Automatic data recovery method and device, electronic equipment and storage medium

Abstract

The invention provides an automatic data recovery method and device, electronic equipment and a storage medium. The automatic data recovery method comprises the following steps: acquiring vehicle driving data used for representing driving habits in a vehicle driving process in real time; if it is detected that the vehicle driving data meets a triggering condition for recording judgment on a driving style, recording the vehicle driving data; marking a current driving style corresponding to the vehicle driving data to obtain marking data; the trigger conditions are generated by triggering corresponding record judgment conditions by using different driving styles; and uploading the marking data to a cloud end, so that the cloud end recycles the marking data. Therefore, data acquisition is performed in a more targeted manner, an application scene of vehicle driving is met, the data volume is not too large, and resource waste and time cost are reduced.

Description

Technical Field

[0001] The present invention relates to the field of data acquisition technology, and in particular to a method, device, electronic device and storage medium for automatic data recovery. Background Art

[0002] Among related technologies, end-to-end autonomous driving solutions rely on deep learning technology, acquiring data from sensors such as cameras, radar, and lidar, and using complex AI (artificial intelligence) models to analyze and make decisions without human intervention. This solution can better handle complex traffic environments and optimize vehicle driving behavior in various scenarios.

[0003] However, the model training samples of related technologies require a sufficient number of samples to ensure the generalization ability of the model and avoid overfitting. However, too much data may lead to waste of resources and increase time costs. Summary of the Invention

[0004] The present application provides an improved data automatic recovery method, device, electronic device and storage medium.

[0005] This application provides a method for automatic data recovery, including:

[0006] Real-time acquisition of vehicle driving data used to characterize driving habits during vehicle driving;

[0007] If it is detected that the vehicle driving data meets a trigger condition for recording a driving style determination, recording the vehicle driving data;

[0008] Marking the current driving style corresponding to the vehicle driving data to obtain marked data; the trigger condition is generated by using different driving styles to trigger corresponding record determination conditions;

[0009] The marking data is uploaded to the cloud, so that the cloud can retrieve the marking data.

[0010] Furthermore, the vehicle driving data includes current vehicle signals and current vehicle driving mode;

[0011] The method further comprises:

[0012] Determining a driving style judgment condition corresponding to the current vehicle driving mode; wherein the parameter threshold of the corresponding driving style judgment condition is different under different vehicle driving modes;

[0013] Based on the current vehicle signal of the vehicle driving data, a current driving style corresponding to the current vehicle signal is determined according to a corresponding driving style judgment condition.

[0014] Furthermore, the vehicle driving data includes current vehicle signals and current vehicle driving mode;

[0015] If it is detected that the vehicle driving data satisfies a trigger condition for recording a driving style determination, recording the vehicle driving data includes:

[0016] If a current driving style corresponding to the current vehicle signal is determined according to the current driving style judgment condition, determining whether the current driving style meets the trigger condition, and determining whether to capture vehicle driving data within a time period corresponding to the current driving style;

[0017] Record the captured data.

[0018] Furthermore, determining whether to capture vehicle driving data within a time period corresponding to the current driving style based on whether the current driving style meets the triggering condition for recording the driving style determination includes:

[0019] If the current driving style can represent driving habits in a normal scenario, capturing vehicle driving data within a time period corresponding to the current driving style; and recording the captured data;

[0020] If the current driving style can represent the driving habits in a specific scenario, then the capture of the vehicle driving data in the time period corresponding to the current driving style is stopped and discarded.

[0021] Furthermore, if the vehicle driving data is detected to meet a trigger condition for recording a driving style determination, recording the vehicle driving data includes:

[0022] If a current driving style corresponding to the current vehicle signal is determined according to the current driving style judgment condition, then capturing vehicle driving data within a time period corresponding to the current driving style;

[0023] Record the captured data.

[0024] Furthermore, the driving style includes: an aggressive driving type; the aggressive driving type includes an aggressive driving type in a normal scenario and / or an aggressive driving type in a specific scenario;

[0025] The determining of a current driving style corresponding to the current vehicle signal includes:

[0026] determining an aggressive driving type corresponding to the current vehicle signal;

[0027] capturing vehicle driving data within a time period corresponding to the determined aggressive driving type according to the determined aggressive driving type;

[0028] Record the captured data.

[0029] Furthermore, the determining of the driving style judgment condition corresponding to the current vehicle driving mode includes:

[0030] If it is determined based on the vehicle driving data that the current vehicle driving mode is different from the previous vehicle driving mode, the driving style judgment condition corresponding to the previous vehicle driving mode is switched to the driving style judgment condition corresponding to the current vehicle driving mode.

[0031] Furthermore, the corresponding driving style judgment conditions include the correlation between the vehicle signal and different driving styles;

[0032] The determining of a current driving style corresponding to the current vehicle signal based on the vehicle driving data and according to a corresponding driving style judgment condition includes:

[0033] Based on the current vehicle signal of the vehicle driving data, a current driving style corresponding to the current vehicle signal is determined according to associations between the vehicle signal and different driving styles.

[0034] Furthermore, the association relationship includes a conditional relationship of driving habits and a conditional relationship of non-intelligent driving activation; wherein the conditional relationship of driving habits includes one or more of vehicle speed data, braking value, steering angle, double flash, and perception data;

[0035] and / or,

[0036] The association relationship includes a conditional relationship for customized environmental judgment, corresponding to which the driving style includes a driving style corresponding to a customized driving scenario; the conditional relationship for customized environmental judgment is a changeable conditional relationship for environmental judgment.

[0037] The present application provides an automatic data recovery device, comprising:

[0038] A data acquisition module is used to acquire vehicle driving data used to characterize driving habits during vehicle driving in real time;

[0039] a trigger recording module, configured to record the vehicle driving data if it is detected that the vehicle driving data satisfies a trigger condition for recording a determination of a driving style;

[0040] a marking module, configured to mark the current driving style corresponding to the vehicle driving data to obtain marking data; wherein the trigger condition is generated by using different driving styles to trigger corresponding recording determination conditions;

[0041] The uploading module is used to upload the marking data to the cloud so that the cloud can recover the marking data.

[0042] The present application provides an electronic device, comprising one or more processors, for implementing any of the methods described above.

[0043] The present application provides a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, the method described in any one of the above items is implemented.

[0044] The present application provides a computer program product, comprising a computer program / instruction, which implements any of the above methods when executed by a processor.

[0045] In some embodiments, the automatic data recovery method of the present application collects data during the vehicle driving process. The captured vehicle driving data is more suitable for human driving habits. At the same time, the data collection is more targeted, which is in line with the application scenarios of vehicle driving. The amount of data will not be too large, reducing resource waste and time costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 Schematic diagram of the structure of the automatic data recovery method according to an embodiment of the present application;

[0047] Figure 2 Schematic diagram of the process of automatic data recovery method provided by the embodiment of the present application;

[0048] Figure 3 Shown Figure 2 Another flow chart of the automatic data recovery method shown;

[0049] Figure 4 FIG2 is a schematic diagram of the structure of the automatic data recovery device provided in an embodiment of the present application;

[0050] Figure 5 Shown is a schematic structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with one or more embodiments of this specification. Rather, they are merely examples of apparatuses and methods consistent with certain aspects of one or more embodiments of this specification, as detailed in the appended claims.

[0052] It should be noted that in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the method may include more or fewer steps than those described in this specification. In addition, a single step described in this specification may be broken down into multiple steps for description in other embodiments, and multiple steps described in this specification may be combined into a single step for description in other embodiments.

[0053] Model training samples in related technologies require a sufficient number of samples to ensure the model's generalization ability and avoid overfitting. However, too much data may lead to wasted resources and increased time costs.

[0054] Since the relevant technology does not determine the collection conditions for the driving style of human drivers, the amount of data obtained is not targeted and does not meet the application scenarios of vehicle driving.

[0055] In order to solve the technical problem that the training sample is too large, resulting in waste of resources and increased time costs, the embodiment of the present application provides a method for automatic data recovery, which is collected during the vehicle driving process. The captured vehicle driving data is more suitable for human driving habits. At the same time, data collection is more targeted, in line with the application scenario of vehicle driving, and the amount of data will not be too large, reducing resource waste and time costs. Furthermore, the labeled data in this article makes the collected data more consistent with the model algorithm. Furthermore, the driving data of different driving styles are distinguished to achieve the association between people and vehicles, which is more in line with the actual application scenario of vehicle driving.

[0056] Figure 1 Shown is a structural diagram of the application of the automatic data recovery method according to an embodiment of the present application.

[0057] like Figure 1 As shown, the present application provides a method for automatic data recovery, which is applied to electronic devices. In one example, the electronic device may include a processor for processing and analyzing data. In another example, the electronic device may include a DC (Data Collection system for automated driving, vehicle data collection and recovery system). The vehicle data collection and recovery system includes the above-mentioned processor for executing the method of this article. In other examples, this electronic device may include, but is not limited to, a vehicle-mounted terminal connected to the vehicle and a mobile terminal independent of the vehicle. The vehicle-mounted terminal connected to the vehicle may be, but is not limited to, a body processor, a center console, or a car HUD head-up display. The mobile terminal independent of the vehicle may be, but is not limited to, a smartphone, a smart phone watch, a tablet computer, or a laptop computer.

[0058] Continue to see Figure 1As shown, the electronic device 11 establishes a connection with the cloud 12 for uploading the marking data of the document to the cloud 12 so that the cloud 12 can recover the marking data.

[0059] Figure 2 The figure shows a flow chart of the automatic data recovery method provided in an embodiment of the present application.

[0060] like Figure 2 As shown, the automatic data recovery method may include but is not limited to the following steps 110 to 140:

[0061] Step 110 , obtaining vehicle driving data used to characterize driving habits during vehicle driving in real time.

[0062] Vehicle driving data refers to vehicle data collected during the current driving process. For example, vehicle driving data includes, but is not limited to, vehicle driving patterns. Another example includes, but is not limited to, vehicle signals during driving.

[0063] Step 120 : If the detected vehicle driving data satisfies the triggering condition for recording the driving style determination, the vehicle driving data is recorded.

[0064] In related technologies, human drivers often perceive the paths planned by autonomous driving systems as unnatural and inconsistent with their personal driving preferences. Subsequently, autonomous driving requires control using a driver-specific model. The model algorithm calculates the driver's preferred nodes, ensuring that the autonomous driving system behaves in a manner consistent with human driving habits.

[0065] The driving style classification in this article refers to the division of driving styles into different types based on factors such as a driver's driving habits, driving skills, and driving attitude. Common driving style classifications may include, but are not limited to, one or more of the following: safe driving type, aggressive driving type, slow driving type, and experienced driving type. By assessing the driving style of human drivers, data is collected and subsequently used to train the autonomous driving model, making the autonomous driving behavior more consistent with that of human drivers. This is crucial because valid data is a prerequisite for generating intelligent algorithms.

[0066] Based on this, when a driving style is determined and meets the driving style recording conditions, recording of the driving style is triggered. This condition is called a trigger condition. Meeting the trigger condition reflects that the driving style meets the conditions of the vehicle driving application scenario or the preset conditions of the vehicle driving application scenario.

[0067] For example, when no driving mode is enabled, vehicle driving is used to represent driving habits in normal scenarios, which is more in line with general public usage. Accordingly, preset conditions for normal driving scenarios are set for these scenarios. Normal scenarios are used to represent scenarios with relatively small dynamic changes, such as roads or parking lots. Thus, vehicle signal changes are relatively minimal. Normal scenarios may include, but are not limited to, one or more of urban road driving, highway driving, rural road driving, and parking lot driving.

[0068] For another example, when the vehicle turns on the vehicle driving mode, the vehicle driving is used to characterize the driving habits in the specific scenario, which is more suitable for the application of the specific scenario, and the preset conditions for the specific scenario of the vehicle driving are set accordingly.

[0069] For example, when the vehicle driving mode is not enabled and the external environment is determined to be a special driving environment, the vehicle driving is used to characterize the driving habits in this specific scenario, which is more suitable for the application of the specific scenario, and the preset conditions for the specific driving scenario of the vehicle are set accordingly. Among them, special driving scenarios are used to characterize driving conditions that are different from regular road driving and are challenging or extreme. For example, special driving scenarios may include, but are not limited to, driving in rainy and snowy weather, driving at night, driving on mountainous or rugged roads, driving in deserts or snow, and one or more of the following autonomous driving scenarios.

[0070] Specific scenarios are used to represent situations where the external environment experiences significant changes, or where dynamic factors such as the vehicle's driving mode experience significant changes. Consequently, vehicle signals experience significant changes. A driving mode refers to how a vehicle's powertrain, suspension, steering, braking, and other aspects are adjusted to provide an optimal driving experience and performance, based on specific driving needs or driving environments. Different driving modes allow drivers to choose the most appropriate mode based on factors such as road conditions, weather, and driving style, improving driving comfort, controllability, and safety.

[0071] Among them, the vehicle driving mode may include but is not limited to one or more of economic mode (Eco mode), sports mode (Sport mode), comfort mode (Comfort mode), snow mode (Snow mode), off-road mode (Off-road mode), and personalized mode (Individual mode).

[0072] In the above-mentioned economic mode, by optimizing the throttle response, transmission system, air conditioning and other functions, the vehicle can improve fuel efficiency. It is suitable for long-distance driving or congested urban roads, saving fuel costs.

[0073] In Sport mode, the vehicle's powertrain, transmission, and suspension are optimized for sharper acceleration and faster response, making it ideal for driving pleasure. This mode is typically used on highways or when the driver wants to enjoy the controls.

[0074] In the above-mentioned comfort mode (Comfort mode), the focus is on comfort, which usually makes the suspension system softer and reduces body vibration, making it suitable for long-term driving.

[0075] In the above-mentioned Snow / Off-road mode, the vehicle will adjust traction control and power output to adapt to wet, snowy, muddy or uneven roads to ensure stable driving of the vehicle.

[0076] In individual mode, the mode is implemented according to user customization.

[0077] Step 130: Label the current driving style corresponding to the vehicle driving data to obtain labeled data. Trigger conditions are generated by using different driving styles to trigger corresponding record determination conditions. The trigger conditions herein may also be referred to as triggers when configuring a vehicle.

[0078] It should be noted that driving style in this article refers to the unique driving methods, habits, and behavioral patterns exhibited by drivers in different traffic environments. It is used to characterize a person's decision-making, driving skills, reaction speed, and interaction with other traffic participants while driving. This driving style can reflect the driver's psychology, attitude, and safety awareness through physical operations such as acceleration, braking, and steering, thereby demonstrating the driver's unique driving style.

[0079] Step 140 : Upload the marking data to the cloud so that the cloud can retrieve the marking data.

[0080] The vehicle driving data may include, but is not limited to, specific action data executed at a specific time and in a specific external environment, or specific action data executed under a specific external environment. Therefore, the labeled data includes vehicle driving data and information such as the current driving style annotated with the vehicle driving data to indicate the specific action data executed at a specific time and in a specific external environment, or under a specific external environment. This facilitates subsequent modeling and, during autonomous driving, learns what specific actions are performed under each driving style at a specific time and in a specific external environment. The specific action data may include, but is not limited to, vehicle signals.

[0081] The above-mentioned vehicle signals may include but are not limited to signals generated by the driving control device and signals collected by the sensor system.

[0082] The driving control device may include, but is not limited to, one or more of a steering wheel, brakes, accelerator, and lights. In one example, if the driving style is a safe driving type, the vehicle chooses to stop and wait when the yellow light of a traffic light with 3 seconds remaining. In another example, if the driving style is an aggressive driving type, the vehicle chooses to continue driving when the yellow light of a traffic light with 3 seconds remaining.

[0083] The sensing data of the sensing system may include, but is not limited to, one or more of the distance data between the vehicle and other vehicles collected by the distance sensor, and the distance data between the vehicle and obstacles collected by the distance sensor.

[0084] Combine Figure 1 As shown, in an optional embodiment, the vehicle driving data includes the current vehicle signal. The driving style is characterized in the specific vehicle signal, and the specific vehicle signal may include but is not limited to one or more of vehicle speed, braking, steering, double flash, perception data, etc. Accordingly, the above method may also include but is not limited to the following two steps: Step 1, determine the driving scene of the vehicle. Step 2, based on the current vehicle signal of the vehicle driving data, determine the current driving style corresponding to the current vehicle signal according to the driving style judgment condition corresponding to the driving scene. In different driving scenes, the parameter thresholds of the corresponding driving style judgment conditions are different. In this way, the correlation between different driving styles and vehicle signals can be determined. This optional embodiment can perform the above two steps after step 110. Of course, this optional embodiment can perform the above two steps before the above step 120.

[0085] The driving scenario is used to represent the external environment or the vehicle driving mode. The driving scenario may include but is not limited to the conventional scenario and / or the specific scenario.

[0086] In this regard, the above-mentioned first step can determine the driving scene of the vehicle through the external environment of the vehicle and / or the driving mode of the vehicle. Example 1: Through the image captured by the vehicle's camera, it is detected whether the external environment is a conventional environment. If it is not a conventional environment, it is determined that the driving scene of the vehicle is a specific scene. If it is a conventional environment, it is determined that the driving scene of the vehicle is a conventional scene. Example 2: Determine whether the vehicle driving mode is turned on. If not turned on, it is determined that the driving scene of the vehicle is a conventional scene. If turned on, it is determined that the driving scene of the vehicle is a specific scene. Of course, the above two examples can be executed separately or in combination.

[0087] The parameter thresholds used in this article are used to characterize the driving style judgment criteria for each vehicle signal. The parameter thresholds for general scenarios and specific scenarios differ. The parameter thresholds herein may include, but are not limited to, one or more of an acceleration threshold, a corner angle threshold, and a vehicle speed threshold.

[0088] In an embodiment of the present application, different driving style judgment conditions are generated for different driving scenarios, and then the current driving style is determined to clearly determine the different driving needs of the driver.

[0089] For example, 1. Safe driving type and corresponding algorithm requirements require collecting driving habit data of safe driving human drivers to enable machine learning to have conservative control logic. This can reduce vehicle accidents and make the machine learning algorithm's misjudgment more humane.

[0090] 2. Aggressive driving style and corresponding algorithm requirements. It is necessary to collect driving habit data of aggressive human drivers so that the algorithm can learn control logic that matches aggressive driving styles and meet driving requirements in racing and competitive scenarios to determine autonomous driving racers.

[0091] 3. Other special scenarios and corresponding algorithm requirements. These special scenarios are determined by the external environment. These external environments may include, but are not limited to, one or more of weather, road conditions, and traffic conditions. The external environment is not limited to the data requirements for special driving scenarios such as tunnels, nighttime driving, congested roads, and mountain roads.

[0092] Figure 3 Shown Figure 2 Another flowchart of the automatic data recycling method is shown.

[0093] like Figure 3 As shown, in an optional embodiment, the vehicle driving data includes the current vehicle signal and the current vehicle driving mode. Accordingly, the method may further include but is not limited to the following two steps: step 111 and step 112:

[0094] Step 111 determines the driving style judgment condition corresponding to the current vehicle driving mode. Different driving modes have different parameter thresholds for the corresponding driving style judgment condition. Different driving modes reflect different driving scenarios. Thus, by enabling the vehicle driving mode, the driving style judgment condition corresponding to the vehicle driving mode is determined. This automatic data recovery method can thus associate vehicle signals with the vehicle, and then with the vehicle driving mode.

[0095] Step 111 can determine the driving style determination condition corresponding to the current vehicle driving mode using at least one of the following optional methods: In a first optional method, the current vehicle driving mode is directly determined based on the vehicle driving data. In a second optional method, if the vehicle driving data determines that the current vehicle driving mode is different from the previous vehicle driving mode, the driving style determination condition corresponding to the previous vehicle driving mode is switched to the driving style determination condition corresponding to the current vehicle driving mode. In this way, the driving style determination condition can be switched as the vehicle driving mode is switched.

[0096] Step 112 : Based on the current vehicle signal of the vehicle driving data and according to the corresponding driving style judgment condition, determine the current driving style corresponding to the current vehicle signal.

[0097] This optional embodiment can perform the above two steps after step 110. Of course, this optional embodiment can perform the above two steps before step 120.

[0098] Continue to see Figure 3 As shown, the corresponding driving style determination condition includes the association between the vehicle signal and different driving styles. Step 112 may further include, but is not limited to, determining a current driving style corresponding to the current vehicle signal based on the current vehicle signal of the vehicle driving data and the association between the vehicle signal and different driving styles.

[0099] Among them, the above-mentioned association relationship may include but is not limited to the conditional relationship of driving habits and the conditional relationship of non-intelligent driving activation; among them, the conditional relationship of driving habits may include but is not limited to one or more of vehicle speed data, braking value, steering angle, double flash and perception data.

[0100] Alternatively, the aforementioned associations may include, but are not limited to, custom environmental judgment conditional relationships, whereby the driving style includes the driving style corresponding to the custom driving scenario; the custom environmental judgment conditional relationships are changeable. In this way, custom driving scenarios can be set according to user needs, and data can be collected and recorded.

[0101] Among them, the vehicle driving mode and the driving style judgment condition can correspond one to one or one to multiple. The parameter threshold in this article is used to characterize the driving style judgment condition of each vehicle signal, corresponding to each vehicle driving mode. The economic mode threshold is smaller than the parameter threshold of the sports mode. Since the vehicle signal in the sports mode changes greatly, the adjustment threshold of the sports mode needs to be larger, and correspondingly, the parameter threshold is larger. In this regard, the driving style judgment condition can include at least any of the following four conditions A, B, C, and D, and the order is not distinguished:

[0102] A. The instantaneous longitudinal acceleration of the vehicle is less than or equal to the maximum acceleration threshold, which is ≤ 2 meters per second squared (m / s 2 ). Or, one of rapid acceleration of 20 KPH (kilometers per hour) within 2 seconds and rapid deceleration of 10 KPH (kilometers per hour) within 2 seconds is also called aggressive emergency braking.

[0103] B. The vehicle steering wheel angle is less than or equal to a maximum angle threshold, such as ≤±4 degrees (°).

[0104] C. The vehicle speed is greater than or equal to the vehicle speed threshold, and the maximum vehicle speed threshold is greater than 5 meters per second (m / s).

[0105] D. Non-intelligent driving activation status.

[0106] Among them, A, B, and C are all conditions that can characterize driving habits. D is for determining manual driving. When it is determined that at least D conditions are not met, it means that the driving style judgment conditions are not met, and the current driving style corresponding to the current vehicle signal is determined. For example, based on A, B, C, and D in the driving style judgment conditions, if it is determined that A, B, and C are all met and only D is not met, then the current driving style corresponding to the current vehicle signal is determined to be no driving style, that is, the current driving style is empty. For another example, based on A, B, C, and D in the driving style judgment conditions, if it is determined that A, B, C, and D are all not met, then the current driving style corresponding to the current vehicle signal is determined to be no driving style, that is, the current driving style is empty.

[0107] To this end, the present invention can obtain the vehicle signals required for the aforementioned driving style judgment conditions A, B, C, and D, and perform judgments based on these conditions. D can be used as a separate driving style judgment condition to determine whether manual driving is occurring. For example, D can be prioritized before A, B, and C. In this way, vehicle driving data can be obtained during manual driving to implement the solution of this application.

[0108] If D is not satisfied, the vehicle's current driving state is not manual driving, that is, the vehicle's current driving state is intelligent driving. In this case, the article can obtain vehicle driving data and determine the corresponding driving style conditions based on the current vehicle signal. If the current driving style corresponding to the current vehicle signal is empty, and the above trigger conditions cannot be met subsequently, the above vehicle driving data does not need to be captured and will not be recorded.

[0109] If the driving style determination conditions include only one of conditions A+B+C, then if this condition is met, the current driving style is determined to be a safe driving type. Conversely, if this condition is not met, then the current driving style is determined to be an aggressive driving type.

[0110] If the driving style judgment condition includes only any two of conditions A+B+C, then these two conditions are met, and the current driving style is determined to be a safe driving type. Conversely, if at least one of these two conditions is not met, the current driving style is determined to be an aggressive driving type.

[0111] If the driving style judgment condition includes the three conditions A+B+C, then if these three conditions are met, the driving style is determined to be a safe driving type. Conversely, if at least one of these three conditions is not met, the current driving style is determined to be an aggressive driving type.

[0112] If the driving style determination criteria include the four conditions A+B+C+D, and these four conditions are met, the driving style is determined to be safe. Conversely, if D is not met, the other conditions A+B+C may or may not be met, and the current driving style is determined to be null. If at least one of the four conditions A+B+C is not met, the current driving style is determined to be aggressive. For example, if none of the conditions A+B+C+D are met, the current driving style is determined to be aggressive.

[0113] Related technologies do not distinguish between scenarios, collect end-to-end data indiscriminately, cannot distinguish driving data of different driving styles, and fail to achieve human-vehicle correlation. Compared with related technologies, the automatic data recovery method provided in the embodiments of this application can distinguish driving scenarios by different vehicle driving modes, collect end-to-end data differentially, distinguish driving data of different driving styles, achieve human-vehicle correlation, and thus realize targeted collection for different driving styles.

[0114] Combine Figure 1 As shown, the vehicle driving data includes the current vehicle signal and the current vehicle driving mode. The above step 120 can be implemented by at least one of the following optional implementation methods:

[0115] In a first optional implementation, the above step 120 may further include but is not limited to the following steps (1) and (2):

[0116] (1) If the current driving style corresponding to the current vehicle signal is determined based on the current driving style judgment condition, it is determined whether the current driving style meets the trigger condition and whether to capture the vehicle driving data within the time period corresponding to the current driving style.

[0117] The time period herein refers to the time associated with the current driving style. This time period may include, but is not limited to, the duration of the current driving style and / or a predetermined period of time before and after the current driving style. The predetermined period of time before and after the current driving style may be, for example, a period of 10 seconds before and after the current driving style.

[0118] In the above step (1), it is determined that the current driving style meets the trigger condition, and the vehicle driving data within the time period corresponding to the current driving style is captured. If it is determined that the current driving style does not meet the trigger condition, the vehicle driving data within the time period corresponding to the current driving style is not captured. In this way, human driving style data is collected under specific conditions, and the captured data is more suitable for human driving habits, providing real data for subsequent simulation model training.

[0119] (2) Record the captured data.

[0120] In an embodiment of the present application, vehicle driving data corresponding to the driving style of the required scenario can be selectively captured.

[0121] Furthermore, based on whether the current driving style meets the triggering conditions for recording and determining the driving style, determining whether to capture vehicle driving data within the time period corresponding to the current driving style may include, but is not limited to, the following steps 1) and (2): 1) If the current driving style can represent driving habits in a general scenario, capturing vehicle driving data within the time period corresponding to the current driving style; and recording the captured data. 2) If the current driving style can represent driving habits in a specific scenario, stopping capturing and discarding vehicle driving data within the time period corresponding to the current driving style.

[0122] In an embodiment of the present application, vehicle driving data corresponding to driving styles in conventional scenarios can be selectively captured, which makes it easier to obtain conventional vehicle driving data in order to train conventional big data commonly used by drivers and discard data from some specific scenarios, thereby improving the efficiency of data collection.

[0123] In a second optional implementation, step 120 may further include, but is not limited to, the following steps [1] and [2]: [1] If a current driving style corresponding to the current vehicle signal is determined based on the current driving style judgment condition, then capturing vehicle driving data within a time period corresponding to the current driving style. [2] Recording the captured data.

[0124] In the embodiment of the present application, once the current driving style is determined, the vehicle driving data is recorded according to the current driving style. Subsequently, the specific driving style classification and training are applied to the specific driving style, which is more conducive to targeted training for specific scenarios.

[0125] Furthermore, the aforementioned driving styles include: aggressive driving types; aggressive driving types include aggressive driving types in conventional scenarios and / or aggressive driving types in specific scenarios; accordingly, determining the current driving style corresponding to the current vehicle signal may include, but is not limited to, the following steps 1] to 3]: 1] Determine the aggressive driving type corresponding to the current vehicle signal. 2] Based on the determined aggressive driving type, capture vehicle driving data within the time period corresponding to the determined aggressive driving type. 3] Record the captured data. In this way, the actual operating habit data of human drivers is captured and collected, providing data support for subsequent model algorithms that realistically simulate human drivers, such as algorithms corresponding to aggressive driving types.

[0126] When the following conditions are met: safe driving type A+B+C+D; aggressive driving type other than A+B+C+D; other driving scenarios such as E, or custom driving scenarios, that is, when the vehicle signal represents the driving type corresponding to the vehicle's behavior after the driver's control, the vehicle data for that time period is captured and packaged for recycling. The captured data is labeled data, including data corresponding to safe driving type and / or data corresponding to aggressive driving type. This data is uploaded to the cloud, where it can be used to simulate and train algorithm models. This vehicle driving data can serve as a standardized data pool for different algorithm requirements (algorithm updates, related to algorithm frameworks and algorithm training results updates, such as software V1.0 and V2.0) to verify algorithm accuracy. This data capture transcends traditional event types. Subsequent data learning eliminates the need to passively accept problems as they arise, but instead allows for preemptive research by obtaining real data before an accident occurs.

[0127] Based on the same inventive concept as the above method, the embodiment of the present application also provides a data automatic recovery device, such as Figure 4 As shown, the data automatic recovery device may include the following modules:

[0128] The data acquisition module 31 is used to acquire vehicle driving data used to characterize driving habits during vehicle driving in real time;

[0129] a trigger recording module 32 for recording the vehicle driving data if the detected vehicle driving data satisfies a trigger condition for recording a determination of the driving style;

[0130] The marking module 33 is used to mark the current driving style corresponding to the vehicle driving data to obtain marking data; the trigger condition is generated by using different driving styles to trigger the corresponding record determination conditions;

[0131] The uploading module 34 is used to upload the marking data to the cloud so that the cloud can recover the marking data.

[0132] In one embodiment, the vehicle driving data includes a current vehicle signal and a current vehicle driving mode; accordingly, the apparatus further includes: a driving style judgment condition determination module, configured to determine a driving style judgment condition corresponding to the current vehicle driving mode; the parameter thresholds of the corresponding driving style judgment condition differ under different vehicle driving modes;

[0133] The current driving style determination module is used to determine the current driving style corresponding to the current vehicle signal based on the current vehicle signal of the vehicle driving data and according to the corresponding driving style judgment condition.

[0134] As an embodiment, the vehicle driving data includes the current vehicle signal and the current vehicle driving mode. The trigger recording module includes:

[0135] a first judgment unit configured to, if a current driving style corresponding to a current vehicle signal is determined based on a current driving style judgment condition, determine whether the current driving style meets a trigger condition, and determine whether to capture vehicle driving data within a time period corresponding to the current driving style;

[0136] And, a first recording unit is used to record the captured data.

[0137] As an embodiment, the judgment unit is specifically configured to: if the current driving style can represent driving habits in a normal scenario, capture vehicle driving data within a time period corresponding to the current driving style; and record the captured data;

[0138] If the current driving style can represent the driving habits in a specific scenario, then the vehicle driving data within the time period corresponding to the current driving style is stopped and discarded.

[0139] As an embodiment, the trigger recording module includes: a second judgment unit, which is used to capture vehicle driving data within a time period corresponding to the current driving style if a current driving style corresponding to the current vehicle signal is determined based on the current driving style judgment condition; and a second recording unit, which is used to record the captured data.

[0140] As an embodiment, driving styles include: aggressive driving types; aggressive driving types include aggressive driving types in conventional scenarios and / or aggressive driving types in specific scenarios; a current driving style determination module is specifically used to determine an aggressive driving type corresponding to a current vehicle signal; based on the determined aggressive driving type, vehicle driving data within a time period corresponding to the determined aggressive driving type is captured; and the captured data is recorded.

[0141] As an embodiment, the driving style judgment condition determination module is specifically used to: if it is determined based on the vehicle driving data that the current vehicle driving mode is different from the previous vehicle driving mode, then the driving style judgment condition corresponding to the previous vehicle driving mode is switched to the driving style judgment condition corresponding to the current vehicle driving mode.

[0142] As an embodiment, the corresponding driving style judgment condition includes the association between the vehicle signal and different driving styles; the current driving style determination module is specifically used to: based on the current vehicle signal of the vehicle driving data, determine the current driving style corresponding to the current vehicle signal according to the association between the vehicle signal and different driving styles.

[0143] The implementation process of the functions and effects of each module / unit in the above-mentioned device is specifically detailed in the implementation process of the corresponding steps in the above-mentioned method, which can achieve the same technical effect and will not be repeated here.

[0144] An embodiment of the present application provides an electronic device including the above-mentioned automatic data recovery device.

[0145] Figure 5 Shown is a schematic structural diagram of an electronic device 50 provided in an embodiment of the present application.

[0146] like Figure 5 As shown, the electronic device 50 includes one or more processors 51 for implementing the above automatic data recovery method.

[0147] In some embodiments, the electronic device 50 may include a storage medium 59. For example, the computer-readable storage medium may store a program that can be called by the processor 51, and may include a non-volatile storage medium. In some embodiments, the electronic device 50 may include a memory 58 and an interface 57. In some embodiments, the electronic device 50 may also include other hardware depending on the actual application.

[0148] The computer-readable storage medium of the embodiment of the present application stores a program thereon, which, when executed by the processor 51, is used to implement the automatic data recovery method described above.

[0149] The present application provides a computer program product, comprising a computer program / instruction, which implements any of the above methods when executed by a processor.

[0150] The present application also provides a computer program, which is stored in a computer-readable storage medium, for example, Figure 5 The computer program is stored in a storage medium 59, and when the processor executes the computer program, the processor 51 is prompted to execute the method described above.

[0151] The present application may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing program code. Computer-readable storage media include permanent and non-permanent, removable and non-removable media, and information storage may be implemented by any method or technology. Information may be computer-readable instructions, data structures, modules of a program, or other data. Examples of computer-readable storage media include but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0152] The above description is only a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of this specification should be included in the scope of protection of this specification.

[0153] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, the phrase "comprises a ..." defining an element does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

Claims

1. A method for automatic data recovery, characterized in that: include: Real-time acquisition of vehicle driving data used to characterize driving habits during vehicle driving; If it is detected that the vehicle driving data meets a trigger condition for recording a driving style determination, recording the vehicle driving data; Marking the current driving style corresponding to the vehicle driving data to obtain marked data; The trigger condition is generated by using different driving styles to trigger corresponding record determination conditions; The marking data is uploaded to the cloud, so that the cloud can retrieve the marking data.

2. The automatic data recovery method according to claim 1, wherein: The vehicle driving data includes current vehicle signals and current vehicle driving mode; The method further comprises: Determining a driving style judgment condition corresponding to the current vehicle driving mode; wherein the parameter threshold of the corresponding driving style judgment condition is different under different vehicle driving modes; Based on the current vehicle signal of the vehicle driving data, a current driving style corresponding to the current vehicle signal is determined according to a corresponding driving style judgment condition.

3. The automatic data recovery method according to claim 1 or 2, characterized in that: The vehicle driving data includes current vehicle signals and current vehicle driving mode; If it is detected that the vehicle driving data satisfies a trigger condition for recording a driving style determination, recording the vehicle driving data includes: If a current driving style corresponding to the current vehicle signal is determined according to the current driving style judgment condition, determining whether the current driving style meets the trigger condition, and determining whether to capture vehicle driving data within a time period corresponding to the current driving style; Record the captured data.

4. The automatic data recovery method according to claim 3, wherein: The determining whether to capture vehicle driving data within a time period corresponding to the current driving style based on whether the current driving style meets the triggering condition for recording the driving style determination includes: If the current driving style can represent driving habits in a normal scenario, capturing vehicle driving data within a time period corresponding to the current driving style; and recording the captured data; If the current driving style can represent the driving habits in a specific scenario, then the capture of the vehicle driving data in the time period corresponding to the current driving style is stopped and discarded.

5. The automatic data recovery method according to claim 1 or 2, characterized in that: If it is detected that the vehicle driving data satisfies a trigger condition for recording a driving style determination, recording the vehicle driving data includes: If a current driving style corresponding to the current vehicle signal is determined according to the current driving style judgment condition, then capturing vehicle driving data within a time period corresponding to the current driving style; Record the captured data.

6. The automatic data recovery method according to claim 2, wherein: The driving style includes: an aggressive driving type; the aggressive driving type includes an aggressive driving type in a normal scenario and / or an aggressive driving type in a specific scenario; The determining of a current driving style corresponding to the current vehicle signal includes: determining an aggressive driving type corresponding to the current vehicle signal; capturing vehicle driving data within a time period corresponding to the determined aggressive driving type according to the determined aggressive driving type; Record the captured data.

7. The automatic data recovery method according to claim 2, wherein: The determining of the driving style judgment condition corresponding to the current vehicle driving mode includes: If it is determined based on the vehicle driving data that the current vehicle driving mode is different from the previous vehicle driving mode, the driving style judgment condition corresponding to the previous vehicle driving mode is switched to the driving style judgment condition corresponding to the current vehicle driving mode.

8. The automatic data recovery method according to claim 2, wherein: The corresponding driving style judgment conditions include the correlation between vehicle signals and different driving styles; The determining of a current driving style corresponding to the current vehicle signal based on the vehicle driving data and according to a corresponding driving style judgment condition includes: Based on the current vehicle signal of the vehicle driving data, a current driving style corresponding to the current vehicle signal is determined according to associations between the vehicle signal and different driving styles.

9. The automatic data recovery method according to claim 8, wherein: The association relationship includes a conditional relationship of driving habits and a conditional relationship of non-intelligent driving activation; wherein the conditional relationship of driving habits includes one or more of vehicle speed data, braking value, steering angle, double flash, and perception data; and / or, The association relationship includes a conditional relationship for customized environmental judgment, corresponding to which the driving style includes a driving style corresponding to a customized driving scenario; the conditional relationship for customized environmental judgment is a changeable conditional relationship for environmental judgment.

10. A data automatic recovery device, characterized in that: include: A data acquisition module is used to acquire vehicle driving data used to characterize driving habits during vehicle driving in real time; a trigger recording module, configured to record the vehicle driving data if it is detected that the vehicle driving data satisfies a trigger condition for recording a determination of a driving style; a marking module, configured to mark the current driving style corresponding to the vehicle driving data to obtain marked data; The trigger condition is generated by using different driving styles to trigger corresponding record determination conditions; The uploading module is used to upload the marking data to the cloud so that the cloud can recover the marking data.

11. An electronic device, characterized in that: The method comprises one or more processors for implementing the automatic data recycling method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that A program is stored thereon, and when the program is executed by a processor, the automatic data recovery method according to any one of claims 1 to 9 is implemented.

Images