Data processing method, electronic equipment and storage medium

Abstract

The invention relates to the technical field of vehicles, in particular to a data processing method, electronic equipment and a storage medium. The method comprises the following steps: acquiring to-be-analyzed driving data; the to-be-analyzed driving data is data collected when the vehicle is abnormal; the to-be-analyzed driving data comprises environment perception data, driving decision data and execution action data; performing scene recognition based on the driving decision data to obtain a target scene corresponding to the to-be-analyzed driving data; obtaining an analysis rule corresponding to the target scene; and based on an analysis rule corresponding to the target scene, performing anomaly analysis on the environment perception data, the driving decision data and the execution action data to obtain an analysis result. The target scene corresponding to the to-be-analyzed driving data is obtained through scene recognition to obtain the analysis rule corresponding to the target scene, so that the to-be-analyzed driving data can be automatically analyzed based on the analysis rule, the data analysis efficiency and the accuracy of an analysis result are improved, and meanwhile, the analysis cost is reduced.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to a data processing method, an electronic device and a storage medium. Background Art

[0002] With the advancement of science and technology, autonomous driving technology has made rapid progress. Although there is still some distance to the full application of advanced autonomous driving technology (L3 and above), basic autonomous driving technology, such as Advanced Driving Assistance System (ADAS), has been widely used. ADAS uses a variety of sensors installed on the vehicle (such as millimeter wave radar, lidar, single / binocular camera, satellite navigation, etc.) to perceive the surrounding environment during vehicle driving, and combines navigation map data to perform systematic calculations and analysis on the perceived data, so that the driver can be aware of possible dangers in advance, which can effectively increase the comfort and safety of vehicle driving.

[0003] Before ADAS is put into mass production and put on the market, its reliability usually needs to be tested. During the test, a test vehicle equipped with ADAS is usually used to drive on a test road to obtain road test data, and then the reliability of ADAS is evaluated by analyzing the road test data. In related technologies, when analyzing road test data, it is usually analyzed manually. However, this analysis method not only has high analysis costs and low analysis efficiency, but also has difficulty in ensuring accuracy. Summary of the invention

[0004] In order to solve the above technical problems, the embodiments of the present application provide a data processing method, an electronic device and a storage medium.

[0005] In a first aspect, an embodiment of the present application provides a data processing method, the method comprising:

[0006] Acquire driving data to be analyzed; the driving data to be analyzed is data collected when an abnormality occurs in the vehicle; the driving data to be analyzed includes environmental perception data, driving decision data, and execution action data;

[0007] Perform scene recognition based on driving decision data to obtain the target scene corresponding to the driving data to be analyzed;

[0008] Obtain the analysis rules corresponding to the target scenario;

[0009] Based on the analysis rules corresponding to the target scenario, an abnormal analysis is performed on the environmental perception data, driving decision data and execution action data to obtain the analysis results.

[0010] In some optional embodiments, obtaining driving data to be analyzed includes:

[0011] Acquire original driving data; the original driving data is data encoded according to a preset encoding format;

[0012] The original driving data is decoded and processed to obtain the driving data to be analyzed.

[0013] In some optional embodiments, scene recognition is performed based on the driving decision data to obtain a target scene corresponding to the driving data to be analyzed, including:

[0014] Matching the driving decision data with the scene recognition rules in the scene recognition rule library to obtain a matching result; the scene recognition rule library is configured with at least one scene recognition rule, and each scene recognition rule corresponds to a preset scene;

[0015] When the matching result indicates that the driving decision data matches the target scene recognition rule, the preset scene corresponding to the target scene recognition rule is determined as the target scene corresponding to the driving data to be analyzed; the target scene recognition rule is one of the at least one scene recognition rules.

[0016] In some optional embodiments, the driving data to be analyzed further includes a vehicle version identifier; based on the analysis rules corresponding to the target scenario, an abnormal analysis is performed on the environmental perception data, the driving decision data, and the execution action data to obtain analysis results, including:

[0017] Based on the vehicle version identifier, obtain vehicle configuration information of the vehicle;

[0018] Based on the analysis rules and vehicle configuration information, an abnormal analysis is performed on the environmental perception data, driving decision data and execution action data to obtain the analysis results.

[0019] In some optional embodiments, the driving data to be analyzed also includes environmental perception true value data, and the environmental perception true value data is used to verify the accuracy of the environmental perception data; based on the analysis rules and the vehicle configuration information, the environmental perception data, the driving decision data, and the execution action data are analyzed for abnormalities to obtain analysis results, including:

[0020] Compare the environmental perception true value data with the environmental perception data to obtain a comparison result;

[0021] When the comparison result indicates that the environmental perception data is accurate, an abnormal analysis is performed on the driving decision data and the execution action data based on the vehicle configuration information and the environmental perception data to obtain an analysis result.

[0022] In some optional embodiments, when the comparison result indicates that the environmental perception data is accurate, the driving decision data and the execution action data are analyzed for abnormalities based on the vehicle configuration information and the environmental perception data to obtain analysis results, including:

[0023] When the comparison result indicates that the environmental perception data is accurate, determining an expected motion model based on the environmental perception data;

[0024] Based on the expected motion model and vehicle configuration information, the driving decision data and execution action data are analyzed for abnormalities to obtain analysis results.

[0025] In some optional embodiments, the method further comprises:

[0026] Generate analysis reports based on target scenarios and analysis results.

[0027] In some optional embodiments, the analysis result includes the abnormal cause of the abnormality; and an analysis report is generated based on the target scenario and the analysis result, including:

[0028] Obtain the knowledge base corresponding to the target scenario;

[0029] Identify improvement suggestions in the knowledge base based on the cause of the anomaly;

[0030] Generate analysis reports based on target scenarios, abnormal causes and improvement suggestions.

[0031] In a second aspect, an embodiment of the present application provides an electronic device, comprising a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded by the processor and executes the data processing method as described above.

[0032] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, in which at least one instruction or at least one program is stored, and the at least one instruction or at least one program is loaded and executed by a processor to implement the data processing method as described above.

[0033] In a fourth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the data processing method described above.

[0034] The technical solution provided by the embodiment of the present application has the following technical effects:

[0035] The data processing method described in the embodiment of the present application obtains the target scene corresponding to the driving data to be analyzed through scene recognition to obtain the analysis rules corresponding to the target scene, so that the driving data to be analyzed can be automatically analyzed based on the analysis rules, thereby improving the data analysis efficiency and the accuracy of the analysis results, while reducing the analysis cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1A schematic diagram of an exemplary application scenario provided for an embodiment of the present application;

[0037] Figure 2 A schematic diagram of a data analyzer provided in an embodiment of the present application;

[0038] Figure 3 A schematic diagram of an interface of a data analyzer provided in an embodiment of the present application;

[0039] Figure 4 A flowchart of a data processing method provided in an embodiment of the present application;

[0040] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

[0041] Figure 6 A hardware structure block diagram of a server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0043] It should be noted that the terms "first", "second", etc. in the specification and claims of the embodiments of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. Thus, the features defined as "first" and "second" may expressly or implicitly include one or more of the features. In addition, unless otherwise specified, the meaning of "multiple" is two or more. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or server comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0044] In order to make the purpose, technical solution and advantages disclosed in the embodiments of the present application more clearly understood, the embodiments of the present application are further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the embodiments of the present application and are not used to limit the embodiments of the present application.

[0045] At present, vehicles are generally equipped with ADAS, such as Autonomous Emergency Braking (AEB) system, Lane Departure Warning (LDW) system, Adaptive Cruise Control (ACC) system, and Emergency Lane Keeping (ELK) system. In order to ensure the safety and comfort of the vehicle, the developed ADAS needs to be tested during the ADAS development process. When testing ADAS, the developed ADAS is usually installed on a test vehicle, and then the test vehicle is used for road testing to obtain ADAS road test data. After obtaining the road test data, the safety and reliability of the developed ADAS are determined by analyzing the road test data. At present, due to the variety of ADAS, when analyzing road test data, manual analysis is generally used to adapt to the analysis requirements of different ADAS. However, this analysis method not only has high analysis cost and low analysis efficiency, but also the analysis process is prone to errors.

[0046] In view of this, an embodiment of the present application provides a data processing method, which can analyze a large amount of driving data in an automated manner, thereby improving data analysis efficiency and ensuring the accuracy of analysis results.

[0047] The application scenarios of the embodiments of the present application are introduced below.

[0048] Figure 1 A schematic diagram of an exemplary application scenario provided in the embodiment of the present application, refer to Figure 1 As shown, the application scenario may include a vehicle 10 and an electronic device 20 .

[0049] The vehicle 10 may be provided with a perception system for perceiving the environment, such as a laser radar, a millimeter wave radar, a single / binocular camera, an infrared thermal imaging sensor, a locator (such as a GPS locator), etc. The vehicle 10 may also be equipped with an automatic driving system to be tested, such as an ADAS system. Road test data may be obtained by controlling the vehicle 10 to travel on a road.

[0050] The electronic device 20 may be installed with an application program for analyzing the road test data, such as a data analyzer, so that the road test data can be analyzed. Optionally, the electronic device includes but is not limited to a desktop computer, a laptop computer, a tablet computer, a mobile phone, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), etc. When the electronic device 20 analyzes the road test data, the road test data can be input into the data analyzer, and the data analyzer can automatically analyze the road test data and automatically generate an analysis report. In addition, the data analyzer can also visualize the analysis report.

[0051] Specifically, Figure 2 A schematic diagram of a data analyzer provided in an embodiment of the present application, referring to Figure 2 As shown, the road test data for data analysis may be data obtained under a variety of test scenarios, such as AEB test scenarios, LDW test scenarios, ACC test scenarios, ELK test scenarios, etc. The content contained in the road test data may include test scenario data and true value data. The test scenario data may include vehicle version identification, environmental perception data, driving decision data, execution action data and other data. The true value data may include environmental perception true value data. The true value data can be used as a set of control data to verify the accuracy of environmental perception data. The test scenario data and the true value data may be data encoded in a preset encoding format, thereby ensuring the security of the data and facilitating the transmission and storage of the data.

[0052] By inputting the test scenario data and true value data into the data analyzer in the electronic device, the automated analysis of the perception data, driving decision data, and execution action data can be achieved. Automated analysis means that the data analyzer can analyze a large amount of test data (i.e., perception data, driving decision data, and execution action data) in an automated manner, thereby improving the efficiency of the analysis. Automated analysis can include steps such as signal processing, scenario detection, and false trigger detection, thereby reducing the burden of manual analysis.

[0053] refer to Figure 2 As shown in the figure, the data analyzer has data adaptability, which can adaptively analyze road test data generated by different vehicle models, different test environments and different test scenarios. Through flexible configuration and parameter adjustment, it can adapt to diverse test data sources. In addition, the data analyzer can also automatically filter and align data. Automatically filtering and aligning data means automatically aligning the installation timestamps of perception data, driving decision data, execution action data and other data.

[0054] refer to Figure 2 As shown, the data analyzer can first perform scene classification detection during the automated analysis of perception data, driving decision data, and execution action data. The data analyzer can automatically identify and classify different preset scenarios, such as AEB test scenarios, LDW test scenarios, etc. For each preset scenario, the data analyzer can execute specific analysis rules to analyze the cause of the abnormality. When the data analyzer analyzes the perception data, driving decision data, and execution action data, it can adopt partial analysis rules for different ADAS. In some embodiments, machine learning models, pattern recognition technology, or other advanced algorithms can also be combined to accurately detect and identify the cause of ADAS abnormalities.

[0055] refer to Figure 2 As shown in the figure, in order to improve the efficiency of data analysis, the data analyzer can efficiently process and store a large amount of road test data through data batch processing, and can maintain efficient performance even when processing tens of thousands of road test data. The data analyzer can generate customized analysis reports, each of which corresponds to a road test data. The analysis report can clearly point out the cause of the abnormality, the corresponding preset scenario, and the recommended improvement measures, etc., so as to provide more specific guidance and information. In addition, the data analyzer can provide a user-friendly interface and visualization tools so that users can intuitively understand the analysis results. The analysis results can be displayed in interactive charts, graphical interfaces, etc., so that users can more easily understand and use the analysis results output by the data analyzer.

[0056] Figure 3 This is a schematic diagram of an interface of a data analyzer provided in an embodiment of the present application, referring to Figure 3As shown, the interface displayed by the data analyzer has a control for selecting the original driving data folder. The user can trigger the control to select the storage path of the original driving data folder. The original driving data folder stores the original driving data (i.e., road test data). By selecting the storage path of the original driving data folder, the original driving data can be imported into the data analyzer. At the same time, the interface displayed by the data analyzer also provides an analysis report saving path control, and the user can trigger the control to select the storage path for generating the analysis report. In the preset scene parameter column, the user can set the sensor deviation (Dx) and the vehicle's own width (Ego-width). After the file path selection is completed and the parameter setting is completed, the user can click the start control to enable the data analyzer to automatically analyze the original driving data and generate an analysis report. The report finally generated can be displayed in the generated analysis report list column in the form of a list. Optionally, the generated analysis report can be a preset file format, such as pdf format, jpg format, xml format, html format, etc. The user can browse the specific content of the corresponding analysis report by clicking the analysis report displayed in the generated analysis report list column.

[0057] The following is an introduction to the data processing method provided in the embodiments of the present application.

[0058] The data processing method provided in the embodiment of the present application can be based on Figure 2 For ease of understanding, the following takes the analysis of road test data generated during ADAS road testing as an example to introduce in detail the data processing method provided in the embodiment of the present application. It should be understood that the data processing method provided in the embodiment of the present application can also perform data analysis on autonomous driving data in other scenarios.

[0059] Figure 4 A flowchart of a data processing method provided in an embodiment of the present application, which can be executed by an electronic device, such as Figure 4 As shown, the method may include:

[0060] S101: Acquire driving data to be analyzed; the driving data to be analyzed is data collected when an abnormality occurs in the vehicle; the driving data to be analyzed includes environmental perception data, driving decision data, and execution action data.

[0061] In the embodiment of the present application, the driving data to be analyzed is the data collected when the vehicle is abnormal. Vehicle abnormality means that when the system with automatic driving function (such as ADAS) in the vehicle is turned on, the vehicle has the automatic driving function of the system mis-triggered, no triggering or the vehicle has an accident. At this time, the data when the abnormality occurs is analyzed to determine the cause of the abnormality.

[0062] For ADAS road testing, when conducting road testing on the ADAS to be tested (such as AEB system, LDW system, ACC system, etc.), the ADAS to be tested can be installed on the test vehicle, and the corresponding road test data can be collected by controlling the test vehicle to drive on the road. In the process of collecting road test data, if the test vehicle has an abnormal situation, such as false triggering, no triggering of the ADAS to be tested, or an accident of the test vehicle, the road test data within a period of time (such as 10 minutes) before and after the abnormal situation occurs can be extracted for analysis to determine the cause of the abnormality. For the ADAS to be tested, by analyzing the road test data and determining the cause of the abnormality, it can provide a reference for the optimization and adjustment of the ADAS to be tested, thereby improving the performance of the ADAS.

[0063] In an embodiment of the present application, the driving data to be analyzed may be obtained by decoding the original driving data (e.g., road test data). The original driving data is data collected when an abnormality occurs in the vehicle. The original driving data may be data encoded in a preset encoding format (e.g., a binary encoding format), and the driving data to be analyzed may be data obtained by decoding the original driving data.

[0064] During the ADAS road test, the environmental perception system in the vehicle will perceive the environment and obtain environmental perception data, such as radar data collected by millimeter-wave radar, image data collected by single / binocular cameras, positioning data collected by locators, etc. At the same time, the ADAS to be tested will make decisions based on the environmental perception data during the road test and generate decision data, such as system operation logs, received / sent command signals, etc. In addition, the actuators on the test vehicle (such as the power system, braking system, etc.) will generate actuator data, such as received command signals, actions performed according to command signals, changes in the vehicle's operating state after the action is performed, and other data. In order to ensure the security of these data, these data are usually encoded according to preset encoding rules to obtain the original driving data, and then the ADAS road test obtained after encoding can be processed by data storage, transmission, analysis, etc.

[0065] Therefore, when performing abnormality analysis, the original driving data can be obtained first, and then the driving data to be analyzed can be obtained by decoding the original driving data. As an example, the original driving data is binary data encoded in a binary encoding format. After the original driving data is obtained, the original driving data is decoded according to the decoding rules to decode the meaningless binary data into environmental perception data, driving decision data, and execution action data with physical meaning. By analyzing the environmental perception data, driving decision data, and execution action data, the cause of the abnormality is determined.

[0066] Generally speaking, the amount of original driving data is large. For a large amount of original driving data, the embodiment of the present application uses a data batch processing method to decode the original driving data.

[0067] In the embodiment of the present application, after decoding the original driving data to obtain the driving data to be analyzed, the environmental perception data, driving decision data and execution action data in the driving data to be analyzed can also be aligned by batch processing to facilitate subsequent data analysis.

[0068] The environmental perception data, driving decision data and execution action data in the driving data to be analyzed usually carry the timestamp of data generation. After obtaining the driving data to be analyzed, the environmental perception data, driving decision data and decision feedback data can be aligned according to the timestamps through batch processing, thereby improving the efficiency of subsequent data analysis.

[0069] S102: Perform scene recognition based on the driving decision data to obtain a target scene corresponding to the driving data to be analyzed.

[0070] In the embodiment of the present application, after the driving data to be analyzed is obtained, scene recognition may be performed on the driving data to be analyzed to determine a target scene corresponding to the driving data to be analyzed.

[0071] Specifically, the data processing method provided in the embodiment of the present application can analyze the driving data generated under a variety of preset scenarios (such as AEB test scenarios, ACC test scenarios), and different preset scenarios correspond to different analysis rules. For example, in the AEB test scenario, the driving data to be analyzed corresponds to the AEB system, and the AEB system has its own unique decision-making strategy. Therefore, when performing abnormal analysis on the environmental perception data, driving decision data, and execution action data generated by the AEB test scenario, it is necessary to use the analysis rules corresponding to the AEB test scenario for analysis. In other words, since different preset scenarios correspond to different analysis rules, after obtaining the driving data to be analyzed, the preset scenario corresponding to the driving data to be analyzed can be identified first, and then the identified preset scenario can be used to obtain the corresponding analysis rules, so that the environmental perception data, driving decision data, and execution action data can be analyzed for abnormalities based on the analysis rules.

[0072] When identifying the preset scene corresponding to the driving data to be analyzed, the preset scene recognition can be performed based on the driving decision data in the driving data to be analyzed. Specifically, the driving decision data is matched with the scene recognition rules in the scene recognition rule library to obtain a matching result. Among them, the scene recognition rule library is configured with one or more scene recognition rules, and the scene recognition rules correspond to the preset scenes, that is, one scene recognition rule corresponds to one preset scene, and the same preset scene can correspond to one or more scene recognition rules. In the case where the matching result indicates that the driving decision data matches the target scene recognition rule, the preset scene corresponding to the target scene recognition rule is determined as the target scene corresponding to the driving data to be analyzed. The target scene recognition rule is a scene recognition rule in the scene recognition rule library.

[0073] That is to say, after obtaining the driving data to be analyzed, the driving decision data in the driving data to be analyzed can be matched one by one with the scene recognition rules in the scene recognition rule library. If the driving decision data matches a certain scene recognition rule, the preset scene corresponding to the scene recognition rule can be used as the target scene corresponding to the driving data to be analyzed. As an example, a certain scene recognition rule in the scene recognition rule library is: the AEB system sends a braking command to the vehicle braking system, and the preset scene corresponding to the scene recognition rule is the AEB test scene. If the driving decision data in the driving data to be analyzed is driving decision data, the AEB system sends a braking command to the vehicle braking system, then it can be determined that the target scene corresponding to the driving data to be analyzed is the AEB test scene.

[0074] In some embodiments, in order to improve the accuracy of preset scene recognition, when the preset scene corresponding to the driving data to be analyzed is recognized, the preset scene recognition can also be performed based on the environmental perception data, driving decision data and / or execution action data in the driving data to be analyzed. In the recognition process, rule matching can also be used to implement it.

[0075] In other embodiments, when identifying the preset scenario corresponding to the driving data to be analyzed, logical reasoning may be performed based on the environmental perception data, the driving decision data and / or the execution action data, and the scenario may be recognized based on the logical reasoning result.

[0076] As an example, in an actual scenario, a vehicle is driving on a city street and suddenly finds a stationary obstacle ahead, but the driver does not brake immediately. In the driving data to be analyzed corresponding to the actual scenario, the environmental perception data may include: the obstacle detection parameter is true (e.g., Obstacle_Detected = True); the speed of the obstacle relative to the vehicle is 0 (e.g., Obstacle_Relative_Speed ​​= 0km / h); the distance between the obstacle and the vehicle is 25m (e.g., Obstacle_Distance = 25m). The execution action data may include: no braking signal is received (e.g., Driver_Braking = False). The driving decision data may include: the obstacle detection parameter is true, indicating that there is a stationary obstacle ahead; the relative speed of the obstacle is 0, indicating that the obstacle is stationary; the distance between the obstacle and the vehicle is 25m, indicating that the obstacle is close to the vehicle; the collision time calculated based on the current speed of the vehicle is 1.5s (e.g., Time_To_Collision = 1.5s); it is determined that the collision time prediction is less than the AEB trigger threshold and the driver does not actively step on the brake. Logical reasoning is performed based on the above environmental perception data, driving decision data, and execution action data to obtain the following reasoning result: the vehicle is close to the obstacle, a high collision risk is detected, and the driver does not take active braking action. Based on the reasoning result, the preset scenario corresponding to the driving data to be analyzed is identified as the AEB test scenario.

[0077] As another example, in an actual scenario, a vehicle is cruising on a highway, and there is a vehicle traveling at a slightly lower speed in front of it, and the vehicle turns on the ACC system. In the driving data to be analyzed corresponding to the actual scenario, the environmental perception data may include: the detection of the vehicle in front is true (e.g., Front_Vehicle_Present = True); the relative speed between the vehicle in front and the vehicle is -5km / h (e.g., Front_Vehicle_Relative_Speed ​​= -5km / h); the distance between the vehicle and the vehicle in front is 55m (e.g., Front_Vehicle_Distance = 55m). The execution action data may include: the current set speed of the vehicle (e.g., Current_Speed_Setpoint) is 80km / h, and the actual speed of the vehicle is 85km / h. The driving decision data may include: the detection of the vehicle ahead is true, indicating that there is a vehicle ahead; the relative speed of the vehicle ahead is -5km / h, indicating that the speed of the vehicle ahead is slightly lower than that of the vehicle; the distance between the vehicle ahead and the vehicle ahead is 55m, indicating that the distance between the vehicles is within the ACC safety range; the current set speed of the vehicle (e.g., Current_Speed_Setpoint) is 80km / h, and the actual speed of the vehicle is 85km / h, indicating that the current set speed of the vehicle is close to the actual speed of the vehicle. Logical reasoning is performed based on the above environmental perception data, driving decision data, and execution action data to obtain the reasoning result: the vehicle ahead is detected and there is no collision threat, and the vehicle is maintaining a safe following distance at cruising speed. According to the reasoning result, the preset scene corresponding to the driving data to be analyzed is identified as the ACC test scene.

[0078] In the above examples, after obtaining the reasoning results, the reasoning results and the data involved in the reasoning process (environmental knowledge data, driving decision data and / or execution action data) can also be recorded, and an event record can be generated, so that in the subsequent data analysis process, there is no need to repeatedly extract relevant data and repeat reasoning, thereby improving data analysis efficiency and saving computing resources. For example, in the above-mentioned AEB test scenario reasoning process, the time, location, obstacle location and other information of the AEB test scenario can be recorded. The relevant data is automatically marked as an AEB event and an event record is generated for subsequent analysis. Similarly, for the above-mentioned ACC test scenario reasoning process, the current cruising speed, relative distance, time and other information can be recorded, the relevant data can be marked as an ACC event and an event record can be generated for subsequent analysis.

[0079] S103: Obtain analysis rules corresponding to the target scenario.

[0080] In the embodiment of the present application, after the target scene corresponding to the driving data to be analyzed is identified, the analysis rules corresponding to the target scene can be obtained from the analysis rule library. For example, if the target scene is an AEB test scene, after the target scene is identified, the analysis rules corresponding to the AEB test scene can be obtained from the analysis rule library, so that the environmental perception data, driving decision data, and execution action data can be analyzed for abnormalities based on the analysis rules corresponding to the AEB test scene.

[0081] It should be noted that the analysis rule library is provided with analysis rules corresponding to a variety of preset scenarios, and each preset scenario may correspond to one or more analysis rules.

[0082] S104: Based on the analysis rules corresponding to the target scenario, an abnormality analysis is performed on the environmental perception data, the driving decision data, and the execution action data to obtain an analysis result.

[0083] In an embodiment of the present application, after obtaining the analysis rules corresponding to the target scene, the obtained analysis rules can be used to perform abnormal analysis on the environmental perception data, driving decision data, and execution action data to obtain analysis results.

[0084] For vehicles, different models of vehicles often have different configurations (such as vehicle size, weight, sensors, etc.), and vehicles with different configurations often have different motion characteristics. In addition, even for vehicles of the same model, the versions of the systems in the vehicles (such as ADAS versions) may be different. Therefore, the causes of abnormalities in vehicles of different models and different system versions may be different. Therefore, in order to improve the accuracy of the analysis results, when performing abnormal analysis on environmental perception data, driving decision data, and execution action data based on analysis rules, the vehicle configuration information of the test vehicle can be obtained first, such as the length, width, and height of the vehicle, the location and number of sensors, the power system parameters, the braking system parameters, etc., so that the analysis rules and vehicle configuration information can be combined to perform abnormal analysis on environmental perception data, driving decision data, and execution action data.

[0085] Specifically, the database may be pre-configured with vehicle configuration information of various models and various system versions of vehicles. At the same time, the driving data to be analyzed contains a vehicle version identifier. The vehicle version identifier can be decoded in the above decoding process of the original driving data. When performing abnormal analysis on environmental perception data, driving decision data and execution action data, the vehicle configuration information corresponding to the vehicle version identifier can be obtained from the database based on the vehicle version identifier, and then the environmental perception data, driving decision data and execution action data are analyzed abnormally based on the analysis rules and vehicle configuration information to obtain analysis results. By pre-configuring the vehicle configuration information of various models and various system versions of vehicles in the database, it is possible to perform abnormal analysis on environmental perception data, driving decision data and execution action data generated by vehicles of different models and different system versions, thereby improving the scope of application of data analysis. At the same time, the vehicle configuration information helps to identify and evaluate the motion characteristics and safety of the vehicle in different preset scenarios. For example, the type and installation position of the sensor installed on the vehicle will affect the detection accuracy and coverage of the vehicle to obstacles, which has an important impact on the decision-making process of the ADAS to be tested. Therefore, by combining vehicle configuration information and analysis rules to perform abnormal analysis on environmental perception data, driving decision data, and execution action data, the accuracy of the analysis results can be improved.

[0086] The analysis results include the cause of the abnormality. When performing abnormality analysis on the environmental perception data, driving decision data, and execution action data in combination with the vehicle configuration information and analysis rules, it is possible to first determine whether the environmental perception data in the driving data to be analyzed is reliable, thereby determining whether the cause of the abnormality of the vehicle is caused by unreliable environmental perception data.

[0087] Specifically, in some scenarios, for example, when the test vehicle is tested on a flat road, the millimeter-wave radar on the test vehicle will not be blocked by the road surface when sensing the environment, so the collected data is more reliable. However, when the test vehicle is tested on a rugged road, if the road slope is large, the undulating road surface may interfere with the work of the millimeter-wave radar, resulting in large errors in the data collected by the millimeter-wave radar. At this time, if the test vehicle is abnormal, the cause of the abnormality may be due to inaccurate detection data of the millimeter-wave radar.

[0088] In order to accurately determine whether the cause of the vehicle abnormality is due to inaccurate environmental perception data, when performing abnormality analysis on environmental perception data, driving decision data and execution action data, the environmental perception true value data can be used to verify the environmental perception data to determine whether the environmental perception data in the driving data to be analyzed is accurate, thereby determining whether the cause of the vehicle abnormality is inaccurate environmental perception data.

[0089] Specifically, the driving data to be analyzed also includes environmental perception true value data, which is used to verify the accuracy of environmental perception data. Environmental perception true value data can also be collected during the road test of the ADAS to be tested. Optionally, environmental perception true value data can be collected using a more accurate perception method, or obtained through a pre-designed test method. Taking distance perception in environmental perception data as an example, a laser radar is additionally installed on the test vehicle to collect the distance to the front target (such as an obstacle) as environmental perception true value data. Alternatively, during the road test of the ADAS to be tested, other vehicles can be controlled to travel in front of the test vehicle. After the road test is completed, the distance to the front vehicle is calculated as environmental perception true value data by obtaining the positioning data of the locator (such as a GPS locator) on other vehicles and the positioning data collected by the locator on the test vehicle. Similarly, the original driving data contains environmental perception true value data, and the environmental perception true value data can be obtained by parsing the original driving data. When performing anomaly analysis on environmental perception data, driving decision data, and execution action data, the environmental perception true value data and the environmental perception data may be compared first to obtain a comparison result. If the comparison result indicates that the environmental perception data is inconsistent with the environmental perception true value data, it may be determined that the cause of the abnormality that caused the abnormality includes inaccurate environmental perception data. If the comparison result indicates that the environmental perception data is accurate, the driving decision data and the execution action data may be further analyzed for abnormalities based on the vehicle configuration information and the environmental perception data to obtain an analysis result.

[0090] When determining whether the cause of the vehicle abnormality is inaccurate environmental perception data, it can also be determined by using sensor data fusion. Sensor data fusion confirmation refers to the consistency verification of environmental perception data collected by different sensors on the vehicle at the same timestamp to determine whether the sensor has failed. As an example, in the AEB test scenario, the AEB braking trigger is often based on the environmental perception data of multiple sensors for decision making. AEB braking is falsely triggered, which may occur when abnormal data appears in a sensor and other sensors do not provide consistent verification. For example, the test vehicle is driving at a normal high speed of 100km / h. The camera on the test vehicle detects that there is an obstacle very close to the test vehicle (for example, less than 40m) in the lane where the test vehicle is located, and the obstacle is traveling at a speed of 40km / h, but the millimeter-wave radar or lidar on the test vehicle does not detect the obstacle. If the AEB braking is triggered, the situation can be marked as a false trigger, and it is determined that the abnormal cause of the AEB braking false trigger is due to inaccurate environmental perception data.

[0091] If it is determined that the vehicle abnormality is not caused by inaccurate environmental perception data, that is, if the environmental perception data is accurate, it can be further analyzed whether the cause of the abnormality is that the decision algorithm of the ADAS to be tested is unreasonable.

[0092] Specifically, when the environmental perception data is accurate, the decision algorithm of the ADAS to be tested can be further analyzed to see whether it is unreasonable by determining the expected motion model. The expected motion model refers to the decision-making process of the ADAS to be tested to be re-mimicked based on the perception data in the driving data to be analyzed based on the target scene, so as to determine whether the driving decision data in the driving data to be analyzed is wrong. That is to say, when the environmental perception data is accurate, the expected motion model can be determined based on the environmental perception data. Then, based on the expected motion model and the vehicle configuration information, the driving decision data and the execution action data are analyzed for abnormalities to obtain the analysis results. As an example, in the AEB test scenario, when predicting the motion trajectory of the front object, the AEB to be tested will make a prediction based on the moving speed and direction of the front object. If it is predicted according to the expected motion model that the front object and the test vehicle will not intersect (such as pedestrians or vehicles moving in a direction perpendicular to the driving direction of the test vehicle), the AEB to be tested should not trigger braking. If the driving decision data in the driving data to be analyzed includes that the front object and the test vehicle may intersect, or the AEB to be tested triggers braking, it can be determined that the abnormal cause is that the decision algorithm of the ADAS to be tested is unreasonable.

[0093] In the embodiment of the present application, after obtaining the analysis results, an analysis report can also be generated based on the target scenario and the analysis results. The analysis report corresponds to the original driving data, that is, each analysis report corresponds to one (road test data). When generating an analysis report based on the target scenario and the analysis results, the target scenario corresponding to the driving data to be analyzed, the event record, and the abnormal cause and other information are summarized to generate an analysis report.

[0094] In some embodiments, in addition to indicating the identified target scenario and abnormal cause, the analysis report may also include improvement suggestions to provide more specific guidance and information. Therefore, when generating an analysis report, improvement suggestions may also be determined based on the abnormal cause and added to the analysis report. Specifically, when generating an analysis report, a knowledge base (such as a knowledge graph) corresponding to the target scenario may be obtained, and then improvement suggestions may be determined in the knowledge base based on the abnormal cause, and an analysis report may be generated based on the target scenario, the abnormal cause, and the improvement suggestions.

[0095] In an embodiment of the present application, the analysis report can summarize the identified target scene and whether the ADAS to be tested has false triggering and other analysis contents. The analysis report may include target scenes, vehicle actuator data, environmental perception data, event records, abnormal causes, improvement suggestions and other contents. For improvement suggestions, they can be found in the knowledge base according to the abnormal causes. For example, the abnormal cause may be attributed to insufficient sensor accuracy, insufficient recognition conditions of the ADAS decision algorithm to be tested, etc. The corresponding improvement suggestions may be to improve the type of sensor, adjust the trigger threshold or optimize the ADAS decision algorithm, etc.

[0096] In the embodiment of the present application, after the analysis report is generated, the analysis report can also be visualized.

[0097] An embodiment of the present application also provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded by the processor and executes the data processing method as described above.

[0098] The data processing method provided in the embodiment of the present application can be applied to any electronic device, including but not limited to a mobile station (MS), a mobile terminal (MT), etc. For example, the electronic device can be a mobile phone, a smart TV, a wearable device, a tablet computer (Pad), a desktop computer, a laptop computer, a virtual reality (VR) device, an augmented reality (AR) device, a terminal in industrial control, a terminal in self-driving, a terminal in remote medical surgery, a terminal in a smart grid, a terminal in transportation safety, a terminal in a smart city, a terminal in a smart home, etc. The embodiment of the present application does not limit the specific form of the electronic device.

[0099] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown in FIG. Figure 5As shown, the electronic device may include one or more processors 504, a system control logic 508 connected to at least one of the processors 504, a system memory 512 connected to the system control logic 508, a non-volatile memory (NVM) 516 connected to the system control logic 508, and a network interface 520 connected to the system control logic 508.

[0100] In some embodiments, the processor 504 may include one or more single-core or multi-core processors. In some embodiments, the processor 504 may include any combination of general-purpose processors and special-purpose processors (eg, graphics processors, application processors, baseband processors, etc.).

[0101] In some embodiments, system control logic 508 may include any suitable interface controller to provide any suitable interface to at least one of processors 504 and / or any suitable device or component in communication with system control logic 508 .

[0102] In some embodiments, the system control logic 508 may include one or more memory controllers to provide an interface to the system memory 512. The system memory 512 may be used to load and store data and / or instructions. In some embodiments, the memory 512 of the electronic device may include any suitable volatile memory, such as a suitable dynamic random access memory (DRAM).

[0103] The non-volatile memory 516 may include one or more tangible, non-transitory readable storage media for storing data and / or instructions. In some embodiments, the non-volatile memory 516 may include any suitable non-volatile memory such as flash memory and / or any suitable non-volatile storage device, such as a hard disk drive (HDD), a compact disc (CD) drive, and a digital versatile disc (DVD) drive.

[0104] The non-volatile memory 516 may include a portion of storage resources on a device on which the electronic device is installed, or it may be accessible by the device but is not necessarily a portion of the device. For example, the non-volatile memory 516 may be accessed over a network via the network interface 520 .

[0105] In particular, the system memory 512 and the non-volatile memory 516 may include, respectively, a temporary copy and a permanent copy of the instructions 524. The instructions 524 may include instructions that, when executed by at least one of the processors 504, cause the electronic device to implement the data processing method described above. In some embodiments, the instructions 524, hardware, firmware, and / or software components thereof may be additionally / alternatively placed in the system control logic 508, the network interface 520, and / or the processor 504.

[0106] The network interface 520 may include a transceiver for providing a radio interface for the electronic device to communicate with any other suitable device (such as a front-end module, an antenna, etc.) through one or more networks. In some embodiments, the network interface 520 may be integrated with other components of the electronic device.

[0107] The network interface 520 may further include any suitable hardware and / or firmware to provide a multiple-input multiple-output radio interface. For example, the network interface 520 may be a network adapter, a wireless network adapter, a telephone modem and / or a wireless modem.

[0108] In one embodiment, at least one of the processors 504 may be packaged with logic for one or more controllers of the system control logic 508 to form a system-in-package (Session Initiation Protocol, SiP). In one embodiment, at least one of the processors 504 may be integrated with logic for one or more controllers of the system control logic 508 on the same die to form a system on chip (System On Chip, SoC).

[0109] The electronic device may further include: an input / output (I / O) device 532. The I / O device 532 may include a user interface that enables a user to interact with the electronic device; the design of the peripheral component interface enables the peripheral component to also interact with the electronic device. In some embodiments, the electronic device further includes a sensor for determining at least one of an environmental condition and location information related to the electronic device.

[0110] In some embodiments, the user interface may include, but is not limited to, a display (e.g., an LCD display, a touch screen display, etc.), a speaker, a microphone, one or more cameras (e.g., a still image camera and / or a video camera), a flashlight (e.g., an LED flash), and a keyboard.

[0111] In some embodiments, the peripheral component interface may include, but is not limited to, a non-volatile memory port, an audio jack, and a power interface.

[0112] An embodiment of the present application also provides a computer-readable storage medium, in which at least one instruction or at least one program is stored, and the at least one instruction or at least one program is loaded and executed by a processor to implement the data processing method as described above.

[0113] Optionally, in the embodiment of this specification, the storage medium may be located in at least one of the multiple network servers of the computer network. Optionally, in this embodiment, the above storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

[0114] The memory of the embodiment of this specification can be used to store software programs and modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, applications required for functions, etc.; the data storage area may store data created according to the use of the device, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory may also include a memory controller to provide the processor with access to the memory.

[0115] The embodiment of the present application also provides a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the data processing method provided by the above method embodiment.

[0116] The data processing method provided in the embodiment of the present application can be executed in a terminal, a computer terminal, a server or a similar computing device. Taking running on a server as an example, Figure 6 The hardware structure block diagram of a server provided in the embodiment of the present application is shown in FIG. Figure 6As shown, the server 300 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 610 (the central processing unit 610 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 630 for storing data, and one or more storage media 620 (such as one or more mass storage devices) for storing application programs 626 or data 622. Among them, the memory 630 and the storage medium 620 can be short-term storage or permanent storage. The program stored in the storage medium 620 may include one or more modules, each of which may include a series of instruction operations on the server. Furthermore, the central processing unit 610 can be configured to communicate with the storage medium 620 and execute a series of instruction operations in the storage medium 620 on the server 600. The server 600 may also include one or more power supplies 660, one or more wired or wireless network interfaces 650, one or more input and output interfaces 640, and / or one or more operating systems 621, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0117] The input / output interface 640 may be used to receive or send data via a network. The specific example of the network may include a wireless network provided by a communication provider of the server 600. In one example, the input / output interface 640 includes a network adapter (Network Interface Controller, NIC), which may be connected to other network devices via a base station so as to communicate with the Internet. In one example, the input / output interface 640 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0118] It can be understood by those skilled in the art that Figure 6 The structure shown is for illustration only and does not limit the structure of the above electronic device. Figure 6 More or fewer components as shown, or with Figure 6 Different configurations are shown.

[0119] It should be noted that the above-mentioned sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above-mentioned specific embodiments of this specification are described. Other embodiments are within the scope of the attached claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0120] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device and server embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0121] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or by a program to instruct the relevant hardware, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0122] The above are only preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A data processing method, characterized in that: The method comprises: Acquire driving data to be analyzed; the driving data to be analyzed is data collected when an abnormality occurs in the vehicle; the driving data to be analyzed includes environmental perception data, driving decision data, and execution action data; Performing scene recognition based on the driving decision data to obtain a target scene corresponding to the driving data to be analyzed; Obtaining analysis rules corresponding to the target scenario; Based on the analysis rules corresponding to the target scene corresponding to the target scene, an abnormality analysis is performed on the environmental perception data, the driving decision data and the execution action data to obtain an analysis result.

2. The method according to claim 1, characterized in that The obtaining of the driving data to be analyzed includes: Acquire original driving data; the original driving data is data encoded according to a preset encoding format; The original driving data is decoded to obtain the driving data to be analyzed.

3. The method according to claim 1, characterized in that The performing scene recognition based on the driving decision data to obtain a target scene corresponding to the driving data to be analyzed includes: Matching the driving decision data with a scene recognition rule in a scene recognition rule library to obtain a matching result; the scene recognition rule library is configured with at least one scene recognition rule, and each scene recognition rule corresponds to a preset scene; When the matching result indicates that the driving decision data matches the target scene recognition rule, the preset scene corresponding to the target scene recognition rule is determined as the target scene corresponding to the driving data to be analyzed; the target scene recognition rule is one of the at least one scene recognition rules.

4. The method according to claim 1, characterized in that: The driving data to be analyzed also includes a vehicle version identifier; the analysis rule corresponding to the target scene corresponding to the target scene is based on which the environmental perception data, the driving decision data, and the execution action data are analyzed for abnormalities to obtain analysis results, including: Based on the vehicle version identifier, obtaining vehicle configuration information of the vehicle; Based on the analysis rules and the vehicle configuration information, an abnormality analysis is performed on the environmental perception data, the driving decision data and the execution action data to obtain an analysis result.

5. The method according to claim 4, characterized in that The driving data to be analyzed also includes environmental perception true value data, and the environmental perception true value data is used to verify the accuracy of the environmental perception data; based on the analysis rules and the vehicle configuration information, the environmental perception data, the driving decision data and the execution action data are analyzed for abnormalities to obtain analysis results, including: Comparing the environmental perception true value data with the environmental perception data to obtain a comparison result; When the comparison result indicates that the environmental perception data is accurate, an abnormality analysis is performed on the driving decision data and the execution action data based on the vehicle configuration information and the environmental perception data to obtain an analysis result.

6. The method according to claim 5, characterized in that When the comparison result indicates that the environmental perception data is accurate, based on the vehicle configuration information and the environmental perception data, an abnormality analysis is performed on the driving decision data and the execution action data to obtain an analysis result, including: If the comparison result indicates that the environmental perception data is accurate, determining an expected motion model based on the environmental perception data; Based on the expected motion model and the vehicle configuration information, an abnormality analysis is performed on the driving decision data and the execution action data to obtain an analysis result.

7. The method according to any one of claims 1 to 6, characterized in that: The method further comprises: An analysis report is generated based on the target scenario and the analysis results.

8. The method according to claim 7, characterized in that The analysis result includes the abnormal cause of the abnormality; and the generating of the analysis report based on the target scenario and the analysis result includes: Acquire a knowledge base corresponding to the target scenario; determining improvement suggestions in the knowledge base based on the abnormal cause; Generate an analysis report based on the target scenario, the abnormal cause and the improvement suggestions.

9. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded by the processor and executes the data processing method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The storage medium stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the data processing method according to any one of claims 1 to 8.

11. A computer program product, characterized in that The method comprises a computer program, which, when executed by a processor, implements the data processing method according to any one of claims 1 to 8.

Images