Data management method, device, computer device and storage medium

Abstract

The present disclosure relates to a data management method, device, computer device and storage medium. It relates to high-level auxiliary driving technology of automobile, and solves the problems of low data utilization and large storage resource consumption. The method comprises: generating a unique collection data code of the current trigger event in the case of detecting a preset trigger event; collecting signal data and / or image data of the trigger event; and storing the signal data and / or the image data in association with the collection data code. The technical solution provided by the present disclosure is suitable for vehicle system event monitoring, and realizes efficient management of vehicle scene data.

Description

Technical Field

[0001] This disclosure relates to advanced driver assistance technologies for automobiles, and more particularly to a data management method, apparatus, computer device, and storage medium. Background Technology

[0002] Advanced driver assistance systems (ADAS) face infinitely changing road scenarios. Before mass production of vehicles, the generalization coverage of road scenarios is insufficient, which leads to problems in ADAS during use, causing safety risks or sudden requests for safety takeover, affecting the user experience.

[0003] To address the issue of scenario generalization, a data closed-loop approach can be used to promptly trigger and record unrecognizable scenario-related data for data simulation, training, and analysis, forming a cycle of "problem discovery - optimization iteration" to update the vehicle's advanced driver assistance system and achieve coverage of new scenarios.

[0004] The recorded data contains multiple modalities, such as sensor signal data and images. The mixed storage of data from different modalities and dimensions results in a massive amount of stored data. Furthermore, it is difficult to filter and utilize this data, making it impossible to extract useful information from the vast amount of data.

[0005] In summary, the lack of a management mechanism for road-related data leads to low data utilization. Summary of the Invention

[0006] To overcome the problems existing in related technologies, this disclosure provides a data management method, apparatus, computer device, and storage medium. Data acquisition is initiated based on a trigger event, and signal data and image data are associated and stored through a unique acquisition data code. This achieves efficient management of vehicle scene data, solves the problems of low data utilization and high storage resource consumption, and provides a high-quality data foundation for fully utilizing data to obtain effective information to improve vehicle performance.

[0007] According to a first aspect of the embodiments of this disclosure, a data management method is provided, comprising: Upon detecting the occurrence of a preset trigger event, a unique data encoding for the current trigger event is generated; Collect signal data and / or image data of the triggering event; The signal data and / or the image data are encoded and associated with the acquired data and stored.

[0008] Furthermore, the step of encoding and associating the signal data and / or the image data with the acquired data for storage includes: The signal data and the image data are packaged and stored separately, and then associated through the encoding of the acquired data. The image data storage structure for storing the image data includes at least one or more of the following fields: Data collection includes encoding, data type, trigger time, sensor data image, and system timestamp. The signal data storage structure for storing the signal data includes at least one or more of the following fields: The data collected includes the encoding, data type, trigger time, vehicle information number, control status signal, target information signal, sensing data signal, and system timestamp.

[0009] Furthermore, the step of generating a unique acquisition data code for the current trigger event when a preset trigger event is detected includes: If any one of a set of preset trigger events is detected, generate the acquired data encoding containing at least one or more of the following fields: Vehicle code, trigger code, trigger time, event code; The trigger code matches the current trigger event, and the trigger code includes at least one or more of the following fields: Trigger condition encoding, vehicle information signal, control status signal, target information signal, sensing camera acquisition encoding, sensing radar acquisition encoding, sensing lidar acquisition encoding, sensing ultrasonic radar acquisition encoding, sensing inertial measurement unit acquisition encoding, and sensing map acquisition encoding.

[0010] Furthermore, the step of acquiring the signal data and / or image data of the triggering event includes: Using the acquired data encoding as an index, and according to the trigger encoding, obtain the signal data acquisition requirements and / or image data acquisition requirements. The signal data acquisition requirements indicate the range of acquired signal data content and the return storage configuration, and the image data acquisition requirements indicate the range of acquired image data content and the return storage configuration. The acquisition of signal data is performed according to the signal data acquisition requirements and / or the acquisition of image data is performed according to the image data acquisition requirements. The signal data is acquired according to a first data acquisition cycle and the image data is acquired according to a second data acquisition cycle.

[0011] Furthermore, the method also includes: Configure the signal data acquisition requirements and / or the image data acquisition requirements.

[0012] Furthermore, the method also includes: Obtain basic information about the event to be traced, wherein the basic information includes at least one or more of the following: Vehicle code, trigger time; Based on the basic information, obtain the collected data code that matches the event to be traced; Based on the collected data encoding, obtain the signal data and / or image data associated with the event to be traced, and reconstruct the occurrence scene of the event to be traced based on the signal data and / or the image data.

[0013] Furthermore, the step of obtaining signal data and / or image data associated with the event to be traced based on the collected data encoding, and reconstructing the occurrence scene of the event to be traced based on the signal data and / or the image data, includes: Based on the trigger time, obtain the first time interval and / or the second time interval; Acquire at least one set of signal data collected within the first time interval and / or at least one set of image data collected within the second time interval; Based on the signal data, the system state changes during the occurrence of the event to be traced can be reconstructed, and / or based on the image data, the environmental changes of the vehicle during the occurrence of the event to be traced can be reconstructed; The occurrence scenario of the event to be traced can be reconstructed based on the changes in the system state and / or the changes in the environment in which the vehicle is located.

[0014] According to a second aspect of the embodiments of this disclosure, a data management apparatus is provided, comprising: The data acquisition encoding generation module is used to generate a unique data acquisition encoding for the current trigger event when a preset trigger event is detected. The data acquisition module is used to acquire signal data and / or image data of the triggering event; A data storage module is used to encode and store the signal data and / or the image data in association with the acquired data.

[0015] According to a third aspect of the embodiments of this disclosure, a computer apparatus is provided, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to execute the aforementioned data management method.

[0016] According to a fourth aspect of the embodiments of the present disclosure, a non-transitory computer-readable storage medium is provided, which, when instructions in the storage medium are executed by a computer's processor, enables the computer to perform the data management method described above.

[0017] The technical solutions provided by the embodiments of this disclosure can include the following beneficial effects: upon detecting the occurrence of a preset trigger event, a unique acquisition data code for the current trigger event is generated; then, signal data and / or image data of the trigger event are acquired; and finally, the signal data and / or the image data are associated and stored with the acquisition data code. By initiating data acquisition based on the trigger event and associating and storing signal data and image data through a unique acquisition data code, efficient management of vehicle scene data is achieved, solving the problems of low data utilization and high storage resource consumption. This provides a high-quality data foundation for fully utilizing data to obtain effective information to improve vehicle performance.

[0018] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0020] Figure 1 This is a flowchart illustrating a data management method according to an exemplary embodiment.

[0021] Figure 2 This is a flowchart illustrating yet another data management method according to an exemplary embodiment.

[0022] Figure 3 This is a flowchart illustrating yet another data management method according to an exemplary embodiment.

[0023] Figure 4 This is a flowchart illustrating yet another data management method according to an exemplary embodiment.

[0024] Figure 5 This is a flowchart illustrating yet another data management method according to an exemplary embodiment.

[0025] Figure 6 This is a block diagram illustrating a data management device according to an exemplary embodiment. Detailed Implementation

[0026] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0027] To address the issue of scenario generalization, a data closed-loop approach can be used to promptly trigger and record unrecognizable scenario-related data for data simulation, training, and analysis, forming a cycle of "problem discovery - optimization iteration" to update the vehicle's advanced driver assistance system and achieve coverage of new scenarios.

[0028] The recorded data contains multiple modalities, such as sensor signal data and images. The mixed storage of data from different modalities and dimensions results in a massive amount of stored data. Furthermore, it is difficult to filter and utilize this data, making it impossible to extract useful information from the vast amount of data.

[0029] In summary, the lack of a management mechanism for road-related data leads to low data utilization.

[0030] The current data closed-loop approach focuses solely on collecting problematic data to meet the basic needs of simulation training, completely neglecting further demands such as problem tracing and responsibility clarification in the after-sales service of mass-produced vehicles. Due to the lack of a standardized coding system and data association logic, OEMs struggle to quickly match the complete data chain corresponding to abnormal events when faced with massive amounts of operational data after mass production. They are unable to accurately reconstruct the vehicle state, environmental perception information, and control logic at the time of the event, leading to difficulties in locating the root cause of abnormal problems. This not only affects the efficient handling of after-sales issues but also restricts the data closed-loop's ability to support algorithm optimization, ultimately preventing the full realization of data value.

[0031] To address the aforementioned problems, embodiments of this disclosure provide a data management method, apparatus, computer device, and storage medium. By initiating data acquisition based on a trigger event and associating and storing signal data and image data through a unique acquisition data code, efficient management of vehicle scene data is achieved. This solves the problems of low data utilization and high storage resource consumption, providing a high-quality data foundation for fully utilizing data to obtain effective information to improve vehicle performance.

[0032] An exemplary embodiment of this disclosure provides a data management method, the process of which uses this method to manage vehicle scene data is as follows: Figure 1 As shown, it includes: Step 101: When a preset trigger event is detected, generate a unique acquisition data code for the current trigger event.

[0033] In this step, if any one of the preset trigger events is detected, the collected data code is generated, which contains at least one or more of the following fields: Vehicle code, trigger code, trigger time, event code.

[0034] Table 1 shows an example of a data structure for encoding collected data.

[0035] Among them, the vehicle code is the Vehicle Identification Number (VIN), used to uniquely identify the target vehicle, and contains basic information such as the production batch and configuration parameters of the key vehicle; the trigger code is used to associate the specific scenario that triggers data collection, and information such as the trigger conditions and data collection requirements can be obtained through the trigger code; the trigger time (also known as the trigger time) is the trigger time of data collection, in the format YYYYMMDDHHMMSS (e.g., 20251020143520 represents 14:35:20 on October 20, 2025), used to achieve data synchronization in the time dimension, ensuring that signal data and image data are aligned on the time axis, and together with the timestamp information, it can maintain time synchronization after the signal data and image data are stored separately.

[0036] Event coding defines a numerical range from 0000 to 9999, with each value corresponding to a specific data combination. Different data combinations can be defined according to application requirements. The coding description of trigger coding follows the same principle and will not be repeated here.

[0037] Table 1: Data Encoding

[0038] The trigger code matches the current trigger event, and the trigger code includes at least one or more of the following fields: Trigger condition encoding, vehicle information signal, control status signal, target information signal, sensing camera acquisition encoding, sensing radar acquisition encoding, sensing lidar acquisition encoding, sensing ultrasonic radar acquisition encoding, sensing inertial measurement unit acquisition encoding, and sensing map acquisition encoding.

[0039] The meanings of each field in the trigger code are as follows: 1. Trigger condition code: Used to define the trigger event for triggering data acquisition, pointing to the specific scene conditions and trigger event (such as "high-speed cruise + forward Radar signal interruption" or "low-speed parking + obstacle approaching"). It is the core judgment basis for triggering acquisition, and each trigger condition code corresponds to a unique trigger scene / trigger event.

[0040] 2. Vehicle Information Signals: Used to associate with the basic vehicle status signals to be collected (such as vehicle speed, tire pressure, fault codes, etc.). The encoded value of the vehicle information signal corresponds to the specific combination of vehicle signal collection to ensure accurate collection of core vehicle status data after triggering.

[0041] 3. Control status signals: Used to associate the control signals of the driver assistance system to be collected (such as ACC cruise activation status, lane keeping status, takeover request trigger status, etc.) and to restore the control logic state of the system when the trigger event occurs.

[0042] 4. Target Information Signal: Used to associate the collection requirements of environmental target data, clarify the environmental target data to be collected (such as the type, distance, speed, etc. of the target ahead), and collect data on the surrounding environmental target range to be monitored in the triggered scenario to support subsequent scene reconstruction.

[0043] 5. Perception Camera Acquisition Code: Used to define the acquisition requirements for cameras, clarifying the rules for camera data acquisition. The encoding value of the perception camera acquisition code corresponds to the combination of camera types to be acquired (such as forward-looking far / medium / near + DMS camera).

[0044] 6. Perception Radar Acquisition Code: Used to define the data acquisition rules of the radar. The coded value corresponds to the radar's detection range, signal frequency and other acquisition parameters to ensure accurate acquisition of dynamic target detection data after triggering.

[0045] 7. Perception LiDAR Acquisition Code: Used to define the data acquisition rules of LiDAR. The encoding value of the perception LiDAR acquisition code corresponds to whether LiDAR data is acquired and the range of point cloud data acquired (such as road outline, obstacle position), supporting 3D environment reconstruction.

[0046] 8. Sensing Ultrasonic Radar Acquisition Code: Used to define the data acquisition rules of ultrasonic radar (USS). The code value of the sensing ultrasonic radar acquisition code corresponds to whether USS data is acquired.

[0047] 9. Sensing Inertial Measurement Unit Acquisition Encoding: Define the IMU (Inertial Measurement Unit) data acquisition rules. The encoded values ​​correspond to the IMU data types to be acquired (such as vehicle attitude, acceleration, angular velocity) and are used to reconstruct the vehicle's motion state.

[0048] 10. Perception Map Acquisition Encoding: Define the IMU (Inertial Measurement Unit) data acquisition rules. The encoded values ​​correspond to the IMU data types to be acquired (such as vehicle attitude, acceleration, and angular velocity) and are used to reconstruct the vehicle's motion state.

[0049] Table 2 illustrates an exemplary data structure for trigger encoding. Here, "Perception_Camera" represents the encoding field for camera acquisition, "Perception_Radar" represents the encoding field for radar acquisition, "Perception_Lidar" represents the encoding field for lidar acquisition, "Perception_USS" represents the encoding field for ultrasonic radar acquisition, "Perception_IMU" represents the encoding field for inertial measurement unit acquisition, and "Perception_Map" represents the encoding field for map acquisition. Each field has a numerical encoding description, and each value corresponds to a specific data combination. The data combination methods represented by different values ​​can be defined according to the application scenario requirements.

[0050] Table 2: Trigger Encoding

[0051] In this step, the data acquisition process is triggered by detecting a trigger event. First, a data acquisition code is generated to mark the trigger event. Subsequently, the acquired data can be managed using this data acquisition code. The trigger event for data acquisition can be defined through the trigger code, pointing to specific scenario conditions and trigger events (such as "high-speed cruising + forward Radar signal interruption" or "low-speed parking + obstacle approaching").

[0052] Depending on the actual application requirements, various trigger events and corresponding data collection needs can be configured.

[0053] Step 102: Collect signal data and / or image data of the triggering event.

[0054] In this step, data acquisition is initiated upon detecting the trigger event, and signal data and / or image data are acquired according to the acquisition requirements. The specific process is as follows: Figure 2 As shown, it includes: Step 201: Using the acquired data encoding as an index, obtain the signal data acquisition requirements and / or image data acquisition requirements according to the trigger encoding.

[0055] The signal data acquisition requirement indicates the range of signal data content to be acquired and the backhaul storage configuration, while the image data acquisition requirement indicates the range of image data content to be acquired and the backhaul storage configuration.

[0056] In this step, each trigger code value corresponds to a combination of data to be collected. The above-arranged codes define a unique trigger code used to describe a specific trigger event. The trigger code can be invoked at any time based on the trigger event to obtain data acquisition requirements and / or image data acquisition requirements.

[0057] According to one exemplary implementation, the signal data acquisition requirements and / or the image data acquisition requirements can be configured according to actual application needs in this embodiment of the disclosure.

[0058] Step 202: Perform signal data acquisition according to the signal data acquisition requirements and / or perform image data acquisition according to the image data acquisition requirements.

[0059] In this step, data collection is performed according to the data collection requirements.

[0060] According to one exemplary embodiment, the signal data is acquired according to a first data acquisition cycle, and the image data is acquired according to a second data acquisition cycle. For example, signal data is acquired at a cycle of 10ms, and image data is acquired at a cycle of 30ms.

[0061] Step 103: Encode and store the signal data and / or the image data in association with the acquired data.

[0062] In this step, the signal data and the image data are packaged and stored separately, and associated with each other through the data acquisition encoding. The signal data and image data are stored separately, and associated and retrieved through data acquisition encoding, timestamps, and other information.

[0063] The image data storage structure for storing the image data includes at least one or more of the following fields: Collect data encoding, data type, trigger time, perceived data image, and system timestamp.

[0064] Image data is stored in a standardized manner using an image data storage structure.

[0065] Table 3 illustrates an example image data storage structure. Here, "Trigger Time" represents the trigger time field, and "MCU Time Stamp" represents the system timestamp field.

[0066] Table 3. Image Data Storage Structure

[0067] The functions of each field in the image data storage structure are as follows: 1. Acquisition Data Encoding: A unique identifier for this image data, associated with the signal data acquired under the same triggering event, ensuring that image data and signal data can be accurately retrieved based on the same acquisition data encoding.

[0068] 2. Data type: Clearly defined as "image data" to facilitate data classification, storage, filtering, and batch retrieval, avoiding storage redundancy and inefficient retrieval caused by mixing different types of data.

[0069] 3. Trigger Time: Records the trigger time of this image data acquisition, which is consistent with the trigger time of the corresponding signal data. It serves as the benchmark for time synchronization between image data and signal data for the same trigger event.

[0070] 4. Sensing Data Images: Stores the raw image data captured by each camera. According to an exemplary embodiment, the storage structure of the sensing data images is shown in Table 4, where the data captured by each camera are arranged and fused in the order from left to right in the table.

[0071] 5. System Timestamp: A precise timestamp of the system time. For example, the precise timestamp of a microcontroller (MCU). Used to supplement the precision of the time recorded in the trigger time field, enabling synchronization of image data and signal data in a more precise time dimension (e.g., millisecond level), avoiding data timing deviations.

[0072] Table 4. Storage structure of sensory data images

[0073] According to an exemplary implementation, the content of the data collected by each camera in the perceived data image is as follows: 1. Forward-facing camera (far): Stores image data of the road ahead at a distance, which can be used to identify lane lines, traffic lights, and distant obstacles.

[0074] 2. Forward-facing camera (medium): Collects image data of the road ahead at a medium distance to identify targets at medium distance and supplement details at a long distance.

[0075] 3. Forward-facing camera - Close: Stores image data of the road ahead at close range, used to identify obstacles (such as pedestrians and non-motorized vehicles) and lane line details at close range.

[0076] 4. Left front camera: Stores left front side view image data, used to identify vehicles in the left lane, the curb, and the side environment when changing lanes.

[0077] 5. Left rear camera: Stores left rear side view image data, used to identify vehicles approaching from the left rear and the left-side environment when parking.

[0078] 6. Right front camera: Stores right front side view image data, its function is symmetrical to that of the left front camera.

[0079] 7. Right rear camera: Stores right rear side view image data, its function is symmetrical to that of the left rear camera.

[0080] 8. Rearview camera: Stores image data of the area directly behind the vehicle for identifying the environment behind it.

[0081] 9. DMS Camera: A camera in the driver monitoring system that stores image data of the driver's status to monitor the driver's condition and detect dangerous signs such as fatigue and distraction.

[0082] The signal data storage structure for storing the signal data includes at least one or more of the following fields: The data collected includes the encoding, data type, trigger time, vehicle information number, control status signal, target information signal, sensing data signal, and system timestamp.

[0083] The functions of each field are explained below: 1. Acquisition Data Encoding: A unique identifier for this signal data, associated with image data acquired under the same triggering event, ensuring that image data and signal data can be accurately retrieved based on the same acquisition data encoding.

[0084] 2. Data type: Clearly defined as "signal data" to facilitate data classification, storage, filtering, and batch retrieval, avoiding storage redundancy and inefficient retrieval caused by mixing different types of data.

[0085] 3. Trigger Time: Records the trigger time of this signal data acquisition, which is consistent with the trigger time of the corresponding image data. It serves as the benchmark for time synchronization between image data and signal data for the same trigger event.

[0086] 4. Vehicle Information Number: Stores basic vehicle operating status data (such as vehicle speed, tire pressure, fault codes, fuel level, etc.). The encoded value of the vehicle information number corresponds one-to-one with the vehicle information signal code of the collected data. Each code corresponds to a fixed set of signal acquisition combinations to ensure that the collected data accurately matches the requirements defined by the trigger.

[0087] 5. Control status signal: Stores the control logic state of the driver assistance system. The encoded value of the control status signal is associated with the control status signal encoding of the trigger, which is used to restore the control behavior and decision logic of the system when the trigger event occurs.

[0088] 6. Target Information Signals: Stores key information about targets in the vehicle's surrounding environment (such as target type: pedestrian / motor vehicle / non-motor vehicle, distance, speed, azimuth, etc.), providing environmental data support for scene reconstruction. It serves as an important sample for environmental modeling during algorithm training and is also a crucial basis for determining the cause of events in after-sales traceability.

[0089] 7. Sensing data signals: Stores the signal-based sensing data of each sensor (including CAN / Ethernet signal data of Radar, Lidar, USS, IMU, and map).

[0090] 8. System Timestamp: A precise timestamp of the system time. For example, the precise timestamp of a microcontroller (MCU). Used to supplement the precision of the time recorded in the trigger time field, enabling synchronization of image data and signal data in a more precise time dimension (e.g., millisecond level), avoiding data timing deviations.

[0091] Table 5 illustrates an exemplary signal data storage structure.

[0092] Among them, "Trigger Time" represents the trigger time field, and "MCU Time Stamp" represents the system timestamp.

[0093] Table 5. Signal Data Storage Structure

[0094] Sensing data signals include fields such as maps, radar, lidar, ultrasonic radar (USS), and IMU. Table 6 illustrates an exemplary storage structure for sensing data signals.

[0095] Table 6. Storage Structure of Sensed Data Signals

[0096] in,: 1. Map: Stores the signal data of the map, which is associated with the sensing map acquisition code. The signal data of the map is collected or not collected according to the content indicated by the sensing map acquisition code.

[0097] 2. Radar: Stores the radar's sensing signal data and is associated with the sensing radar's acquisition encoding. The encoded value of the sensing radar acquisition encoding defines the radar's acquisition parameters (such as detection range, signal frequency, etc.), ensuring that the acquired data matches the triggering event (such as enhancing long-range radar data in high-speed scenes).

[0098] 3. Lidar: The Lidar's sensing signal data, associated with the sensing LiDAR acquisition code. The encoding value of the sensing LiDAR acquisition code determines whether to acquire data and the density of the acquired point cloud.

[0099] 4. USS: Stores the sensing signal data of the USS, including information such as the distance, position, and outline of nearby obstacles. It is associated with the sensing ultrasonic radar acquisition code, and data is collected or not collected according to the instructions of the sensing ultrasonic radar acquisition code, avoiding redundant data and achieving data lightweighting.

[0100] 5. IMU: Stores the sensing signal data from the IMU, including vehicle acceleration, angular velocity, etc., to reconstruct the vehicle's motion state and driving stability. It is associated with the encoding acquired by the sensing inertial measurement unit. The encoded values ​​of the sensing inertial measurement unit's acquisition encoding define the combination of IMU data types to be acquired, providing a benchmark for vehicle motion state for scene reconstruction.

[0101] By controlling the acquisition process, data is selectively collected based on specific triggering events and scenarios, employing targeted acquisition strategies to achieve comprehensive yet lightweight data acquisition. A unique data acquisition code associates the signal data and image data of the triggering event, facilitating subsequent management and retrieval. The acquired image and signal data are stored in a database, with packaged and returned image and signal data stored separately. This allows for subsequent data loop simulations and training, as well as problem tracing and accountability clarification, all possible using the database data.

[0102] An exemplary embodiment of this disclosure provides a data management method that, when event tracing is required and after-sales issues are discovered, calls data to restore the event to be traced. The specific process is as follows: Figure 3 As shown, it includes: Step 301: Obtain basic information about the event to be traced.

[0103] The basic information includes at least one or more of the following: Vehicle code, trigger time.

[0104] Step 302: Based on the basic information, obtain the collected data code that matches the event to be traced.

[0105] In this step, based on the basic information, the data code pointing to the event to be traced is obtained by matching the stored data codes.

[0106] Step 303: Based on the collected data encoding, obtain the signal data and / or image data associated with the event to be traced, and reconstruct the occurrence scene of the event to be traced based on the signal data and / or the image data.

[0107] This step is as follows: Figure 4 As shown, it includes: Step 401: Obtain the first time interval and / or the second time interval based on the trigger time.

[0108] In this step, the first time interval for calling signal data and the second time interval for calling image data are obtained.

[0109] Step 402: Acquire at least one set of signal data collected within the first time interval and / or at least one set of image data collected within the second time interval.

[0110] For example, the first time interval is the 1-second time interval that covers the trigger time. With the signal data collected at a 10ms cycle, a total of 100 sets of signal data are acquired.

[0111] The second time interval is a 90ms time interval that covers the trigger time. With image data collected at a 30ms cycle, a total of 3 sets of image data were acquired.

[0112] Step 403: Reconstruct the system state changes during the occurrence of the traceable event based on the signal data, and / or reconstruct the environmental changes of the vehicle during the occurrence of the traceable event based on the image data.

[0113] In this step, the system state changes of the vehicle during the occurrence of the event to be traced are reconstructed based on the signal data, such as the sudden drop in the forward radar signal strength triggering a takeover warning.

[0114] Image data can provide more detailed environmental information and also verify changes in system status.

[0115] Step 404: Reconstruct the occurrence scenario of the event to be traced based on the changes in the system state and / or the changes in the environment in which the vehicle is located.

[0116] This step reconstructs the incident scenario, providing environmental and vehicle information at the time the event occurred. Based on this scenario, the root cause and responsibility for the problem can be analyzed, enabling the tracing of quality issues and clarification of responsibility, thus providing strong data support for improving vehicle performance.

[0117] An exemplary embodiment of this disclosure also provides a data management method that can simultaneously address data loop triggering, recording, and traceability and correlation of after-sales issues in mass production of vehicles.

[0118] In this embodiment, the data management method includes a data storage structure and a data encoding method. The data storage structure addresses the issues of data storage content and arrangement, and designs a method for storing image data and signal data separately. Data is synchronized and combined / retrieved in the cloud or when the data is used via data IDs and timestamps. The data encoding method includes a data ID encoding method and encoding methods for each data type, facilitating data identification, retrieval, filtering, and extraction. The data storage structure and data encoding method are used in combination, collecting only relevant data according to specific functional scenarios. This includes relevant data used in data closure (simulation / training) and data required for responsibility clarification for data after-sales issues, achieving data lightweighting and precision. This allows the data closure-loop data recording system to simultaneously address the development of advanced driver assistance systems and the clarification of responsibility for mass production after-sales issues.

[0119] Data encoding methods include: acquired data encoding, trigger encoding, signal data storage, sensing data signal storage, image data storage, and sensing image data storage.

[0120] like Figure 5 As shown, data acquisition requirements include image data acquisition and signal data acquisition. Through data encoding and association, the associated data can be synchronously accessed, combined, and filtered, facilitating subsequent data closure and after-sales problem analysis. Data feedback is triggered, stored, and returned based on acquisition requirements, including packaged image data and packaged signal data. After data feedback, it is stored in the first database, awaiting data retrieval. Data closure utilizes the feedback data, and problem tracing and responsibility clarification also utilize the feedback data.

[0121] The second database contains trigger codes, vehicle information signals, control status signals, target information signals, sensing data signals, and sensing image signals. The trigger codes are used for data encoding, while the other information is used to configure data acquisition requirements.

[0122] The collected data encoding includes vehicle VIN encoding, trigger encoding, and event encoding.

[0123] Trigger encoding includes trigger condition encoding, vehicle information signal encoding, control status signal encoding, target information signal encoding, perception-camera acquisition encoding, perception-Radar acquisition encoding, perception-Lidar acquisition encoding, perception-USS acquisition encoding, perception-IMU acquisition encoding, and perception-map acquisition encoding.

[0124] The system includes trigger condition encoding, vehicle information signal encoding, control status signal encoding, target information signal encoding, perception-camera acquisition encoding, perception-Radar acquisition encoding, perception-Lidar acquisition encoding, perception-USS acquisition encoding, perception-IMU acquisition encoding, and perception-map acquisition encoding. Each encoding corresponds to a specific combination of data to be collected. This arrangement of encodings defines a unique trigger encoding. Trigger encodings can be invoked at any time.

[0125] Due to the special nature of image data, the signal data and image data acquired in the data closed loop are stored separately. The signal data and image data are associated through the acquired data encoding. The data time synchronization combination is achieved by storing the trigger time and MCU TimeStamp in the data, which is used for simulation and training of the data closed loop, and for tracing and clarifying responsibility for after-sales issues.

[0126] The signal data is stored by recording and storing the acquired data encoding, data type (signal / image), trigger time, vehicle information signal, control status signal, target information signal, sensing data signal and MCU TimeStamp in chronological order. For example, if the minimum signal period is 10ms, then a complete set of data is recorded and stored every 10ms, and so on.

[0127] Sensing signal data refers to sensor-sensed data stored in CAN / Ethernet signal format, including Sensing_Radar, Sensing_Lidar, Sensing_USS, Sensing_IMU, and Sensing_Map.

[0128] The image data is stored by recording and storing the acquired data encoding, data type (signal / image), trigger time, perceived data image, and MCU TimeStamp in chronological order. For example, if the minimum signal period is 30ms, then a complete set of data is recorded and stored every 30ms, and so on.

[0129] Image data storage can be arranged and stored in the following order: front-view camera (far), front-view camera (medium), front-view camera (near), left front camera, left rear camera, right front camera, right rear camera, rear-view camera, and DMS camera.

[0130] All data is linked through data collection codes. By analyzing and extracting data from closed-loop systems based on the meaning and format of the codes, after-sales issues and responsibilities can be accurately identified, thus achieving fully digital management.

[0131] An exemplary embodiment of this disclosure also provides a data management device, the structure of which is as follows: Figure 6 As shown, it includes: The data acquisition encoding generation module 601 is used to generate a unique data acquisition encoding for the current trigger event when a preset trigger event is detected. Data acquisition module 602 is used to acquire signal data and / or image data of the triggering event; The data storage module 603 is used to encode and store the signal data and / or the image data in association with the acquired data.

[0132] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0133] An exemplary embodiment of this disclosure also provides a computer apparatus, including: processor; Memory used to store processor-executable instructions; The processor is configured to execute the data management method provided in the embodiments of this disclosure.

[0134] An exemplary embodiment of this disclosure also provides a non-transitory computer-readable storage medium, wherein instructions in the storage medium, when executed by a computer's processor, enable the computer to perform the data management method provided by embodiments of this disclosure.

[0135] Embodiments of this disclosure provide a data management method, apparatus, computer device, and storage medium. Upon detecting a preset trigger event, a unique acquisition data code for the current trigger event is generated. Then, signal data and / or image data of the trigger event are acquired, and the signal data and / or image data are associated with and stored using the acquisition data code. By initiating data acquisition based on a trigger event and associating signal data and image data with a unique acquisition data code, efficient management of vehicle scene data is achieved. This solves the problems of low data utilization and high storage resource consumption, providing a high-quality data foundation for fully utilizing data to obtain effective information to improve vehicle performance.

[0136] During the data collection process, data sources and data storage formats are selected in a targeted manner according to the collection requirements, so as to achieve comprehensive and lightweight data collection, improve data collection and management efficiency, reduce system pressure and resource consumption, and further improve system performance and reliability.

[0137] During the invocation process, the accurate retrieval of related data is achieved through data encoding, providing strong data support for more complex data utilization needs such as quality problem tracing and responsibility clarification, and laying the foundation for accurate analysis and iterative upgrades of vehicle systems.

[0138] All data is linked through data coding. By analyzing and extracting data from closed-loop systems, the meaning and format of the coding can be used to accurately pinpoint after-sales issues and clarify responsibilities, thus achieving fully digital management.

[0139] Those skilled in the art will also understand that the various illustrative logical blocks and steps listed in the embodiments of this disclosure can be implemented by electronic hardware, computer software, or a combination of both. Whether such functionality is implemented in hardware or software depends on the specific application and the overall system design requirements. Those skilled in the art can implement the described functionality using various methods for each specific application, but such implementation should not be construed as exceeding the scope of protection of the embodiments of this disclosure.

[0140] Furthermore, the term “exemplary” is used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as “exemplary” is not necessarily to be construed as advantageous compared to other aspects or designs. Rather, the use of the term “exemplary” is intended to present the concept in a concrete manner. As used herein, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or.” That is, unless otherwise specified or clear from the context, “X applies A or B” is intended to mean any of the natural inclusive arrangements. That is, “X applies A or B” satisfies any of the foregoing instances if X applies A; X applies B; or both X applies A and B. Additionally, unless otherwise specified or clear from the context to refer to the singular form, the articles “a” and “an” as used in this application and the appended claims are generally understood to mean “one or more.”

[0141] Similarly, although this disclosure has been shown and described with respect to one or more implementations, equivalent variations and modifications will occur to those skilled in the art upon reading and understanding this specification and the accompanying drawings. This disclosure includes all such modifications and variations and is limited only by the scope of the claims. In particular, with respect to the various functions performed by the components described above (e.g., elements, resources, etc.), unless otherwise indicated, the terminology used to describe such components is intended to correspond to any component (functionally equivalent) that performs the specific function of the described component, even if structurally not equivalent to the disclosed structure. Furthermore, although specific features of this disclosure may have been disclosed with respect to only one of several implementations, such features may be combined with one or more other features of other implementations, as may be desired and advantageous to any given or particular application. Moreover, with regard to the terms “comprising,” “owning,” “having,” “having,” or variations thereof as used in the detailed description or claims, such terms are intended to be inclusive in a manner similar to the term “including.”

[0142] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0143] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A data management method, characterized in that, include: Upon detecting the occurrence of a preset trigger event, a unique data encoding for the current trigger event is generated; Collect signal data and / or image data of the triggering event; The signal data and / or the image data are encoded and associated with the acquired data and stored.

2. The data management method according to claim 1, characterized in that, The step of encoding and associating the signal data and / or the image data with the acquired data for storage includes: The signal data and the image data are packaged and stored separately, and then associated through the encoding of the acquired data. The image data storage structure for storing the image data includes at least one or more of the following fields: Data collection includes encoding, data type, trigger time, sensor data image, and system timestamp. The signal data storage structure for storing the signal data includes at least one or more of the following fields: The data collected includes the encoding, data type, trigger time, vehicle information number, control status signal, target information signal, sensing data signal, and system timestamp.

3. The data management method according to claim 1, characterized in that, When a preset trigger event is detected, the step of generating a unique acquisition data code for the current trigger event includes: If any one of a set of preset trigger events is detected, generate the acquired data encoding containing at least one or more of the following fields: Vehicle code, trigger code, trigger time, event code; The trigger code matches the current trigger event, and the trigger code includes at least one or more of the following fields: Trigger condition encoding, vehicle information signal, control status signal, target information signal, sensing camera acquisition encoding, sensing radar acquisition encoding, sensing lidar acquisition encoding, sensing ultrasonic radar acquisition encoding, sensing inertial measurement unit acquisition encoding, and sensing map acquisition encoding.

4. The data management method according to claim 3, characterized in that, The step of acquiring the signal data and / or image data of the triggering event includes: Using the acquired data encoding as an index, and according to the trigger encoding, obtain the signal data acquisition requirements and / or image data acquisition requirements. The signal data acquisition requirements indicate the range of acquired signal data content and the return storage configuration, and the image data acquisition requirements indicate the range of acquired image data content and the return storage configuration. The acquisition of signal data is performed according to the signal data acquisition requirements and / or the acquisition of image data is performed according to the image data acquisition requirements. The signal data is acquired according to a first data acquisition cycle and the image data is acquired according to a second data acquisition cycle.

5. The data management method according to claim 4, characterized in that, The method further includes: Configure the signal data acquisition requirements and / or the image data acquisition requirements.

6. The data management method according to claim 1, characterized in that, The method further includes: Obtain basic information about the event to be traced, wherein the basic information includes at least one or more of the following: Vehicle code, trigger time; Based on the basic information, obtain the collected data code that matches the event to be traced; Based on the collected data encoding, obtain the signal data and / or image data associated with the event to be traced, and reconstruct the occurrence scene of the event to be traced based on the signal data and / or the image data.

7. The data management method according to claim 6, characterized in that, The step of obtaining signal data and / or image data associated with the event to be traced based on the collected data encoding, and reconstructing the occurrence scene of the event to be traced based on the signal data and / or the image data, includes: Based on the trigger time, obtain the first time interval and / or the second time interval; Acquire at least one set of signal data collected within the first time interval and / or at least one set of image data collected within the second time interval; Based on the signal data, the system state changes during the occurrence of the event to be traced can be reconstructed, and / or based on the image data, the environmental changes of the vehicle during the occurrence of the event to be traced can be reconstructed; The occurrence scenario of the event to be traced can be reconstructed based on the changes in the system state and / or the changes in the environment in which the vehicle is located.

8. A data management device, characterized in that, include: The data acquisition encoding generation module is used to generate a unique data acquisition encoding for the current trigger event when a preset trigger event is detected. The data acquisition module is used to acquire signal data and / or image data of the triggering event; A data storage module is used to encode and store the signal data and / or the image data in association with the acquired data.

9. A computer device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to perform the data management method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of a computer, the computer is able to perform the data management method as described in any one of claims 1 to 7.

Images