Data recording system and method for autonomous vehicle

Abstract

The invention provides a data recording system and method of an automatic driving vehicle, the data recording system comprises at least one automatic driving control unit, and the automatic driving control unit comprises a data storage module and an intelligent driving domain control processing module; wherein the data storage module is used for storing data information of an automatic driving vehicle; the intelligent driving domain control processing module comprises a data processing module and a first data management module; the data processing module is used for receiving and processing vehicle record data of a plurality of information modules of the automatic driving vehicle; the first data management module is used for monitoring the vehicle record data of the plurality of information modules; when the data processing module receives a collision event or collision risk event trigger signal, the vehicle record data of a plurality of information modules in a standard storage period before the moment is stored in the data storage module.

Description

Technical Field

[0001] The present application relates to the technical field of automobiles, and in particular, to a data recording system and method for an autonomous vehicle. Background Art

[0002] With the rapid development of autonomous driving technology, its safety has become the focus of public attention. In this context, the role of the Data Storage System for Automated Driving (DSSAD) has become increasingly prominent. It can help restore the accident scene by recording detailed data, thereby clarifying the responsibility division between the autonomous driving system and the driver. However, most of the current autonomous driving data storage systems on the market adopt an independent design. Such an independently set data recording system not only includes core hardware such as a memory and a controller, but also auxiliary components such as a housing and a connector, resulting in a relatively high production cost. In addition, the independent system usually uses a metal housing to meet the heat dissipation requirements, which makes its overall weight relatively large, and thus has a negative impact on the cruising range of electric vehicles. Moreover, the independent system uses an independent power supply. Once this power supply fails or is damaged, the data recording function of the system may be interrupted. At a critical moment, if an accident occurs, the integrity and accuracy of the data cannot be guaranteed, thus affecting the accurate division of accident liability.

[0003] The existing problem is that the currently independently set data recording system not only has a relatively high production cost, but also uses an independent housing, occupying a relatively large space, which is not conducive to the overall vehicle layout. Summary of the Invention

[0004] The present application solves the technical problem that the currently independently set data recording system not only has a relatively high production cost, but also uses an independent housing, occupying a relatively large space, which is not conducive to the overall vehicle layout. The present application reduces the setting of a large number of electronic components and structural parts by using the domain control processing module of the autonomous vehicle, saving the production cost and the installation space.

[0005] To solve the above problems, the present application provides a data recording system for an autonomous vehicle. The data recording system includes at least one autonomous driving control unit. The autonomous driving control unit includes a data storage module and an intelligent driving domain control processing module. Among them, the data storage module is used to store the data information of the autonomous vehicle. The intelligent driving domain control processing module includes a data processing module and a first data management module. The data processing module is used to receive and process the vehicle record data of several information modules of the autonomous vehicle. The first data management module is used to monitor the vehicle record data of several information modules. When the data processing module receives a trigger signal for a collision event or a collision risk event, it stores the vehicle record data of several information modules within a standard save period before this moment into the data storage module.

[0006] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: when receiving the triggering signal of a collision event or a collision risk event, the vehicle record data of several information modules before the event occurs is automatically stored. This mechanism ensures that important data information can be obtained at a critical moment, providing a reliable basis for subsequent accident analysis and liability determination. Moreover, by setting a standard storage period, the critical data before the collision event can be completely stored, effectively avoiding data loss caused by system failures or human operation errors, and improving the integrity and reliability of the data. At the same time, the setting of the intelligent driving domain control processing module separates the data processing and management functions, enhancing the flexibility and scalability of the system. Different information modules can work independently, facilitating subsequent function expansion and system upgrade. The introduction of the first data management module makes the monitoring of the vehicle record data of several information modules more efficient. Through centralized management, potential risks can be quickly identified and responded to, enhancing the safety of the autonomous driving system.

[0007] In a possible design, the data storage module includes an Embedded MultiMediaCard (EMMC) storage module and a second data management module; among them, the second data management module is used to receive and manage the data transmitted from the intelligent driving domain control processing module, and store the data into the EMMC storage module; the data backup method of the second data management module includes data reading at the Personal Computer (PC) end and data uploading to the cloud.

[0008] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: The Embedded Multimedia Card (EMMC) storage module has a high read and write speed and low latency, enabling fast storage and retrieval of a large amount of data, improving the efficiency of data storage, and meeting the requirements of the autonomous driving system for real-time performance and high efficiency. Moreover, the second data management module is responsible for receiving and managing data from the intelligent driving domain control processing module, capable of flexibly processing different types of data streams, ensuring the orderly storage and management of data, and enhancing the overall performance of the system. At the same time, through the management of the second data management module, the system can effectively monitor the data reception and storage processes, reduce the risk of data loss or damage, and enhance the integrity and security of data. Concentrating the data storage and management functions in the EMMC storage module and the second data management module simplifies the system architecture, reduces the complexity of the system, and facilitates subsequent maintenance and upgrade. Additionally, by simultaneously adopting the methods of PC-side data reading and cloud data uploading, the system realizes multiple backups of data, reduces the risk of data loss, and ensures the security and reliability of critical data. Moreover, PC-side data reading provides a convenient local access method, allowing users to quickly obtain and analyze data, which is suitable for scenarios requiring quick response; while cloud data uploading supports remote access, facilitating data sharing and collaboration.

[0009] In a possible design, the EMMC storage module includes an autonomous driving storage area and an accident storage area; among them, the data transmitted by the intelligent driving domain control processing module includes non-locked event data normally recorded and locked event data recorded when a collision event or a collision risk event triggering signal is received; the non-locked event data is stored in the autonomous driving storage area, and the locked event data is stored in the accident storage area.

[0010] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By storing the non-locked event data and the locked event data in different areas respectively, the system realizes the classified management of data, facilitating subsequent data retrieval and analysis, and improving the efficiency of data management. Moreover, through reasonable division of the storage area, the system can utilize the storage resources more effectively, avoid waste of storage space caused by data mixing, and improve the storage efficiency. Among them, the locked event data is stored in a dedicated accident storage area, ensuring that these critical data are not overwritten or accidentally deleted, enhancing the security and integrity of data, and providing a reliable basis for accident analysis. Additionally, separating different types of data for storage can reduce interference during data retrieval, improve the efficiency of data processing, especially when quickly accessing accident-related data, enabling faster location and extraction of information. When a collision event or a collision risk event triggering signal is received, the system can promptly lock the relevant data and store it in the accident storage area, ensuring that the critical data is recorded in a timely manner when the event occurs and reducing the risk of data loss.

[0011] In a possible design, the second data management module can monitor the storage space status of the Embedded Multimedia Card (EMMC) storage module; if the remaining storage space in the accident storage area is insufficient, the second data management module sends a space shortage signal to the autonomous vehicle.

[0012] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: through real-time monitoring of the storage space status, the system can timely detect the problem of insufficient storage space, ensuring continuous recording of accident-related data and improving the reliability of the system. Moreover, when the space in the accident storage area is insufficient, a warning signal is sent in a timely manner, which can prompt the driver or the system to take measures to effectively prevent the loss of important data caused by insufficient storage space. By directly feeding back the space shortage signal to the vehicle, the driver can make corresponding decisions according to the actual situation, such as choosing to clean unnecessary data or perform data backup to ensure the normal operation of the system. Additionally, the method of storage space warning enables the driver to timely understand the storage status of the vehicle, enhancing the initiative of the user and the sense of control over the vehicle status, and improving the driving experience. At the same time, through the active monitoring and warning mechanism, the system can take preventive measures before potential problems occur, improving the overall reliability and safety of the autonomous driving system.

[0013] In a possible design, the second data management module stores the locked event data with frame dropping.

[0014] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: through frame dropping storage, the system can effectively reduce the storage requirements for the locked event data, saving space in the accident storage area, so that important event data can still be retained even when the storage space is limited. Moreover, frame dropping storage can reduce the storage of redundant data on the premise of ensuring the key features and important information of the event, ensuring that sufficient information can still be obtained for analysis and investigation in case of an accident. At the same time, through the frame dropping process of the locked event data, the system can more centrally record the data at critical moments, helping accident investigators better understand the process and causes of the event and improving the accuracy of accident analysis.

[0015] In a possible design, the autonomous driving storage area overwrites data in a first-in, first-out manner; the data in the accident storage area can only be overwritten when preset conditions are met.

[0016] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: The advanced first-in, first-out method is adopted in the autonomous driving storage area, which can ensure that the latest data is stored first, avoiding waste of storage space and improving the efficiency of data management. Moreover, the first-in, first-out mechanism enables the autonomous driving storage area to dynamically manage the storage space, ensuring that the system can still operate efficiently in the case of a large amount of data and avoiding excessive occupation of the storage space. The data in the accident storage area cannot be overwritten before meeting the preset conditions, ensuring the integrity and security of key accident data and providing a reliable basis for subsequent accident analysis and liability determination.

[0017] In a possible design, the intelligent driving domain control processing module further includes a power management module, and the power management module is used to control and manage the power supply of the autonomous driving vehicle to power the data recording system.

[0018] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: The power management module can intelligently adjust the power supply according to the actual needs of the data recording system, ensuring that the power requirements of the system in different working states are met and improving the power utilization efficiency. Moreover, by centrally managing the power supply, the power management module can effectively monitor the power supply status, timely detect and handle power anomalies, improve the stability and reliability of the system, and reduce the risk of system failures caused by power problems. At the same time, through a stable power supply, it is ensured that the data recording system can work properly at critical moments, improving the reliability and integrity of data recording.

[0019] In a possible design, the several information modules specifically include five information modules, namely basic information, vehicle status and dynamic information, autonomous driving system operation information, driving environment information, and driver operation and status information.

[0020] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By integrating the five information modules, the system can comprehensively collect various types of data related to autonomous driving, ensuring comprehensive monitoring of the vehicle operation status, environmental changes, and driver behavior, and improving the richness and diversity of data. Moreover, the data collected by the five information modules can provide a basis for subsequent analysis and optimization, can restore the specific situation of the accident scene, help clarify the responsibility division between the autonomous driving system and the driver, and help the autonomous driving R & D team continuously improve the autonomous driving algorithm and system performance, promoting technological progress.

[0021] In a possible design, the data recording system uses two sets of system-on-chip (SoC) of the autonomous driving control unit, vehicle sensors, and two sets of power supplies of the autonomous driving vehicle.

[0022] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By reusing existing hardware resources such as system-on-chip (SoC), sensors, and power supplies, the need for additional hardware is reduced, thereby lowering the overall cost of the system and improving its economy. Moreover, integrating multiple functional modules on the same platform reduces the complexity of the system, simplifies the design and implementation processes, and improves the integration and reliability of the system. At the same time, by reusing two sets of autonomous driving control units, when one of them fails, the data recording system can still operate on the other autonomous driving control unit, ensuring the stability of the system operation.

[0023] This application provides a data recording method for an autonomous driving vehicle, which is applied to a data recording system. The data recording system includes at least one autonomous driving control unit, and the autonomous driving control unit includes a data storage module and an intelligent driving domain control processing module. The data storage module includes an embedded multimedia card (EMMC) storage module and a second data management module. The embedded multimedia card (EMMC) storage module includes an autonomous driving storage area and an accident storage area. The data recording method includes: The data processing module receives vehicle recording data of several information modules of the autonomous driving vehicle; the data processing module performs data type arbitration on the vehicle recording data, calibrates them as non-locked event data and locked event data respectively, and transmits them to the data storage module; the second data management module stores the non-locked event data in the autonomous driving storage area and stores the locked event data in the accident storage area.

[0024] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By performing data type arbitration on the vehicle recording data, the data is divided into non-locked event data and locked event data. This classification management method enables different types of data to be effectively distinguished and processed, improving the efficiency and accuracy of data management. The autonomous driving storage area and the accident storage area in the embedded multimedia card (EMMC) storage module are used to store non-locked event data and locked event data respectively. This dedicated storage design ensures the security and integrity of key accident data, avoiding confusion with other non-critical data, and thus improving the traceability of the data. By classifying and specially storing the data, it provides a basis for subsequent intelligent decision-making. By analyzing the locked event data, the cause of the accident can be better understood, and then the autonomous driving algorithm and safety strategy can be optimized.

[0025] In a possible design, when the data processing module performs data type arbitration on the vehicle recording data and calibrates them as non-locked event data and locked event data respectively, it includes: When the data processing module receives a collision event or a collision risk event trigger signal, the vehicle recording data within the standard save period before that moment is calibrated as locked event data; if the data processing module receives a vehicle normal signal, all the received vehicle recording data is calibrated as non-locked event data.

[0026] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By receiving the triggering signals of collision events or collision risk events in real time, the data processing module can dynamically calibrate the data. This real-time response mechanism ensures that important data can be accurately identified and saved at critical moments, improving the timeliness and relevance of the data. The vehicle record data before the collision event is calibrated as locked event data to ensure that these data will not be overwritten or deleted after an accident. This protection mechanism improves the integrity of accident data and provides a reliable basis for subsequent accident analysis and liability determination. By calibrating the vehicle record data in the normal driving state as non-locked event data, the data in the normal and abnormal states can be effectively distinguished.

[0027] In a possible design, the method further includes: The second data management module can monitor the storage space status of the embedded multimedia card (EMMC) storage module, including: The second data management module detects the EMMC storage module. If the storage space of the accident storage area is insufficient for data storage for a specified number of times, a space shortage signal is sent to the autonomous vehicle; if the storage space of the accident storage area can meet the data storage for the specified number of times, the locked event data is stored normally.

[0028] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: The second data management module can detect the storage space status of the EMMC storage module in real time to ensure that there is sufficient storage space to record the locked event data when an accident occurs. This real-time monitoring mechanism improves the reliability of data storage and avoids data loss caused by insufficient storage space. When the storage space of the accident storage area is insufficient, the system can timely send a space shortage signal to the autonomous vehicle. This proactive warning mechanism enables the vehicle to take timely measures, such as clearing unnecessary data or adjusting the storage strategy, to ensure that the storage of critical data is not affected. By monitoring the storage space status, the second data management module can dynamically adjust the data storage strategy according to the actual situation of the storage space. When the storage space is sufficient, the locked event data is stored normally; when the storage space is insufficient, the system can preferentially retain important data and optimize data management. Description of the Drawings

[0029] Figure 1 Schematic architecture of a data recording system provided by an embodiment of the present application Figure 1 ; Figure 2 Schematic architecture of a data recording system provided by an embodiment of the present application Figure 2 ; Figure 3 Timing diagram of data storage provided by an embodiment of the present application; Figure 4 The data storage process provided by the embodiment of the present application Figure 1 ; Figure 5 The data storage process provided by the embodiment of the present application Figure 2 ; Figure 6 It is the data type arbitration flow chart provided by the embodiment of the present application. Specific implementation manners

[0030] To make the above objects, features, and advantages of the present application more obvious and understandable, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0031] Refer to Figures 1 to 6 , this embodiment provides a data recording system for an autonomous vehicle. The data recording system includes at least one autonomous driving control unit, and the autonomous driving control unit includes a data storage module and an intelligent driving domain control processing module; wherein, The data storage module is used to store the data information of the autonomous vehicle; the intelligent driving domain control processing module includes a data processing module and a first data management module; the data processing module is used to receive and process the vehicle record data of several information modules of the autonomous vehicle; the first data management module is used to monitor the vehicle record data of several information modules; when the data processing module receives a trigger signal for a collision event or a collision risk event, the vehicle record data of several information modules within a standard save period before that moment is stored in the data storage module.

[0032] Specifically, in this embodiment, the autonomous vehicle is an autonomous vehicle above L3 level. Among them, L3-level autonomous driving is an important level of autonomous driving technology and belongs to the autonomous driving classification standard defined by the International Organization of Automotive Engineers. According to this standard, at the L3 level, the vehicle can achieve full autonomous driving under specific conditions, and the driver does not need to continuously monitor the operation of the vehicle, but still needs to be able to take over control when the system requests. Autonomous vehicles above L3 level will have a dual power supply redundant design, which can use the redundant power supply design of the main and secondary autonomous driving domain controllers to ensure that after a collision accident or system failure, the other power supply can still support the normal operation of the autonomous driving data recording system and meet the data recording requirements before, during, and after the accident. Among them, refer to Figure 1 shown, the main autonomous driving controller and the secondary autonomous driving controller are each an autonomous driving control unit, and a data storage module and an intelligent driving domain control processing module are both provided in the main autonomous driving controller or the secondary autonomous driving controller.

[0033] Specifically, the key components required by traditional stand-alone data recording systems include a System on Chip (SoC), flash memory, Random Access Memory (RAM), and an Embedded Multimedia Card (EMMC), etc. The total cost of these components is relatively high. Moreover, the stand-alone autonomous driving data recording system needs to be equipped with a metal and plastic housing, an independent Printed Circuit Board (PCB) inside, connectors such as Ethernet and power supply, and also needs to be connected to sensors such as the intelligent driving domain controller or camera through an additional dedicated wiring harness, and its structural component cost is relatively high. In addition, to ensure that vehicle data can still be effectively recorded when the main circuit is powered off in an accident, this system usually needs to be equipped with a backup battery to power the data recording system when the main battery is damaged. The data recording system of this embodiment integrates the data recording function into the main and secondary autonomous driving domain controllers, which not only meets the data recording requirements but also eliminates the cost and complexity of an external stand-alone data recording system.

[0034] In an embodiment of the present application, the data storage module includes an Embedded Multimedia Card (EMMC) storage module and a second data management module; wherein, the second data management module is used to receive and manage the data transmitted from the intelligent driving domain control processing module, and store the data into the Embedded Multimedia Card (EMMC) storage module; the data backup methods of the second data management module include data reading on the PC side and data uploading to the cloud.

[0035] Specifically, in this embodiment, the Embedded Multimedia Card (EMMC) storage module, that is, the Embedded Multimedia Card storage module, is an integrated storage solution, usually used to store the operating system, application programs, and data. EMMC can provide fast data reading and writing speeds, thus ensuring the real-time performance and response ability of the system, and the low-power consumption characteristic of EMMC helps to extend the battery life of the vehicle. Among them, the second data management module is responsible for receiving and managing the data transmitted from the intelligent driving domain control processing module. The second data management module will store the received data in a specific storage area of the Embedded Multimedia Card (EMMC) storage module, and the second data management module can manage the data uploading to the cloud and downloading in the Embedded Multimedia Card (EMMC) storage module, as well as the processing of data reading permissions, etc. Data reading on the PC side refers to extracting and backing up the data in the autonomous driving vehicle by connecting to a personal computer. Cloud data uploading is to upload the data generated by the vehicle to the cloud server in real time or regularly. Among them, data reading on the PC side and cloud data uploading complement each other. Data reading on the PC side is suitable for local data management and analysis, while cloud data uploading provides convenience for the centralized storage and sharing of data. These two backup methods together improve the data management ability and flexibility of the autonomous driving system.

[0036] In an embodiment of the present application, the Embedded MultiMedia Card (EMMC) storage module includes an autonomous driving storage area and an accident storage area; wherein, the data transmitted by the intelligent driving domain control module includes non-locked event data recorded normally and locked event data recorded when a collision event or a collision risk event trigger signal is received; the non-locked event data is stored in the autonomous driving storage area, and the locked event data is stored in the accident storage area.

[0037] Specifically, in this embodiment, refer to Figure 3 and Figure 4 As shown, when the driver activates the autonomous driving system, the data recording system is activated accordingly and real-time collects and arbitrates vehicle information and video data; when a collision event or a collision risk event occurs, the second data management module of the system will arbitrate that this type of event needs to be stored according to the locked event type and stored in the accident storage area of the Embedded MultiMedia Card (EMMC) storage module, and the stored locked event data will not overwrite each other until the data in the accident storage area is uploaded to the cloud or securely stored locally, then it can be overwritten. On the contrary, for non-collision events or non-collision risk events, the system will arbitrate them as non-locked events and directly store them in the autonomous driving storage area of the Embedded MultiMedia Card (EMMC) storage module, and upload the data to the cloud or overwrite the stored data according to the principle of first-in, first-out and overwriteable.

[0038] In an embodiment of the present application, the second data management module can monitor the storage space status of the Embedded MultiMedia Card (EMMC) storage module; if there is insufficient remaining storage space in the accident storage area, the second data management module will send a space shortage signal to the autonomous driving vehicle.

[0039] Specifically, in this embodiment, refer to Figure 5 As shown, the second data management module will real-time monitor the storage space status of the accident storage area in the Embedded MultiMedia Card (EMMC) storage module. When there is insufficient remaining storage space in the accident storage area, the second data management module will send a space shortage alarm message to the dashboard of the autonomous driving vehicle to prompt the driver to migrate the stored locked event data, that is, upload the locked event data to the platform or server.

[0040] In an embodiment of the present application, the second data management module stores the locked event data with frame dropping.

[0041] Specifically, in this embodiment, refer to Figure 6As shown in the data type arbitration flowchart, vehicle external video data is an essential recorded data element in the data recording system. Vehicle external video data is mainly used for accident analysis and liability determination. Therefore, the minimum storage frequency requirements for vehicle external video data are different from those of the autonomous driving domain controller. The storage frequency of autonomous driving video data is generally 30 frames per second, but the minimum frame rate requirement of the data recording system is only 4 frames per second. Therefore, considering the storage space and erasure life of the EMMC, the autonomous driving domain controller with a data recording system needs to simplify the frame rate in the original data packet to achieve efficient storage. For the locked event data, the frame rate is reduced from the original 30 frames per second to 4 frames per second, and then the data is compressed and packed and stored in the accident storage area. Among them, frame rate reduction storage can reduce the storage of redundant data while ensuring the key features and important information of the event, thus saving the space of the accident storage area.

[0042] In an embodiment of the present application, the autonomous driving storage area overwrites data in a first-in, first-out manner; the data in the accident storage area can only be overwritten when preset conditions are met.

[0043] Specifically, in this embodiment, the accident storage area can store the locked event data of at least five collision events or collision risk events and cannot be directly overwritten. Only when the data in the accident storage area meets the preset conditions can it be overwritten. The preset conditions include uploading the data in the accident storage area to the platform or the server. Therefore, only when the vehicle manufacturer securely stores all the locked event data on the enterprise platform or the server can the locked event data be overwritten. Among them, since the data in the accident storage area needs to be uploaded before it can be overwritten, it can prompt the vehicle manufacturer to upload the data to the enterprise platform in a timely manner when an accident occurs, ensuring the timely acquisition and analysis of key data.

[0044] In an embodiment of the present application, the intelligent driving domain control processing module further includes a power management module, which is used to control and manage the power supply of the autonomous driving vehicle to supply power to the data recording system.

[0045] Specifically, in this embodiment, the power management module is responsible for managing the dual-channel power input provided by the autonomous driving vehicle and ensuring that when one of the power supplies fails, the other power supply can provide input in a timely manner, thereby ensuring the integrity of the data recording of the data recording system. In addition, the power management module can also provide warnings to the driver through instruments or the central control screen about the power failure information.

[0046] In an embodiment of the present application, several information modules specifically include five information modules, namely basic information, vehicle status and dynamic information, autonomous driving system operation information, driving environment information, and driver operation and status information.

[0047] Specifically, in this embodiment, refer to Figure 2 As shown, the five information modules of the autonomous vehicle each perform important functions. Among them, the basic information module: records the basic parameters of the vehicle, such as vehicle model, manufacturer, vehicle identification number, registration information, etc., to provide the identity information of the vehicle for management and tracking. The vehicle status and dynamic information module: records the current status of the vehicle, including speed, acceleration, direction, position, fuel quantity, battery status, etc., to monitor the running status of the vehicle in real time, ensure that the vehicle is driving within a safe range, and provide necessary dynamic data support for the autonomous driving system. The autonomous driving system operation information module: records the working status of the autonomous driving system, including algorithm operation conditions, sensor data, decision-making processes, control instructions, etc., to monitor the performance and stability of the autonomous driving system, ensure that the system can correctly understand and respond to environmental changes, and improve the safety and reliability of autonomous driving. The driving environment information module: records the data of the surrounding environment, including road conditions, traffic signs, pedestrians, other vehicles, weather conditions, etc., to provide environmental perception ability for the autonomous driving system, help the vehicle make safe driving decisions, and avoid potential dangers. The driver operation and status information module: records the operation behaviors and physiological states of the driver, such as the use of the steering wheel, accelerator, and brake, and the driver's fatigue, attention, etc., to monitor the status of the driver and ensure that the control can be taken over in a timely manner when needed.

[0048] In an embodiment of the present application, the data recording system uses two sets of system-on-chip (SoC) of the autonomous driving control unit, vehicle sensors, and two sets of power supplies of the autonomous vehicle.

[0049] Specifically, in this embodiment, the two sets of autonomous driving control units are respectively the main autonomous driving domain controller and the secondary autonomous driving domain controller. The system-on-chip (SoC) refers to the technology of integrating multiple functional modules on a single chip. This integrated design can significantly improve the performance and efficiency of the system, reduce power consumption and costs. The system-on-chip (SoC) usually integrates functional modules such as a processor, a graphics processing unit, a digital signal processor, a memory, and various interfaces, and can process complex computing tasks, such as sensor data processing, environmental perception, decision-making, etc. Among them, the vehicle sensors used by the data recording system include vehicle cameras. By multiplexing the data transmitted by the vehicle cameras to the main and secondary autonomous driving domain controllers, the transmission path of the camera video data from the autonomous driving domain controller to the data storage and recording system is omitted, improving the data transmission efficiency.

[0050] The present application provides a data recording method for an autonomous vehicle, which is applied to a data recording system. The data recording system includes at least one autonomous driving control unit. The autonomous driving control unit includes a data storage module and an intelligent driving domain control processing module. The data storage module includes an embedded multimedia card (EMMC) storage module and a second data management module. The embedded multimedia card (EMMC) storage module includes an autonomous driving storage area and an accident storage area. The data recording method includes: a data processing module receives vehicle recording data of several information modules of the autonomous vehicle; the data processing module performs data type arbitration on the vehicle recording data, calibrates them as non-locked event data and locked event data respectively, and transmits them to the data storage module; the second data management module stores the non-locked event data in the autonomous driving storage area and stores the locked event data in the accident storage area.

[0051] Specifically, by performing data type arbitration on the vehicle recording data, the data is divided into non-locked event data and locked event data, which enables different types of data to be effectively distinguished and processed, improving the efficiency and accuracy of data management.

[0052] In an embodiment of the present application, the data processing module performs data type arbitration on the vehicle recording data, calibrates them as non-locked event data and locked event data respectively, including: when the data processing module receives a collision event or a collision risk event trigger signal, the vehicle recording data within the standard save period before that moment is calibrated as locked event data; if the data processing module receives a vehicle normal signal, all the received vehicle recording data is calibrated as non-locked event data.

[0053] Specifically, by receiving a collision event or a collision risk event trigger signal and a vehicle normal signal to calibrate the data type, it can ensure the accurate identification and preservation of important data, improving the timeliness and relevance of the data. Only when the autonomous vehicle detects a collision or a collision risk will it send a collision event or a collision risk event trigger signal to the data processing module to save the key data. When the vehicle is running normally, the autonomous vehicle sends a vehicle normal signal to the data processing module, and the vehicle data will be recorded as non-locked event data. Since the vehicle is in a normal state most of the time, the amount of non-locked event data is very large. Setting the storage method of non-locked event data as a first-in-first-out overwrite mode can ensure that subsequent data can be normally stored in the embedded multimedia card (EMMC) storage module.

[0054] In one embodiment of the present application, the method further includes: The second data management module can monitor the storage space status of the Embedded Multimedia Card (EMMC) storage module, including: The second data management module detects the EMMC storage module. If the storage space of the accident storage area is insufficient for storing data a specified number of times, a space shortage signal is sent to the autonomous vehicle; if the storage space of the accident storage area can meet the data storage requirements for the specified number of times, the locked event data is stored normally.

[0055] Specifically, by monitoring whether the storage space of the accident storage area can meet the storage requirements for a specified number of times, it can be ensured that when a collision event occurs, the locked event data can be normally stored in the accident storage area to prevent data loss.

[0056] Although the present application is disclosed as above, the present application is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present application. Therefore, the protection scope of the present application should be subject to the scope defined by the claims.

Claims

1. A data recording system for an autonomous driving vehicle, characterized in that: The data recording system includes at least one autonomous driving control unit, and the autonomous driving control unit includes a data storage module and an intelligent driving domain control processing module; wherein, The data storage module is used to store data information of the autonomous driving vehicle; The intelligent driving domain control processing module includes a data processing module and a first data management module; the data processing module is used to receive and process vehicle recording data of several information modules of the autonomous driving vehicle; the first data management module is used to monitor the vehicle recording data of several information modules; when the data processing module receives a collision event or a collision risk event triggering signal, the vehicle recording data of several information modules within the standard storage period before that moment is stored in the data storage module.

2. The data recording system for an autonomous driving vehicle according to claim 1, characterized in that: The data storage module includes an embedded multimedia card EMMC storage module and a second data management module; Among them, the second data management module is used to receive and manage data transmitted from the intelligent driving domain control processing module, and store the data in the EMMC storage module; the data backup method of the second data management module includes PC data reading and cloud data uploading.

3. The data recording system for an autonomous driving vehicle according to claim 2, characterized in that: The embedded multimedia card EMMC storage module includes an automatic driving storage area and an accident storage area; Among them, the data transmitted by the intelligent driving domain control processing module includes normally recorded non-locking event data and locked event data recorded when a collision event or collision risk event triggering signal is received; the non-locking event data is stored in the automatic driving storage area, and the locked event data is stored in the accident storage area.

4. The data recording system for an autonomous driving vehicle according to claim 3, characterized in that: The second data management module can monitor the storage space status of the embedded multimedia card EMMC storage module; if there is insufficient remaining storage space in the accident storage area, the second data management module sends an insufficient space signal to the autonomous driving vehicle.

5. The data recording system for an autonomous driving vehicle according to claim 3, characterized in that: The second data management module stores the locking event data in reduced frames.

6. The data recording system for an autonomous driving vehicle according to claim 3, characterized in that: The autonomous driving storage area overwrites data in a first-in-first-out manner; the data in the accident storage area can only be overwritten when preset conditions are met.

7. The data recording system for an autonomous driving vehicle according to claim 1, characterized in that: The intelligent driving domain control processing module also includes a power management module, which is used to control and manage the power supply of the autonomous driving vehicle to power the data recording system.

8. A data recording method for an autonomous driving vehicle, characterized in that: The data recording system applied to any one of claims 1 to 7 comprises at least one autonomous driving control unit, the autonomous driving control unit comprises a data storage module and an intelligent driving domain control processing module, the data storage module comprises an embedded multimedia card EMMC storage module and a second data management module, the embedded multimedia card EMMC storage module comprises an autonomous driving storage area and an accident storage area; the data recording method comprises: The data processing module receives vehicle recording data from a plurality of information modules of the autonomous driving vehicle; The data processing module performs data type arbitration on the vehicle recorded data, marks them as non-locking event data and locking event data, and transmits them to the data storage module; The second data management module stores the non-locking event data in the automatic driving storage area, and stores the locking event data in the accident storage area.

9. The data recording method of the autonomous driving vehicle according to claim 8, characterized in that: The data processing module performs data type arbitration on the vehicle recorded data and marks them as non-locking event data and locking event data, respectively, including: When the data processing module receives a collision event or collision risk event triggering signal, the vehicle record data within the standard storage period before this moment is marked as locking event data; If the data processing module receives a normal vehicle signal, the received vehicle record data are all marked as non-locking event data.

10. The data recording method of the autonomous driving vehicle according to claim 8, characterized in that: The method further comprises: The second data management module detects the embedded multimedia card EMMC storage module, and if there is a storage space in the accident storage area that is insufficient for a specified number of data storages, sends an insufficient space signal to the automatic driving vehicle; If the storage space of the accident storage area can satisfy the data storage of the specified number of times, the locking event data is stored normally.

Images