sample log data, and the first training data includes the second log data, the reference abnormality positioning result, and the car key function.
[0016] Optionally, the third module includes: A third submodule is configured to obtain, for each object in the at least one object, log data of the thread of the object during the period when the object executes the car key function, to obtain third log data; The fourth submodule is configured to remove log data irrelevant to the execution of the vehicle key function from the third log data to obtain the second log data.
[0017] Optionally, the first module includes: A fifth submodule is configured to determine a time interval for log acquisition in response to an abnormality location operation of a user; The sixth submodule is configured to obtain the log data of each object in the vehicle key system within the time interval to obtain the first log data.
[0018] According to a third aspect of an embodiment of the present disclosure, there is provided an abnormality locating device, comprising: processor; a memory for storing processor-executable instructions; The processor is configured to execute the steps of any one of the methods described in the first aspect.
[0019] According to a fourth aspect of an embodiment of the present disclosure, there is provided a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the methods described in the first aspect are implemented.
[0020] According to a fifth aspect of an embodiment of the present disclosure, a computer program product is provided, including a computer program, which implements the steps of any one of the methods in the first aspect when executed by a processor.
[0021] In the above solution, in response to the user's abnormality locating operation, the log data of at least one object in the car key system can be obtained to obtain the first log data. In this way, the first log data can be input into the abnormality locating system to obtain the abnormality locating result of the car key system output by the abnormality locating system.
[0022] In this way, the log data can be processed by the anomaly location system to obtain anomaly location results. Compared with manual analysis of log data and other methods, the above solution has higher data processing efficiency and anomaly location efficiency.
[0023] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.
[0025] Figure 1 It is a schematic diagram of a digital key system architecture according to an exemplary embodiment.
[0026] Figure 2 The invention is a flowchart showing an execution flow of a system function according to an exemplary embodiment.
[0027] Figure 3 The figure is a flow chart of an abnormality locating method according to an exemplary embodiment.
[0028] Figure 4 The figure is a flow chart of an abnormality locating method according to an exemplary embodiment.
[0029] Figure 5 The figure is a block diagram of an abnormality locating device according to an exemplary embodiment.
[0030] Figure 6 It is a block diagram of a device for abnormality location according to an exemplary embodiment. DETAILED DESCRIPTION
[0031] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.
[0032] It should be noted that all actions of acquiring signals, information or data in the present disclosure are carried out in compliance with the relevant data protection laws and policies of the country where the device is located and with the authorization given by the owner of the corresponding device.
[0033] Before introducing the anomaly locating method, device, storage medium, and program product of the present disclosure, the relevant scenarios of the embodiments of the present disclosure are first introduced.
[0034] At present, digital keys have been widely used in various car models. Usually, a digital key system may include multiple parts such as car manufacturer cloud services, vehicles, and equipment, which makes it difficult to troubleshoot digital key faults.
[0035] Figure 1 is a schematic diagram of a digital key system architecture shown in an exemplary embodiment of the present disclosure, referring to Figure 1, the digital key system architecture may include a vehicle manufacturer cloud, vehicles and equipment.
[0036] The car manufacturer cloud can be a collection of cloud services built by car manufacturers in the digital key business system. For example, it may include key certificate sub-service (responsible for managing certificates and secret keys related to the digital key business), key information sub-service (responsible for managing key information and key status related to the digital key business), user account sub-service (responsible for managing user account information and account status related to the digital key business), vehicle communication sub-service (responsible for managing the communication link between the car manufacturer cloud service and the vehicle), and APP (application) communication sub-service (responsible for managing the communication link between the car manufacturer cloud service and the car manufacturer APP).
[0037] The vehicle may be a vehicle-wide device provided by the car manufacturer in the digital key business system. The physical submodules or software submodules that the vehicle may include include: a network connection module (responsible for realizing the network connection function between the vehicle and the cloud), a central computing platform (responsible for the overall business function scheduling of the vehicle), and a digital key business processor (responsible for the execution of the specific digital key business function logic of the vehicle and the completion of wireless communication with the external digital key device).
[0038] The device may be a device provided by the device manufacturer in the digital key business system. For example: mobile phones, watches, etc. The physical submodules or software submodules that the device may include include: operating system (responsible for managing the overall scheduling between various software on the device), car manufacturer APP (responsible for executing the relevant functions provided by the car manufacturer to the device after the APP is installed), Bluetooth or NFC (NearField Communication) or UWB (Ultra-Wide Band) module (responsible for providing the device with wireless communication capabilities to the vehicle), etc.
[0039] Reference Figure 1 Since the digital key system includes the car manufacturer cloud, vehicles and equipment, when an abnormal problem occurs in the digital key system during the research and development, verification or listing stage, it may be necessary for multiple R&D personnel from the car manufacturer cloud, vehicles and equipment to jointly troubleshoot the problem in order to locate the cause of the fault.
[0040] Exemplarily, the R&D personnel of each sub-service of the car manufacturer's cloud service can collect the relevant logs of each sub-service within a certain time range before and after the problem occurs (for example, 3 minutes before and 3 minutes after the problem occurs, a total of 6 minutes). The R&D personnel of each sub-module of the vehicle can collect the relevant logs of each sub-module within a certain time range before and after the problem occurs (for example, 3 minutes before and 3 minutes after the problem occurs, a total of 6 minutes). The R&D personnel of each sub-module of the user device can collect the relevant logs of each sub-module within a certain time range before and after the problem occurs (for example, 3 minutes before and 3 minutes after the problem occurs, a total of 6 minutes). In this way, the R&D personnel at each end can jointly troubleshoot the problem according to the log status of the sub-module or sub-service they are responsible for.
[0041] Figure 2 is an execution flow chart of a system function shown in an exemplary embodiment of the present disclosure, referring to Figure 2 , the process may include: module A calls the b1 interface of module B through a certain communication method, and after executing the relevant software logic, the b1 interface continues to call the c1 interface of module C. After executing the relevant software logic, the c1 interface returns the execution result to module B. After receiving the return of the c1 interface, module B continues to return the execution result to module A.
[0042] Combination Figure 2 For example, when the system function fails, the fault troubleshooting and locating process may include: The A module developer queries the A module's log file to see if there is a record of calling the B module b1 interface. If not, the problem can be located in the A module.
[0043] If module A has a record of calling module B's b1 interface, the B module developer can query the log file of module B to see if there is a record of module A requesting to call b1 interface. If not, the problem can be located in the AB module call communication link.
[0044] If module B has a record of module A requesting to call interface b1, then continue to check whether there is a record of calling interface c1 of module C. If not, the problem can be located in module B.
[0045] If there is a record of module B calling module C's c1 interface, the C module developer can query the C module's log file to determine whether there is a record of module B requesting to call the c1 interface. If not, the problem can be located in the BC module call communication link; If there is a record in module C that module B requests to call the c1 interface, you can continue to check whether there is a record that module C returns the execution result to module B. If not, the problem can be located in module C.
[0046] If there is a record of module C returning the execution result to module B, the B module developer can query the B module log file to check whether there is a record of module C returning the execution result. If not, the problem can be located in the BC module call communication link.
[0047] If module B has a record of module C returning the execution result, you can continue to check whether there is a record of module B returning the execution result to module A. If not, the problem can be located in module B.
[0048] If there is a record of module B returning the execution result to module A, the R&D personnel of module A can check whether there is a record of module B returning the execution result by querying the log file of module A. If not, the problem can be located in the communication link of module AB.
[0049] In this way, R&D personnel from multiple terminals can jointly analyze log data to locate the location of the fault. It is worth noting that the digital key system involves many modules, and the development technology stacks of different modules may have different characteristics, resulting in the need for R&D personnel from multiple terminals to work together to troubleshoot problems. This makes troubleshooting less efficient.
[0050] To this end, the present disclosure provides an abnormality positioning method. Figure 3 is a flowchart of an abnormality location method shown in an exemplary embodiment of the present disclosure, referring to Figure 3 , the method comprising: In step S31 , in response to an abnormal positioning operation by a user, log data of at least one object in a vehicle key system is acquired to obtain first log data.
[0051] In one embodiment, the user can perform an abnormality location operation when the car key system is abnormal. For example, an abnormality location control can be provided in the car key system application, and when the car key system is abnormal, the user can click the control to trigger the abnormality location operation.
[0052] In one implementation, the user may perform anomaly location operations based on demand. For example, the user may perform anomaly location operations periodically.
[0053] In one implementation, each component of the car key system may be regarded as an object, for example, the car manufacturer cloud service is regarded as an object, the user's device is regarded as an object, and the vehicle is regarded as an object.
[0054] In one implementation, the car factory cloud service, the user's device, and the vehicle can also be divided into finer granularity to obtain more objects. Figure 1, the operating system, car manufacturer APP and Bluetooth / NFC / UWB in the user's device can be regarded as an object respectively. The network connection module, central computing platform and digital key business processor in the vehicle can be regarded as an object respectively. Each sub-service in the car manufacturer's cloud service set can be regarded as an object respectively.
[0055] In one implementation, log data of an object in the vehicle key system may be acquired to obtain the first log data.
[0056] In one implementation, log data of multiple objects in the car key system may be obtained to obtain the first log data. In one implementation, log data of each object in the car key system may be obtained to obtain the first log data.
[0057] In a possible implementation, in response to the user's abnormal positioning operation, a time interval for log acquisition may be determined, and log data of each object in the vehicle key system within the time interval may be acquired to obtain the first log data.
[0058] Exemplarily, when an abnormal error occurs in the car key system, it can be determined that an abnormal phenomenon occurs in the car key system. In this way, the user can trigger an abnormal location operation. The execution device of the method can determine the time interval for log acquisition in response to the user's abnormal location operation. As an example, the time interval can be determined according to the moment when the abnormality occurs. For example, the time interval can be determined from 6 minutes before to 6 minutes after the moment when the abnormality occurs. Of course, the length of the time interval can be set based on demand, and the embodiments of the present disclosure are not limited to this.
[0059] In a possible implementation, the time interval may be determined according to the time when the user triggers the abnormality locating operation. For example, the time interval may be determined from 3 minutes before to 3 minutes after the time when the user triggers the abnormality locating operation.
[0060] In step S32, the first log data is input into the abnormality locating system, and the abnormality locating result of the vehicle key system output by the abnormality locating system is obtained.
[0061] In one embodiment, the anomaly location system may include a rule engine, which may include predefined rules or thresholds. For example, a normal range of certain parameters of the log data may be set, and then when the relevant parameters in the first log data exceed the range, it may be determined that an anomaly has occurred.
[0062] In one embodiment, the anomaly location system may include a data mining and clustering engine. For example, the data mining and clustering engine may perform cluster analysis on the data through unsupervised learning methods, such as K-means, DBSCAN, etc. When outliers are identified in the data, it may be determined that an anomaly has occurred.
[0063] In one embodiment, the abnormality locating system includes a first model. Thus, the first log data can be input into the first model to obtain the abnormality locating result of the vehicle key system output by the first model.
[0064] The first model is trained by taking sample log data of at least one object of a sample car key system as input and processing results as output, and the processing results include abnormality positioning results of the sample car key system.
[0065] In a possible implementation, the abnormality positioning result may include a field indicating whether the vehicle key system is abnormal. As an example, the output field of the abnormality positioning result may include "normal; abnormal".
[0066] In a possible implementation, the abnormal location result may include the abnormal cause of the car key system. As an example, the output field of the abnormal location result may include "XX object failure in the car key system". When the car key system does not have an abnormality, the abnormal cause may be empty or "none".
[0067] In a possible implementation, the abnormality positioning result may also include “whether the vehicle key system is abnormal” and “the reason for the abnormality of the vehicle key system”.
[0068] In the above solution, in response to the user's abnormality locating operation, the log data of at least one object in the car key system is obtained to obtain the first log data. In this way, the first log data can be input into the abnormality locating system to obtain the abnormality locating result of the car key system output by the abnormality locating system.
[0069] In this way, the log data can be processed by the anomaly location system to obtain anomaly location results. Compared with manual analysis of log data and other methods, the above solution has higher data processing efficiency and anomaly location efficiency.
[0070] In a possible implementation, the first model can also be trained by taking sample log data of at least one object of the sample car key system as input and processing results as output. The processing results include abnormality positioning results of the sample car key system and functions performed by the sample car key system determined according to the sample log data.
[0071] In this case, the inputting the first log data into the first model to obtain the abnormality positioning result of the vehicle key system output by the first model includes: The first log data is input into the first model to obtain an abnormality positioning result of the vehicle key system output by the first model and a function performed by the vehicle key system determined according to the first log data.
[0072] In this way, the first model can also determine the function performed by the vehicle key system according to the first log data.
[0073] The following is an exemplary description of the training method of the first model. In order to train the first model, the sample log data may be obtained. Figure 4 is a flowchart of an abnormality location method shown in an exemplary embodiment of the present disclosure, referring to Figure 4 , the method is Figure 3 The basis includes: In step S33 , log data of at least one object of the sample car key system when executing a car key function is acquired to obtain second log data.
[0074] In one implementation, the vehicle key system may be used as the sample vehicle key system. In one implementation, a vehicle key system having the same architecture as the vehicle key system may also be used as the sample vehicle key system.
[0075] In one implementation, log data of one or more objects of the sample vehicle key system when performing a vehicle key function may be acquired to obtain second log data.
[0076] It is also worth noting that sometimes, the object may output log data including noise data. For example, for a cloud server running a single application, since the application is often configured with the ability to execute tasks in parallel with multiple threads, it may appear that thread X is executing task X1 and thread Y is simultaneously executing task Y1. In this way, thread X may output log x1, thread Y may output log y1, and thread X may output log x2. In this way, the log file may appear in the order of x1-y1-x2, thereby interfering with model training.
[0077] To this end, in a possible implementation manner, the acquiring log data of at least one object of the sample car key system when performing a car key function to obtain second log data includes: For each object in the at least one object, obtaining log data of the thread of the object during the period when the object executes the car key function, to obtain third log data; Log data irrelevant to the execution of the vehicle key function is removed from the third log data to obtain the second log data.
[0078] For example, in one embodiment, for each of the at least one object, the log data of each thread of the object during the period when the object executes the car key function can be obtained to obtain the third log data. Continuing with the above example, if the object is a cloud server, it includes threads X and Y. In this way, if thread X is executing car key task C1, and thread Y is simultaneously executing other task Y1, it may happen that thread X outputs log c1, thread Y outputs log y1, and thread X outputs log c2. In this way, log data in the order of c1-y1-c2 may appear.
[0079] In this case, the log data y1 can be determined from the log data c1-y1-c2 as log data irrelevant to the execution of the car key function. Therefore, the log data y1 irrelevant to the execution of the car key function can be removed from the log data c1-y1-c2 to obtain the second log data c1-c2.
[0080] In this way, by removing thread log data that is not related to the execution of the car key, the interference with the model training can be reduced.
[0081] In addition, in a possible implementation, the second log data may also include the noisy log data and / or the log data from which the noise is removed. Continuing with the above example, the second log data may include log data c1-y1-c2 and / or c1-c2. In this way, the training samples can be enriched and the generalization ability of the model can be improved.
[0082] In a possible implementation, the second log data may also include thread information. Using the example of log data c1-y1-c2, c1 is output by thread X, y1 is output by thread Y, and c2 is output by thread X. Therefore, the association between the thread and the log data output by the thread can be retained in the second log data. As an example, the second log data including thread information can be expressed as c1(X)-y1(Y)-c2(X), where the thread outputting the log data is in brackets. In this way, by training the model with log data with thread information, the large model's understanding of the concept of threads can be improved, and the ability to classify log outputs of different threads can be obtained, thereby improving the large model's ability to identify and locate problems.
[0083] Reference Figure 4 In step S34, a reference abnormal positioning result is determined, the reference abnormal positioning result includes the abnormal cause and / or execution result of the sample car key system when executing the car key function, and the execution result is configured to describe whether the car key function is successfully executed.
[0084] Exemplarily, the abnormal positioning result of the reference may be determined by manual or machine annotation.
[0085] In a possible implementation, the referenced abnormality location result includes the execution result of the vehicle key function, such as execution success, execution failure, etc. In some implementations, the execution result may include a field code describing execution success or failure, such as 1 for execution success and 0 for execution failure.
[0086] In a possible implementation, the referenced abnormal positioning result may include the abnormal reason for the sample car key system to perform the car key function. For example, when the sample car key system performs the car key function abnormally, the reason for the abnormal execution may be manually determined. Of course, when the sample car key system performs the car key function normally, the abnormal reason may be an empty field, or a related field indicating no abnormality.
[0087] In step S35, first training data is generated, the first training data includes the second log data and the reference anomaly location result, and the sample log data includes the first training data.
[0088] Exemplarily, the second log data and the reference anomaly location result may be associated, and the reference anomaly location result may be used as a label of the second log data, thereby obtaining the first training data.
[0089] In addition, in a possible implementation, the processing result includes an abnormality positioning result of the sample car key system and a function performed by the sample car key system determined according to the sample log data. In this case, the first training data can be generated, and the first training data includes the second log data, the reference abnormality positioning result, and the car key function.
[0090] In this way, by adding the function information of the car key to the first training data, it is helpful for the training model to identify the functions performed by the car key system according to the log data.
[0091] It should be noted that, in a possible implementation, the processing result includes the abnormality positioning result of the sample car key system, and the function performed by the sample car key system determined according to the sample log data. In this case, the first training data may also not include the car key function. For example, the model can be trained to classify the log data through unsupervised learning, and the association relationship between the classification result and the car key function can be established through manual labeling. In this way, when the first model outputs the category of the log data, the corresponding car key function can be determined according to the association relationship.
[0092] After obtaining the first training data, the first model can be obtained by training with the first training data. As an example, the first model can be a large language model (LLM). During training, the second log data in the first training data can be input into the large language model to obtain the processing result of the output of the large language model, such as the abnormal location result of the sample car key system determined based on the second log data. In this way, the output abnormal location result and the reference abnormal location result in the first training data can be compared to calculate the loss value. In this way, the model parameters can be updated according to the loss value, thereby realizing the training of the model and obtaining the first model.
[0093] In this way, the log data can be processed by training the first model, which can improve the processing efficiency of the log data and the efficiency of locating anomalies.
[0094] The following combination Figure 1 and 2 The anomaly location method disclosed in the present invention is described. In order to train the ability of the big model to locate problems in the full-link log of the digital key system, training data can be prepared so that the big model can learn to distinguish between normal full-link logs of the digital key system and abnormal full-link logs of the digital key system, and determine the location and cause of the anomaly from the log data.
[0095] Exemplarily, the training data may include: normal log data without noise, normal log data with noise, abnormal log data without noise, and abnormal log data with noise.
[0096] In order to collect the above log data, a training data collection table can be designed as shown in Table 1:
[0097] Table 1: Training data collection table Referring to Table 1, the meanings of the various fields in the training data collection table may include: The digital key function item is configured to describe the function being performed by the digital key system reflected in this log data. For example, it can be filled in as: Bluetooth digital key creation, NFC digital key deletion, etc.
[0098] Module A, B, and C logs are configured to describe specific log segments of each module. For example, they can be filled in as: cloud log, vehicle log, mobile log, etc., or they can be filled in as: key certificate sub-service module log, key information sub-service module log, central computing platform log, digital key business processor log, device operating system log, device factory APP log, etc. In some implementations, specific log segments can be filled in, or the accessible log segment file storage address can be filled in, and then the log file is read and parsed before model training.
[0099] The normal log label is configured to describe the results of the execution of specific functions by various modules of the system reflected in this log data. For example, it can be filled in as: normal or abnormal.
[0100] Abnormal reason description: If the content of the "Log Normal Label" field is "Abnormal", you can manually fill in this field as an explanation of the abnormal reason for this log data. Figure 2 For example, you can fill in: Module A has a record of calling module B's b1 interface, but module B does not have a record of module A requesting to call the b1 interface. It is suspected that the call link between the AB modules is abnormal.
[0101] Based on the training data collection table, training data collection can be performed, combined with Figure 2 For illustration, the collected training data may include: Normal log data without noise: It should be noted that the logs of various modules of the car key system may be mixed with multi-threaded logs. Taking the cloud service module as an example, on a cloud server running a single application, since the application is often configured with the ability to execute tasks in parallel with multiple threads, it may happen that thread X is executing task X1 and thread Y is executing task Y1 at the same time. In this way, thread X outputs log x1, thread Y outputs log y1, and thread X outputs log x2. Then the log file may appear in the order of x1-y1-x2. Such multi-threaded cross-mixed log files are likely to interfere with the normal log features of large models.
[0102] Therefore, normal log data without noise can be collected. For example, the log files of each participating module can be selected during the time period when the functional target is successfully executed multiple times. In the log files of each module, the output logs of the threads participating in this target function are selected, and the output logs of other threads that are not the target function but are executing at the same time are excluded. In this way, the filtered log files can be archived and filled into the training data collection table.
[0103] As an example, a normal log data without noise can be shown in Table 2:
[0104] Table 2: Normal log data without noise It should be understood that in actual applications, log files may still contain multi-threaded log intersections. Therefore, normal log data with noise can be collected to improve the model's adaptability to noise.
[0105] For example, to facilitate manual troubleshooting of system problems, you can add the thread number of the log before the specific log when outputting the log. Taking the above X and Y thread scenarios as an example, the final output file can be presented as: X - x1; Y - y1; X - x2.
[0106] Collecting noisy log data to train the big model can improve the big model's understanding of threads, obtain the ability to classify log outputs of different threads, and thus improve the big model's ability to solve problems in actual business scenarios. For example, the process of collecting normal log data with noise can include: Select the time period in which the functional objectives are successfully executed multiple times, and fill in the training data collection form based on the log files of each participating module and the original log file archives.
[0107] For example, normal log data containing noise can be shown in Table 3:
[0108] Table 3: Normal log data with noise In addition, we can collect noise-free abnormal log data and manually annotate the log data to train the large model to have the ability to identify abnormal situations in the digital key system. Figure 2 For example, the abnormal data after thread log classification and filtering can be manually annotated with the abnormal cause, for example: "Module A has a record of calling module B's b1 interface, but module B does not have a record of module A requesting to call module b1 interface, and it is suspected that the call link of modules AB is abnormal." Data collection is then performed after annotation.
[0109] Therefore, the mobile phone process of abnormal log data containing noise can include: selecting the log files of each participating module in the time period of multiple abnormal execution function targets. In the log files of each module, select the output logs of the threads participating in this target function, and exclude the output logs of other threads that are not the target function but are executing at the same time. Manually mark the abnormal cause of this abnormal scene, and fill in the training data collection table according to the filtered and marked log file archive.
[0110] As an example, the abnormal log data without noise can be shown in Table 4:
[0111] Table 4: Abnormal log data without noise In addition, abnormal log data containing noise can be collected and manually annotated to train the large model to have the ability to identify abnormal situations of the digital key system. For example, the log files of each relevant participating module in the time period of multiple abnormal execution function targets can be selected. In this way, the abnormal cause of this abnormal scene can be manually annotated, and the training data collection table can be filled out according to the original log file archive.
[0112] As an example, the abnormal log data containing noise can be shown in Table 5:
[0113] Table 5: Abnormal log data with noise After obtaining the training data, the model can be fine-tuned. For example, the organized training data can be provided to the selected basic model (you can also choose multiple basic models to experiment separately and then compare the effects) to complete the training and fine-tuning of the model. Through multiple rounds of fine-tuning, the model can be capable of outputting the following information: Identify and output the digital key function that the current car key system is executing as reflected in the full-link log file of the digital key system; whether there is any abnormality when the car key system executes the digital key function; if there is any abnormality, describe the cause of the abnormality.
[0114] After the training is completed, the trained model can be deployed and an interface can be provided to the outside. As an example, when a problem occurs in the digital key system, the R&D personnel can intercept the log files of the corresponding time period of each module and call the interface for input. In this way, the large model can assist in locating the location and cause of the problem based on the log data input by each end calling the interface.
[0115] Continue to use Figure 2 For example, in one example, the log data input to the large model can be shown in Table 6:
[0116] Table 6: Large model input log data In this way, the model can output: the current system is executing Figure 2 The functional flow shown; the execution of this system functional flow is abnormal; because there is a record of module A calling module B's b1 interface, and module B does not have a record of module A requesting to call the b1 interface, it is suspected that there is an abnormality in the AB module call link.
[0117] By adopting the above solution, digital key system developers can collect log data of the modules they are responsible for in the car key system when problems may occur in the car key system. By inputting the log data into the trained model, the problem location results and causes output by the model can be obtained. In this way, the cost of repeated communication between multiple R&D personnel can be reduced, and the efficiency of log data processing and problem troubleshooting can be improved.
[0118] Based on the same inventive concept, an embodiment of the present disclosure provides an abnormality locating device. Figure 5 is a block diagram of an abnormality locating device shown in an exemplary embodiment of the present disclosure, referring to Figure 5 , the abnormality locating device comprises: The first module 501 is configured to obtain log data of at least one object in the vehicle key system in response to an abnormal positioning operation of the user, and obtain first log data; The processing module 502 is configured to input the first log data into an abnormality locating system, and obtain an abnormality locating result of the vehicle key system output by the abnormality locating system.
[0119] In this way, the log data can be processed by the anomaly location system to obtain anomaly location results. Compared with manual analysis of log data and other methods, the above solution has higher data processing efficiency and anomaly location efficiency.
[0120] Optionally, the processing module 502 includes: A second module is configured to input the first log data into a first model to obtain an abnormality positioning result of the vehicle key system output by the first model; Among them, the anomaly positioning system includes the first model, which is trained by taking sample log data of at least one object of the sample car key system as input and processing results as output, and the processing results include anomaly positioning results of the sample car key system.
[0121] Optionally, the second module includes: a first submodule, configured to input the first log data into the first model, obtain an abnormality locating result of the vehicle key system output by the first model, and a function performed by the vehicle key system determined according to the first log data; Among them, the first model is trained by taking sample log data of at least one object of the sample car key system as input and processing results as output, and the processing results include the abnormality positioning results of the sample car key system and the functions performed by the sample car key system determined according to the sample log data.
[0122] Optionally, the device comprises: A third module is configured to obtain log data of at least one object of the sample car key system when performing a car key function to obtain second log data; a fourth module, configured to determine a reference abnormal positioning result, wherein the reference abnormal positioning result includes an abnormal cause and / or an execution result of the sample car key system when executing the car key function, and the execution result is configured to describe whether the car key function is successfully executed; The fifth module is configured to generate first training data, wherein the first training data includes the second log data and the reference anomaly location result, and the sample log data includes the first training data.
[0123] Optionally, the fifth module includes: The second submodule is configured to generate the first training data when the processing result includes the abnormality positioning result of the sample car key system and the function performed by the sample car key system determined according to the sample log data, and the first training data includes the second log data, the reference abnormality positioning result, and the car key function.
[0124] Optionally, the third module includes: A third submodule is configured to obtain, for each object in the at least one object, log data of the thread of the object during the period when the object executes the car key function, to obtain third log data; The fourth submodule is configured to remove log data irrelevant to the execution of the vehicle key function from the third log data to obtain the second log data.
[0125] Optionally, the first module 501 includes: A fifth submodule is configured to determine a time interval for log acquisition in response to an abnormality location operation of a user; The sixth submodule is configured to obtain the log data of each object in the vehicle key system within the time interval to obtain the first log data.
[0126] The present disclosure provides an abnormality locating device, comprising: processor; a memory for storing processor-executable instructions; The processor is configured to execute the steps of the anomaly locating method provided in at least one embodiment of the present disclosure.
[0127] An embodiment of the present disclosure provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the abnormality locating method provided in at least one embodiment of the present disclosure are implemented.
[0128] An embodiment of the present disclosure provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the abnormality locating method provided in at least one embodiment of the present disclosure.
[0129] Regarding the abnormality locating device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the abnormality locating method, and will not be elaborated here.
[0130] Figure 6 1 is a block diagram of a device 1900 for abnormality location shown according to an exemplary embodiment. For example, the device 1900 can be provided as a server. Referring to FIG. 1900, the device 1900 includes a processing component 1922, which further includes one or more processors, and a memory resource represented by a memory 1932 for storing instructions that can be executed by the processing component 1922, such as an application. The application stored in the memory 1932 may include one or more modules, each of which corresponds to a set of instructions. In addition, the processing component 1922 is configured to execute instructions to perform the above-mentioned log data processing method.
[0131] The device 1900 may also include a power supply component 1926 configured to perform power management of the device 1900, a wired or wireless network interface 1950 configured to connect the device 1900 to a network, and an input / output interface 1958. The device 1900 may operate based on an operating system stored in the memory 1932, such as Windows Server 2000. TM , Mac OS X TM , Unix TM , Linux TM , FreeBSD TM or similar.
[0132] In addition, the word "exemplary" is used herein to indicate serving as an example, instance, or diagram. Any aspect or design described as "exemplary" in this article is not necessarily understood to be advantageous compared to other aspects or designs. On the contrary, the use of the word exemplary is intended to present concepts in a specific way. As used herein, the term "or" is intended to represent an inclusive "or" rather than an exclusive "or". That is, unless otherwise specified or clear from the context, "X applies A or B" is intended to represent any one of the natural inclusive arrangements. That is, if X applies A; X applies B; or X applies both A and B, "X applies A or B" is satisfied under any of the aforementioned examples. In addition, unless otherwise specified or clearly pointed to a singular form from the context, the articles "one" and "an" as used in this application and the appended claims are generally understood to mean "one or more".
[0133] Likewise, although the present disclosure has been shown and described with respect to one or more implementations, equivalent variations and modifications will occur to those skilled in the art after reading and understanding the specification and drawings. The present disclosure includes all such modifications and variations and is limited only by the scope of the claims. In particular, with respect to the various functions performed by the components (e.g., modules) described above, unless otherwise indicated, the terms used to describe such components are intended to correspond to any component (functionally equivalent) that performs the specific functions of the described components, even if the structure is not equivalent to the disclosed structure. In addition, although specific features of the present disclosure may have been disclosed with respect to only one of several implementations, such features may be combined with one or more other features of other implementations as may be desired and beneficial to any given or specific application. In addition, with respect to "including", "having", "having", "having", or variations thereof used in a specific embodiment or claim, such terms are intended to be inclusive in a manner similar to the term "comprising".
[0134] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the appended claims.
[0135] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
[0136] In addition, unless otherwise specified, the features of some embodiments of the present disclosure described herein may be combined with each other. As used herein, the term "and / or" includes any one of the related listed items and any combination of any two or more; similarly, "at least one of . . . " includes any one of the related listed items and any combination of any two or more.
[0137] Although terms such as "first", "second" and "third" may be used herein to describe various modules, these modules are not limited to these terms. On the contrary, these terms are only used to distinguish one module from another. Therefore, without departing from the teachings of the various examples, the first module mentioned in the examples described herein may also be referred to as the second module. In addition, the terms "first" and "second" are used only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description herein, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
Claims
1. A method for locating an abnormality, characterized in that: include: In response to an abnormal positioning operation by a user, acquiring log data of at least one object in the vehicle key system to obtain first log data; The first log data is input into an abnormality locating system, and an abnormality locating result of the vehicle key system output by the abnormality locating system is obtained.
2. The method according to claim 1, characterized in that The step of inputting the first log data into an abnormality locating system to obtain an abnormality locating result of the vehicle key system output by the abnormality locating system includes: Inputting the first log data into a first model to obtain an abnormality locating result of the vehicle key system output by the first model; Among them, the anomaly positioning system includes the first model, which is trained by taking sample log data of at least one object of the sample car key system as input and processing results as output, and the processing results include anomaly positioning results of the sample car key system.
3. The method according to claim 2, characterized in that The step of inputting the first log data into the first model to obtain an abnormality locating result of the vehicle key system output by the first model includes: Inputting the first log data into the first model, obtaining an abnormality locating result of the vehicle key system output by the first model, and a function performed by the vehicle key system determined according to the first log data; Among them, the first model is trained by taking sample log data of at least one object of the sample car key system as input and processing results as output, and the processing results include the abnormality positioning results of the sample car key system and the functions performed by the sample car key system determined according to the sample log data.
4. The method according to claim 2 or 3, characterized in that: The method comprises: Acquire log data of at least one object of the sample car key system when performing a car key function to obtain second log data; Determine a reference abnormal positioning result, wherein the reference abnormal positioning result includes an abnormal cause and / or an execution result of the sample car key system when executing the car key function, and the execution result is configured to describe whether the car key function is successfully executed; First training data is generated, where the first training data includes the second log data and the reference anomaly location result, and the sample log data includes the first training data.
5. The method according to claim 4, characterized in that The generating first training data comprises: When the processing result includes the abnormality positioning result of the sample car key system and the function performed by the sample car key system determined according to the sample log data, the first training data is generated, and the first training data includes the second log data, the reference abnormality positioning result, and the car key function.
6. The method according to claim 4, characterized in that The step of obtaining log data of at least one object of the sample car key system when performing a car key function to obtain second log data includes: For each object in the at least one object, obtaining log data of the thread of the object during the period when the object executes the car key function, to obtain third log data; Log data irrelevant to the execution of the vehicle key function is removed from the third log data to obtain the second log data.
7. The method according to any one of claims 1 to 3, characterized in that The step of acquiring log data of at least one object in the vehicle key system in response to the abnormal positioning operation of the user to obtain first log data includes: In response to the user's abnormal location operation, determine a time interval for log acquisition; The log data of each object in the vehicle key system within the time interval is obtained to obtain the first log data.
8. An abnormality locating device, characterized in that: include: A first module is configured to obtain log data of at least one object in the vehicle key system in response to an abnormal positioning operation of a user, and obtain first log data; The second module is configured to input the first log data into an abnormality locating system to obtain an abnormality locating result of the vehicle key system output by the abnormality locating system.
9. An abnormality locating device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to execute the steps of the method according to any one of claims 1 to 7.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.
11. A computer program product, characterized in that The invention comprises a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 7.