Automatic driving event analysis method and device, equipment and storage medium

Abstract

The invention provides an automatic driving event analysis method and device, equipment and a storage medium, and the method comprises the steps: obtaining a natural language inputted by a user, the natural language form comprises text description, voice description, picture description and the like, and the specific content comprises a data storage path, a signal structure and an event triggering judgment condition; generating a data analysis code according to the signal structure; on the basis of the data analysis code, generating, compiling and deploying of the application service are completed according to the system interface, after deployment is completed, signal data can be obtained from the data storage path, and the signal data are analyzed into natural language data; and determining target data meeting event triggering judgment conditions, and further generating a sample report. Through the scheme, the time required for manually setting parameters and analyzing data is shortened, and the overall efficiency of event analysis is improved. In addition, along with feedback accumulation, the self analysis process is continuously optimized, and it is ensured that high efficiency and accuracy are kept in the face of new data.

Description

Technical Field

[0001] This application relates to the field of vehicle technologies, and particularly to a method, apparatus, device, and storage medium for analyzing autonomous driving events. Background Art

[0002] In the field of autonomous driving, offline analysis of data and scenario event recognition have become key aspects of the progress of autonomous driving technologies. With the rapid development of artificial intelligence technologies, how to efficiently and accurately process massive amounts of data and identify key scenario events from it has become one of the research hotspots.

[0003] Most of the data in vehicles is signal data from sensors rather than image data. When training a model, converting signals into images can improve recognition accuracy, but in some cases, information on some time series features may be lost. Moreover, the method of manually annotating and analyzing signal data not only increases costs but also limits the generalization ability of problem analysis.

[0004] Therefore, how to efficiently mine valuable events from signal data is a challenge. Summary of the Invention

[0005] This application provides a method, apparatus, device, and storage medium for analyzing autonomous driving events, which are used to quickly mine event data from signal data.

[0006] In a first aspect, this application provides a method for analyzing autonomous driving events, and the method includes:

[0007] In response to an operation by the user to enable analysis metrics, obtain natural language input by the user, where the natural language includes a data storage path, a signal structure, and an event trigger determination condition;

[0008] Generate data parsing code according to the signal structure, where the data parsing code is used to parse the signal data of the signal structure;

[0009] Generate an application service based on the data parsing code and a preset system interface specification; where the system interface specification defines the rules, formats, and sequences for interaction and communication between system components;

[0010] Invoke the application service, obtain signal data in machine language from the data storage path, and parse the signal data into natural language data;

[0011] Determine target data that meets the event trigger determination condition from the natural language data;

[0012] Generate a sample report according to the target data;

[0013] Output the sample report in the conversation of the analysis metrics.

[0014] Optionally, the method further includes:

[0015] In the conversation of the analysis metrics, receive a new data storage path input by the user;

[0016] Generate a new sample report according to the new data storage path;

[0017] Output the new sample report.

[0018] Optionally, the signal structure includes a signal interface name and a signal declaration.

[0019] Optionally, the signal structure is in the form of an example signal.

[0020] Optionally, the example signal is natural language in an image;

[0021] Correspondingly, the method further includes:

[0022] Extract the example signal representing the signal structure from the image.

[0023] Optionally, the method further includes:

[0024] Obtain a report keyword input by the user;

[0025] Retrieve the report keyword in all stored sample reports and output the retrieved sample reports.

[0026] Optionally, the sample report includes target data and associated data, and the associated data is natural language data within a preset time period before and after the target data.

[0027] In a second aspect, the present application provides a device for analyzing autonomous driving events, and the device includes:

[0028] An acquisition module, configured to acquire natural language input by the user in response to an operation of the user to start analysis metrics, where the natural language includes a data storage path, a signal structure, and an event trigger determination condition;

[0029] A code generation module, configured to generate data parsing code according to the signal structure, where the data parsing code is used to parse the signal data of the signal structure;

[0030] An application service generation module, configured to generate an application service based on the data parsing code and a preset system interface specification; wherein, the system interface specification defines rules, formats, and sequences for interaction and communication between system components;

[0031] A parsing module that calls the application service to obtain signal data in machine language from the data storage path and parses the signal data into natural language data;

[0032] A data determination module for determining target data that meets the event trigger determination condition from the natural language data;

[0033] A report generation module for generating a sample report based on the target data;

[0034] An output module for outputting the sample report in the conversation of the analysis metrics.

[0035] In a third aspect, the present application provides an electronic device, including: a memory, a processor;

[0036] The memory stores computer execution instructions;

[0037] The processor executes the computer execution instructions stored in the memory, so that the processor executes the method according to any one of the first aspects.

[0038] In a fourth aspect, the present application provides a computer-readable storage medium, in which computer execution instructions are stored, and when the computer execution instructions are executed by a processor, they are used to implement the method according to any one of the first aspects.

[0039] In a fifth aspect, the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the above first aspect and / or various possible implementation manners of the first aspect.

[0040] The method, apparatus, device, and storage medium for autonomous driving event analysis provided by this application. The method includes: responding to the operation of the user to enable the analysis metrics, obtaining the natural language input by the user, where the natural language includes the data storage path, signal structure, and event trigger determination condition; generating data parsing code according to the signal structure, where the data parsing code is used to parse the signal data of the signal structure; based on the data parsing code, completing the generation, compilation, and deployment of the application service according to the system interface. After the deployment is completed, it can obtain the signal data in machine language from the data storage path and parse the signal data into natural language data; determining the target data that meets the event trigger determination condition from the natural language data; generating a sample report according to the target data; and outputting the sample report in the conversation of the analysis metrics. By describing the signal structure and the signal characteristics of the event in natural language, the data is defined and described in a more intuitive and human-thinking-habit way. This semi-automated process greatly reduces the time required for manual parameter setting and data analysis, thus significantly improving the overall efficiency of event analysis. With the accumulation of user feedback and the continuous iteration of the algorithm, it can continuously learn and optimize its own analysis process. This self-evolving ability ensures that it always remains efficient and accurate in the face of new data and challenges. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The drawings here are incorporated into the specification and form a part of this specification, showing embodiments consistent with this application, and are used together with the specification to explain the principles of this application.

[0042] Figure 1 It is a schematic diagram of the autonomous driving event analysis system provided by this application;

[0043] Figure 2 It is a schematic flowchart of Embodiment 1 of the method for autonomous driving event analysis provided by this application;

[0044] Figure 3 It is a schematic diagram of the input image provided by this application;

[0045] Figure 4 It is another schematic diagram of the input image provided by this application;

[0046] Figure 5 It is a schematic flowchart of the autonomous driving event analysis provided by this application;

[0047] Figure 6 It is a schematic structural diagram of an apparatus for autonomous driving event analysis provided by this application;

[0048] Figure 7 It is a schematic structural diagram of the electronic device provided by this application.

[0049] Through the above-mentioned drawings, specific embodiments of the present application have been shown, and will be described in more detail hereinafter. These drawings and written descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by reference to specific embodiments. Detailed Description of the Embodiments

[0050] Here, exemplary embodiments will be described in detail, and examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0051] In the field of autonomous driving, the offline analysis of data and the recognition of scenario events have become key aspects of the progress of autonomous driving technology. With the rapid development of artificial intelligence technology, the combination of artificial intelligence (AI) and big data analysis has become an important force driving the progress of many fields, and large models have demonstrated powerful application potential in multiple fields. Especially in the field of autonomous driving, how to efficiently and accurately process massive amounts of data and identify key scenario events from it has become one of the research hotspots.

[0052] Studies have shown that models such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) can extract useful information from data obtained from devices such as vision sensors and radars; in addition, by converting signals into image form, existing image recognition algorithms can be better utilized to process radar or lidar data.

[0053] However, in the offline analysis of autonomous driving data, the following main defects or deficiencies exist:

[0054] 1) Dependence on a large amount of labeled data: Existing technical solutions usually require a large amount of manual annotation and analysis of data for model training, which not only increases costs but also limits the generalization ability of problem analysis.

[0055] 2) False positive rate in complex scenarios: In complex driving scenarios, the event recognition accuracy of existing solutions still needs to be improved, and there is a high risk of false positives.

[0056] 3) Complex multi-modal data processing: When processing multi-modal data such as images and time series signals, existing solutions require complex model integration and data processing processes, increasing the complexity and computational cost of the system.

[0057] 4) Insufficient mining of natural language event tags: The existing technologies are relatively lacking in mining event time tags described in natural language, which limits the application of the technologies in a wider range of scenarios.

[0058] 5) Frequent changes in the events of concern: With the continuous upgrade of the autonomous driving function, the events of concern are also constantly changing, and the characteristic signals or signal indicators related to the events are also constantly changing. It is necessary to repeatedly explore new events from the existing data and build new event recognition tools;

[0059] In view of this, the present application proposes an autonomous driving event analysis method to solve the above problems. By introducing an artificial intelligence system based on a large model for offline analysis of autonomous driving data and event scenario mining, directly using the signal data of the time series for training, and combining natural language processing technologies, event scenario mining is carried out on the signal data to improve the accuracy, efficiency, and automation degree of autonomous driving data analysis and event recognition. The present application has the following innovation points in the process of solving the above problems:

[0060] 1) Automatic data annotation: By designing a mechanism, it is possible to achieve automatic annotation of autonomous driving data with less manual intervention, thereby reducing costs.

[0061] 2) Direct processing of time series signals: Develop a method that can directly input the original time series signals into the large model to avoid losses during the information conversion process.

[0062] 3) Mining of natural language time tags: Introduce advanced natural language processing technologies and optimize the algorithms to improve the ability to identify key events from text descriptions.

[0063] 4) Semi-automatic data analysis process: Through natural language description of the signal structure and signal characteristics of the event, the system realizes automatic signal parsing, visualization, and event tagging, timestamp positioning, and report generation.

[0064] The technical solutions of the present application and how the technical solutions of the present application solve the above technical problems will be described in detail below with specific embodiments. These specific embodiments below can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0065] Figure 1This is a schematic diagram of the autonomous driving event analysis system provided by this application. The autonomous driving event analysis system mainly consists of four core parts, including a human-computer interaction server: which understands the analyst's instructions, decomposes the instructions into resources available to the computing server, and presents the report results; a computing server: which parses, analyzes, and generates results for the data in the data storage server according to the data analysis instructions, numbers and archives the instructions for subsequent calls; a report server: which generates and stores reports for the analysis results for viewing or for historical archiving backup and viewing; a data storage server: which is used to store the data to be analyzed and provides data interfaces for the computing server, the human-computer interaction server, and the report server.

[0066] The human-computer interaction server, which is the bridge between the user and the system, is mainly responsible for the following aspects of work:

[0067] 1) Instruction understanding and parsing: Receives and analyzes the analysis instructions input by the user to ensure the accuracy of the instruction semantics.

[0068] 2) Task decomposition and scheduling: Decomposes complex analysis instructions into multiple subtasks and assigns these subtasks to the computing server for processing.

[0069] 3) Result presentation and management: Collects the analysis results returned by the computing server and presents them to the user in an intuitive way, while supporting the export and management functions of the results.

[0070] 4) User interface design: Provides a friendly graphical user interface (GUI) so that non-technical personnel can easily start data analysis operations.

[0071] The computing server, which is the core processing unit of the entire system, mainly has the following responsibilities:

[0072] 1) Data processing and analysis: Efficiently parses and processes the data in the data storage server and executes various statistical analysis and machine learning algorithms.

[0073] 2) Result generation and optimization: Generates detailed analysis reports and visualization charts to ensure the accuracy and interpretability of the results.

[0074] 3) Task numbering and archiving: Assigns a unique number to each analysis task and archives and stores the relevant data and results for easy subsequent query and reuse.

[0075] 4) Resource management and scheduling: Dynamically manages computing resources, optimizes the task execution process, and improves the overall processing capacity and response speed of the system.

[0076] The report server focuses on the display and storage of analysis results, and its specific functions are as follows:

[0077] 1) Report generation engine: Automatically generate formatted report documents based on the analysis results provided by the computing server;

[0078] 2) Multi-format support: Support multiple report formats (such as PDF, Excel, HTML, etc.) to meet the needs of different users;

[0079] 3) History record management: Save all generated report copies, build a complete historical database for easy traceability and comparative analysis;

[0080] 4) Permission control and security mechanism: Implement strict access control policies to ensure the security and privacy protection of report data.

[0081] The data storage server is the basic support of the entire system, and its main functions include:

[0082] 1) Data reception and storage: Receive data from different data sources and perform efficient and secure storage management;

[0083] 2) Data interface provision: Provide standardized data access interfaces for the computing server, human-computer interaction server, and report server to ensure data circulation and consistency;

[0084] 3) Data backup and recovery: Regularly back up data and establish a perfect data recovery mechanism to prevent data loss and damage;

[0085] 4) Data security and privacy protection: Adopt encryption technology and access control measures to ensure data security and user privacy.

[0086] Figure 2 The flowchart of Embodiment 1 of the method for analyzing autonomous driving events provided in this application is as Figure 2 shown, and the method includes:

[0087] S101. In response to the user's operation of enabling the analysis index, obtain the natural language input by the user, where the natural language includes the data storage path, signal structure, and event trigger determination condition.

[0088] In this step, the user enables a new round of index analysis of the model on the human-computer interaction server and enters the natural language index definition in the dialog box. The index definition first defines the name of the analysis index. For example, the index for which the user conducts the analysis is abnormal vehicle speed. Then, the path for storing signal data and the signal structure are given, specifically including defining the data type, format, and its various components of the signal (such as the interface name and signal declaration). The user also needs to specify the conditions under which the signal meets the event trigger judgment condition.

[0089] The signal interface name is usually the identifier or name of the signal, which is used to identify and reference the signal in the system. Signal declaration involves defining the signal in the code, usually including the type of the signal (e.g., integer, floating-point number, boolean, etc.) and other information related to the signal. This is also part of the signal structure, indicating how the signal is declared and used in the system. For example, in natural language, the input for signal declaration: float temperature sensor, this is a signal declaration of floating type, indicating that the output of the temperature sensor is a floating value.

[0090] In one implementation, the signal structure is in the form of an example signal, and the large model determines the signal structure based on the example signal.

[0091] In one implementation, instead of directly inputting natural language, the user inputs an image that contains natural language. The natural language in the image can be a statement directly describing the signal structure or an example signal. In this way, it is first necessary to extract the natural language from the image and determine whether it directly represents the signal structure or an example signal. As Figure 3 and Figure 4 shown.

[0092] S102. Generate data parsing code according to the signal structure. The data parsing code is used to parse the signal data of the signal structure.

[0093] To improve the automation efficiency of the whole process, the large model system generates data parsing code and applications based on the provided description of the signal structure at the "computing server side", and can extract and parse the signal, and represent it as natural language.

[0094] S103. Generate application services based on the data parsing code and the preset system interface specifications; among them, the system interface specifications define the rules, formats, and sequences of interaction and communication between system components.

[0095] The system interface specifications include the file names of the generated code and executable files in each business link, the request interfaces of the services (server addresses, ports), the parameters passed by each application or service, the order of invocation of each application or service, etc.

[0096] S104. Invoke the application service to obtain the signal data in machine language from the data storage path, and parse the signal data into natural language data.

[0097] S105. Determine the target data that meets the event trigger determination conditions from the natural language data.

[0098] For example, if the signal structure is a step signal, a pulse signal, or a threshold signal, the model will determine which data points meet the trigger judgment conditions based on the converted natural language data and the specified event trigger judgment conditions, and retain the data that meets the event trigger judgment conditions.

[0099] S106. Generate a sample report based on the target data and output the sample report in the dialogue of the analysis metrics.

[0100] The report will indicate which time period the data meets the event trigger conditions and display the associated data on the sample report. The associated data refers to the signal data values within a preset time period before and after the event trigger, facilitating the user to analyze the context of the trigger event in detail.

[0101] The system will display the generated sample report in the dialogue box of the analysis metrics to help the user understand the details of the signal trigger event and display the relevant data analysis results.

[0102] The user can view the sample report and determine whether the events identified in the report are correct and meet the requirements. If it meets the user's needs, the user can save the entire analysis metrics module, that is, when the user analyzes this metric next time, there is no need to re-enter the natural language, and only the input data needs to be changed, and the model can perform the metric analysis again and generate a sample report.

[0103] If the user believes that there are deficiencies in the event judgment in the sample report, the user can enter the areas that need to be adjusted in the dialogue box, and the large model can make adjustments based on the user's feedback.

[0104] With the accumulation of user feedback and the continuous iteration of the algorithm, it can continuously learn and optimize its own analysis process. This self-evolving ability ensures that it always remains efficient and accurate in the face of new data and challenges.

[0105] Exemplary:

[0106] Suppose the user provides the following information:

[0107] Data storage path: / path / to / data / signals.data

[0108] Signal structure: {signal_name: "temperature", data_type: "float", units: "Celsius"};

[0109] Event trigger determination condition: temperature > 30.0°C;

[0110] Parsing code generation: The parsing code will read the file according to the structure and parse out the data of the temperature signal.

[0111] Data parsing: After reading the file, the data of the temperature signal is converted into natural language. For example:

[0112] Temperature at 10:00 AM: 29.5°C

[0113] Temperature at 10:15 AM: 31.2°C

[0114] Trigger condition determination: According to the condition that the signal value is greater than 30°C, it is identified that the record at 10:15 AM meets the condition.

[0115] Sample report generation: According to the triggered target data, a report is generated, and the content is as follows:

[0116] Event trigger time: 10:15 AM;

[0117] Triggered signal: Temperature;

[0118] Value: 31.2°C;

[0119] Other associated data or charts.

[0120] In this way, the complete process from signal data parsing to event determination and then to sample report generation is completed.

[0121] This embodiment provides a method for analyzing autonomous driving events. The method includes: in response to a user's operation of enabling analysis metrics, obtaining natural language input by the user, where the natural language includes a data storage path, a signal structure, and an event trigger determination condition; generating data parsing code according to the signal structure, where the data parsing code is used to parse the signal data of the signal structure; generating an application service through the data parsing code and a preset system interface specification, calling the application service to obtain the signal data in machine language from the data storage path, and parsing the signal data into natural language data; determining target data that meets the event trigger determination condition from the natural language data; generating a sample report according to the target data; and outputting the sample report in the conversation of the analysis metrics. By describing the signal structure and the signal characteristics of events in natural language, data is defined and described in a more intuitive and human-thinking-habit way. This semi-automated process greatly reduces the time required for manually setting parameters and analyzing data, thus significantly improving the overall efficiency of data analysis. With the accumulation of user feedback and the continuous iteration of algorithms, the present invention can continuously learn and optimize its own analysis process. This self-evolution ability ensures that it always remains efficient and accurate in the face of new data and challenges. In this way, it is possible to avoid excessive implementation processes of underlying data analysis technologies, improve the efficiency of data analysis, and achieve intelligent implementation through a large model in key steps such as signal parsing, visualization, event tagging, and timestamp positioning, effectively reducing potential errors introduced by human operations.

[0122] In addition, the multi-modal data analysis method can directly parse structured signals and analyze signal timing diagrams based on image recognition, achieving efficient and accurate data processing, improving the analysis efficiency and accuracy, and at the same time reducing the maintenance cost of complex systems and the technical threshold of data mining.

[0123] Figure 5 It is a schematic diagram of the process for analyzing autonomous driving events provided by this application. As Figure 5 shown, the main work processes include key links such as starting a new analysis, analyzing new data, and viewing historical reports.

[0124] 1. Through the [New Business] link, confirm the subsequent execution steps, namely [Start a New Analysis], [Analyze New Data], and [View Historical Reports].

[0125] 2. In the "Human-Machine Interaction Server", through the [Start a New Analysis] link, the analyst first starts [Define Analysis Metrics], and through [Describe Signal Structure], in the data log to be analyzed, give the organization form of the data, such as information on vehicle speed, target distance, etc.; the large model system in the "Computing Server" generates data parsing code and applications based on the provided [Describe Signal Structure].

[0126] 3. The analyst provides the

description of signal form

example signals

[0127] 4. Based on the event determination application generated above, test and parse the sample data log and generate a sample report. The analyst initially confirms the feasibility of the process. If there are optimization requirements,

define analysis metrics

[0128] 5. If it is confirmed to be available through the analysis of the sample sample by the analyst, in the

fill in data path

[0129] 6. Subsequently, in the "computing server", collect the above signal parsing and metric judgment applications, and in the

system analysis and generate report

[0130] 7. At the same time, the "computing server" will archive and back up the resources of this test (such as the generated signal parsing application), and mark the ID for subsequent reuse.

[0131] 8. If in the

define analysis metrics

modify historical analysis template

[0132] 9. If in the

new business

analyze new data

fill in data path

[0133] 10. If in the

new business

view historical report

system retrieve historical report

describe report keywords

manual view report

[0134] This application mainly describes the data and scenario cases that need to be analyzed in natural language, and automatically analyzes the data and presents reports. It can be considered as an artificial intelligence system that integrates a dedicated large model for all aspects of intelligent driving log data analysis and event mining business problems. The focus is to generate corresponding executable applications through natural language or sample pictures and other human-understandable signal descriptions, parse, analyze and identify the recorded and transmitted working data of the autonomous driving system, and report.

[0135] Among them, the indicator definition link is manually described in natural language, or a schematic diagram of the signal is provided for the system to identify and understand, and the indicator results are generated and confirmed by humans. This link can also be generated or replaced by other artificial intelligence methods such as reinforcement learning, or simulated learning to simulate and generalize existing evaluation indicators to reduce the degree of human intervention, but at the current stage, the evaluation indicators generated based on reinforcement learning / simulation learning methods have high uncertainty and low usability.

[0136] In addition, in addition to directly giving the indicator definition (natural language description or picture example), the indicator definition stage can also provide dynamic forms such as sample data, that is, the sample data contains the event signal characteristics expected to be detected, and the system understands and judges the data characteristics by itself, or it can provide video streams, other similar report index references, etc. However, at the current stage, the increased complexity of the system will reduce the effectiveness of the evaluation results.

[0137] The review and confirmation of the result report is done manually, mainly for the evaluation of the results. The solution can be expanded to an interactive interface, including validity confirmation and result evaluation, and the evaluation results of the report can be used for the training and feedback improvement of the large model. However, at the current stage, it will introduce more uncertainties, which is not conducive to the stability of the report results.

[0138] The retrieval of historical reports can currently be achieved through traditional search engines or through artificial intelligence-based report summaries. The traditional search engine method produces more direct and detailed results, while the artificial intelligence-based summary method is more conducive to improving the efficiency of report understanding.

[0139] In the specific implementation of large models or artificial intelligence agents in each link, there can be different forms, such as a Transformer-based network framework, or it can be implemented through network model frameworks such as RetNet and Mamba.

[0140] In the above data parsing, analysis, and signal recognition applications, in addition to common C++ code and its applications, different types of code languages such as Python and Rust can also be generated. The current stage report is mainly provided in the form of an HTML / JSON web service, and static files such as PDF and TXT or dynamic files such as video and PPT can also be generated according to needs, depending on the requirements of the analyst or the data storage format of the data log file.

[0141] The data parsing application generated by the above process is mainly used for the data stored on the "data storage server", that is, the log data files generated when various autonomous driving systems work, including but not limited to log logs, JSON data, YAML data, rosbag data, bin data, pcap data, etc.; the data content stored therein includes but not limited to various types of data such as the vehicle's own state and sensor state.

[0142] Currently, limited by the requirements of computing power and real-time performance, this system mainly operates based on cloud servers. In the future, with the improvement of the computing power of terminal devices (such as in-vehicle data collection devices, mobile devices, etc.) and the simplification of models, it can also be deployed on terminal systems to run in real time, realizing preliminary data analysis and event recognition at the vehicle end, reducing the data transmission bandwidth requirements and storage requirements.

[0143] Figure 6 It is a schematic structural diagram of a device for autonomous driving event analysis provided for this application, as Figure 6 shown. The device 40 for autonomous driving event analysis includes:

[0144] An acquisition module 401, configured to obtain the natural language input by the user in response to the user's operation of enabling the analysis metrics, where the natural language includes a data storage path, a signal structure, and an event trigger determination condition;

[0145] A code generation module 402, configured to generate data parsing code according to the signal structure, where the data parsing code is used to parse the signal data of the signal structure;

[0146] An application service generation module 403, which generates an application service based on the data parsing code and a preset system interface specification; where the system interface specification defines the rules, formats, and sequences for interaction and communication between system components;

[0147] A parsing module 404, configured to call the application service to obtain the signal data in machine language from the data storage path and parse the signal data into natural language data;

[0148] A data determination module 405, configured to determine target data that meets the event trigger determination condition from the natural language data;

[0149] A report generation module 406, configured to generate a sample report according to the target data;

[0150] An output module 407, configured to output the sample report in the conversation of the analysis metrics.

[0151] Optionally, the acquisition module 401 is further configured to receive a newly input data storage path from a user in the conversation of the analysis metrics;

[0152] Correspondingly, the report generation module 405 is further configured to generate a new sample report according to the new data storage path;

[0153] Correspondingly, the output module 406 is further configured to output the new sample report.

[0154] Optionally, the signal structure includes a signal interface name and a signal declaration.

[0155] Optionally, the signal structure is in the form of an example signal.

[0156] Optionally, the example signal is natural language in an image;

[0157] Correspondingly, the acquisition module 401 is further configured to:

[0158] Extract an example signal representing the signal structure from the image.

[0159] Optionally, the acquisition module 401 is further configured to acquire a report keyword input by a user.

[0160] Correspondingly, the data determination module 404 is further configured to retrieve the report keyword from all the stored sample reports and output the retrieved sample reports.

[0161] Optionally, the sample report includes target data and associated data, and the associated data is natural language data within a preset time period before and after the target data.

[0162] The device for analyzing autonomous driving events provided in this embodiment can execute the method provided in the above method embodiment, and its implementation principle and technical effects are similar, which will not be elaborated here in this embodiment.

[0163] Figure 7 It is a schematic structural diagram of an electronic device provided in this application. As Figure 7 shown, the electronic device 50 provided in this embodiment includes: at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. Among them, the processor 501, the memory 502, and the communication component 503 are connected through a bus 504.

[0164] In a specific implementation process, at least one processor 501 executes computer-executable instructions stored in a memory 502, so that at least one processor 501 executes the above-mentioned method.

[0165] For the specific implementation process of the processor 501, reference may be made to the above method embodiment, and its implementation principle and technical effects are similar, so they will not be elaborated here in this embodiment.

[0166] In the above embodiment, it should be understood that the processor may be a central processing unit (Central Processing Unit, CPU for short), or other general-purpose processors, digital signal processors (Digital Signal Processor, DSP for short), application specific integrated circuits (Application Specific Integrated Circuit, ASIC for short), etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the invention can be directly embodied as being executed by a hardware processor, or executed by a combination of hardware and software modules in the processor.

[0167] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (Non-volatile Memory, NVM), such as at least one disk memory.

[0168] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, the buses in the drawings of this application are not limited to only one bus or one type of bus.

[0169] This application also provides a computer program product, including a computer program, which implements the above-mentioned method when executed by a processor.

[0170] This application also provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the processor executes the computer-executable instructions, the above-mentioned method is implemented.

[0171] The above-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible by a general-purpose or special-purpose computer.

[0172] An exemplary readable storage medium is coupled to the processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in a device.

[0173] The division of units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be an indirect coupling or communication connection through some interfaces, devices, or units, and can be in electrical, mechanical, or other forms.

[0174] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0175] In addition, the functional units in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0176] If a function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present invention. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, and other various media that can store program codes.

[0177] Those of ordinary skill in the art can understand that all or part of the steps for implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When this program is executed, it executes the steps including the above method embodiments; and the foregoing storage medium includes: ROMs, RAMs, magnetic disks, or optical discs, and other various media that can store program codes.

[0178] Finally, it should be noted that: After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily think of other implementation manners of the present invention. The present invention is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include the common general knowledge or conventional technical means in the technical field not disclosed in the present invention. It is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A method for analyzing an autonomous driving event, characterized in that: The method comprises: In response to a user's operation of starting an analysis indicator, obtaining a natural language input by the user, wherein the natural language includes a data storage path, a signal structure, and an event trigger determination condition; Generate a data parsing code according to the signal structure, wherein the data parsing code is used to parse the signal data of the signal structure; Generate application services based on the data parsing code and a preset system interface specification; wherein the system interface specification defines the rules, format and sequence of interaction and communication between system components; Calling the application service to obtain signal data in machine language from the data storage path, and parsing the signal data into natural language data; Determining target data satisfying the event trigger determination condition from the natural language data; A sample report is generated according to the target data, and the sample report is output in the dialog of the analysis indicator.

2. The method according to claim 1, characterized in that The method further comprises: In the dialog of analyzing the indicator, receiving a new data storage path input by a user; Generate a new sample report according to the new data storage path; The new sample report is output.

3. The method according to claim 1 or 2, characterized in that: The signal structure includes a signal interface name and a signal declaration.

4. The method according to claim 3, characterized in that: The signal structure is in the form of an example signal.

5. The method according to claim 4, characterized in that The example signal is natural language in an image; Accordingly, the method further includes: An example signal representing a signal structure is extracted from the image.

6. The method according to claim 1 or 2, characterized in that: The method further comprises: Get the report keywords entered by the user; The report keyword is searched in all stored sample reports, and the retrieved sample reports are output.

7. The method according to claim 1 or 2, characterized in that: The sample report includes target data and associated data, and the associated data is natural language data within a preset time period before and after the target data.

8. A device for analyzing an automatic driving event, characterized in that: The device comprises: An acquisition module, configured to acquire a natural language input by a user in response to an operation of starting an analysis indicator by the user, wherein the natural language includes a data storage path, a signal structure, and an event trigger determination condition; A code generation module, used for generating a data parsing code according to the signal structure, wherein the data parsing code is used for parsing signal data of the signal structure; An application service generation module, used to generate application services based on the data parsing code and a preset system interface specification; wherein the system interface specification defines the rules, format and sequence of interaction and communication between system components; A parsing module, calling the application service, obtaining signal data in machine language from the data storage path, and parsing the signal data into natural language data; A data determination module, used to determine target data satisfying the event trigger determination condition from the natural language data; A report generation module, used to generate a sample report according to the target data; An output module is used to output the sample report in the dialog of the analysis indicator.

9. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

Images