Automatic driving data recording memory and data reading method

Abstract

The invention belongs to the technical field of automatic driving. According to the automatic driving data recording memory and the data reading method, a data input interface is used for receiving original data from a vehicle sensor, a data processing unit is used for preprocessing the data transmitted from the data input interface, and a storage medium is used for storing the preprocessed data; the power management module is used for performing power supply management and providing power supply for a set time when a vehicle power supply is interrupted, and the communication interface is used for transmitting data to external equipment; and the control and management unit is respectively connected with the data input interface, the data processing unit, the storage medium, the power management module and the communication interface, and is used for controlling and managing each component of the automatic driving data recording memory. According to the invention, complete and accurate storage of the automatic driving data is realized, the automatic driving data cannot be lost even if a vehicle is collided or a network fault suddenly occurs, and the integrity and completeness of the data are ensured.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to an autonomous driving data recording storage device and a data reading method. Background Art

[0002] The statements in this section merely provide background art related to the present invention and do not necessarily constitute prior art.

[0003] Autonomous vehicles can complete driving operations without human drivers by using technologies such as artificial intelligence, visual computing, radar, monitoring devices and global positioning systems. During the operation of the autonomous driving system, it is necessary to record and store the operating status monitoring data. At present, most of them adopt the intelligent driving domain controller storage or intelligent driving cloud platform storage. The intelligent driving domain controller storage generally uses the data storage unit in the autonomous driving domain controller for storage, and the intelligent driving cloud platform storage generally adopts the idle time backhaul solution to transfer the data to the cloud platform storage.

[0004] However, the intelligent driving domain controller generally has a small storage capacity and a limited storage period. It can only read data for a limited period of time through manufacturer-specific equipment and methods, and cannot achieve complete storage of autonomous driving operation status monitoring data. The storage method of uploading to the intelligent driving cloud platform has problems such as data packet loss or unsuccessful upload. Once the vehicle-side data is overwritten, it may affect data analysis. Summary of the invention

[0005] In order to address the deficiencies in the prior art, the present invention provides an autonomous driving data recording storage device and a data reading method, which realize complete and accurate storage of autonomous driving data. Even if a vehicle collides or a sudden network failure occurs, the autonomous driving data will not be lost. When the cloud data is not fully transmitted back, the data stored before the failure can be read through an event reading device, thereby ensuring the integrity and completeness of the data.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides an autonomous driving data recording storage device.

[0008] An autonomous driving data recording storage device, comprising: a data input interface, a data processing unit, a storage medium, a power management module, a communication interface, and a control and management unit;

[0009] The data input interface is used to receive raw data from a vehicle sensor, the data processing unit is used to pre-process the data transmitted from the data input interface, and the storage medium is used to store the pre-processed data;

[0010] The power management module is used to manage power supply and provide power supply for a set time when the vehicle power is interrupted, and the communication interface is used to transmit data to an external device;

[0011] The control and management unit is respectively connected to the data input interface, the data processing unit, the storage medium, the power management module, and the communication interface, and is used to control and manage the various components of the autonomous driving data recording storage.

[0012] As a further limitation of the first aspect of the present invention, the raw data of the vehicle sensor includes: raw data of a camera, a radar, a lidar and a GPS.

[0013] As a further limitation of the first aspect of the present invention, the preprocessing at least includes: data compression, format conversion and data cleaning.

[0014] As a further limitation of the first aspect of the present invention, the communication interface includes: a wired communication interface and a wireless communication interface, and the external device includes a smart driving cloud platform or an external data reading device.

[0015] As a further limitation of the first aspect of the present invention, it also includes an encryption and security module, and the encryption and security module is used to encrypt data before the data is transmitted through the communication interface.

[0016] As a further limitation of the first aspect of the present invention, the data input interface, the data processing unit, the storage medium, the power management module, the communication interface, the data encryption and security module, and the control and management unit are all arranged in an environmental protection housing.

[0017] As a further limitation of the first aspect of the present invention, the communication interface is connected to the intelligent driving cloud platform through a gateway via the Internet of Vehicles, the autonomous driving data recording storage receives raw data and Json strings from the intelligent driving domain controller through the data input interface, the raw data and Json strings should be sent or received in pairs, the intelligent driving cloud platform receives metadata and structured and unstructured raw data sent from the communication interface through the https protocol, and the communication interface is connected to the PKI platform through the Internet of Vehicles.

[0018] As a further limitation of the first aspect of the present invention, the data input interface is connected to the intelligent driving domain controller, the intelligent driving domain controller is connected to each sensor of the vehicle, the intelligent driving domain controller receives sensor information, performs perception fusion, completes intelligent driving path planning, and issues control instructions, and pre-buries a trigger mechanism to save and back up the perception results, planning results, control instructions and original perception data in collision and specific scenarios, and send them to the autonomous driving data recording storage device.

[0019] In a second aspect, the present invention provides a method for reading data from an autonomous driving data recording memory.

[0020] A method for reading data from an autonomous driving data recording memory is used for the autonomous driving data recording memory according to the first aspect of the present invention, wherein an external event reading device is connected to a communication interface, and comprises the following process:

[0021] The event reading device requests to read the time range of the stored autonomous driving data recording system event data. If the return value is invalid or a negative response is returned, the event reading device terminates the process; if the return value is valid, the time range of the event data is obtained;

[0022] The event reading device requests the ECU storing the data to enter the extended mode. If the return value is invalid or a negative response is returned, the event reading device terminates the process;

[0023] The event reading device requests to enter the secure access mode. If the return value is invalid or a negative response is returned, the event reading device terminates the process;

[0024] The event reading device requests the driver ECU to dump the event data to the specified folder and compress it into a compressed package. If the return value is invalid or a negative response is returned, the event reading device terminates the process;

[0025] The event reading device queries the compression progress. If the return value is invalid or a negative response is returned, the event reading device terminates the process;

[0026] The event reading device requests to upload event data. If the return value is invalid or a negative response is returned, the event reading device terminates the process;

[0027] The event reading device requests to start data transmission. If the return value is invalid or a negative response is returned, the event reading device terminates the process;

[0028] After data transmission is completed or when data transmission needs to be interrupted, the event reading device requests to exit data transmission. If the return value is invalid or a negative response is returned, the event reading device terminates the process;

[0029] The event reading device requests to verify whether the transmitted file is complete. If the return value is invalid or a negative response is returned, the event reading device terminates the process; if the verification is successful, the autonomous driving data recording storage device automatically deletes the compressed package.

[0030] As a further limitation of the second aspect of the present invention, when the event reading device requests to verify whether the transmitted file is complete, the request instruction issued by the event reading device includes the file characteristics received by the event reading device when reading the event data, and the return instruction received by the event reading device follows the verification result, and the file characteristics are calculated based on the compressed package.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] 1. The present invention innovatively proposes an autonomous driving data recording storage device, including a data input interface, a data processing unit, a storage medium, a power management module, a communication interface, and a control and management unit, which realizes the collection, preprocessing, storage and output of autonomous driving data. The data input interface is directly connected to the vehicle domain controller to receive complete autonomous driving process data, thereby ensuring the integrity of data storage.

[0033] 2. The present invention innovatively proposes an autonomous driving data recording storage device. The power management module is used to manage the power supply and provide power supply for a set time when the vehicle power is interrupted. Even if the vehicle collides or a sudden network failure occurs, the autonomous driving data will not be lost and can continue to work within the set time period until the autonomous driving data is completely saved.

[0034] 3. The present invention innovatively proposes an autonomous driving data recording storage device, which also includes an encryption and security module. The encryption and security module is used to encrypt data before the data is transmitted through the communication interface, thereby ensuring the security of data transmission and avoiding the leakage of autonomous driving data during the transmission process.

[0035] 4. The present invention innovatively proposes an autonomous driving data recording storage device. The data input interface, data processing unit, storage medium, power management module, communication interface, data encryption and security module, and control and management unit are all arranged in an environmental protection casing. They are all arranged in an environmental protection casing, which meets the requirements of dust and water resistance, EMC and electrical performance, and can withstand the test of harsh environments.

[0036] 5. The present invention innovatively proposes a method for reading data from an autonomous driving data recording storage device. Even if the cloud data has not been fully transmitted back, the data stored before the failure can be read through an event reading device, ensuring the safe, stable storage and readability of the autonomous driving data, thereby enabling the data to be effectively used for subsequent data analysis.

[0037] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0039] Figure 1 A schematic diagram of the automatic driving data recording storage device provided by the present invention;

[0040] Figure 2 This is a schematic diagram of the usage status of the autonomous driving data recording storage provided by the present invention. DETAILED DESCRIPTION

[0041] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0042] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0043] In this implementation, an automatic driving data recording storage is proposed, such as Figure 1 As shown, it includes: a data input interface, a data processing unit, a storage medium, a power management module, an encryption and security module, a communication interface, and a control and management unit.

[0044] The data input interface usually includes multiple interfaces for receiving raw data from vehicle sensors (such as cameras, radars, lidars, GPS, etc.). The data processing unit is responsible for receiving the data from the data input interface and performing preprocessing, such as data compression, format conversion, data cleaning, and feature extraction, to reduce storage requirements and improve storage efficiency.

[0045] This implementation method can select a suitable compression algorithm according to the data type. For example, H.264 or H.265 video compression algorithms can be used for video data; octree, KD tree and other data structures can be used for point cloud data compression; when implementing the compression algorithm, attention should be paid to balancing the compression rate and processing speed to ensure the best effect with limited computing resources.

[0046] In this implementation, preferably, the raw data can be converted into a format more suitable for storage and subsequent processing, for example, converting radar data from raw binary format into a standard CSV or JSON format, and converting video data into a format more suitable for long-term storage (such as MP4).

[0047] In this implementation, preferably, redundant information, noise and erroneous data in the data can be removed, and data cleaning algorithms or rules can be used to ensure the accuracy and consistency of the data.

[0048] In this implementation, preferably, key features can be extracted from the original data for subsequent data analysis and machine learning model training. The feature extraction algorithm should be designed and implemented according to the specific needs of the autonomous driving system.

[0049] In this implementation, the storage medium is used as the actual storage location of the data, and a high-performance solid-state drive (SSD) or flash memory is usually used to store the pre-processed data.

[0050] The power management module is responsible for providing a stable power supply for the entire system and ensuring that the memory can continue to work for a period of time to complete the storage of key data when the vehicle power is interrupted. For example, the following method can be adopted: the power management module should be designed with multiple power input interfaces, including the vehicle's main power supply and backup power supply (such as a battery pack or supercapacitor). When the vehicle's main power supply is normal, the main power supply will power the system; when the vehicle's main power supply is interrupted, it will automatically switch to the backup power supply.

[0051] The communication interface includes wired and wireless communication interfaces, which are used to transmit data to external devices, such as remote servers or maintenance equipment, when needed. The encryption and security module is responsible for data encryption and security, ensuring the safety of data during storage and transmission. The environmental protection shell surrounds the entire autonomous driving data recording storage, providing physical protection to ensure normal operation in harsh environments.

[0052] In this implementation, the control and management unit is responsible for the control and management of the entire system, including data management, system monitoring, fault detection and recovery, etc. Optionally, the control and management unit can use the following process to implement its management and control functions:

[0053] (1) Data management.

[0054] The data generated by the autonomous driving system should be classified, including but not limited to video data, radar data, sensor data, system logs, etc.; a reasonable storage structure should be designed, such as using a database or file system to store data, and setting the data storage path, naming rules, etc.; ensure the integrity and consistency of the data, and improve the reliability of the data through data verification and redundant storage.

[0055] Back up critical data regularly to prevent data loss or damage; backup data should be stored in a secure location, such as a remote server or cloud storage, and provide data recovery capabilities so that data can be restored from backup data when it is lost or damaged; implement strict data access control policies to limit illegal access to data, and record data access logs to track and audit data access.

[0056] (2) System monitoring.

[0057] Monitor the system operation status in real time through monitoring software or hardware tools, including the status of data processing units, storage media, power management modules, etc.; monitor key parameters of the autonomous driving system, such as sensor data, control signals, etc., to ensure the normal operation of the system;

[0058] Record system operation logs, including system startup, shutdown, abnormality and other events, and record key events during autonomous driving, such as emergency braking and system takeover, for subsequent analysis and troubleshooting;

[0059] Set alarm thresholds. When system parameters exceed the thresholds, alarms are triggered. Early warning functions are provided to predict potential system failures or anomalies so that timely measures can be taken to deal with them.

[0060] (3) Fault detection.

[0061] Use hardware self-test or external test tools to test the status of hardware devices, such as whether the power management module, storage medium, etc., are faulty. When a hardware fault is detected, an alarm is triggered and the fault information is recorded;

[0062] Monitor the software running status of the data processing unit, detect whether there are software anomalies or crashes, and identify software failures through log analysis or anomaly detection algorithms;

[0063] Combined with the hardware and software fault detection results, the fault source is comprehensively located and detailed fault information is provided, including fault type, occurrence time, impact range, etc., so that technical personnel can conduct fault troubleshooting and repair.

[0064] (4) Fault recovery.

[0065] For some minor faults, the system can automatically perform recovery operations, such as restarting the data processing unit, restoring the default configuration, etc., and set automatic recovery strategies to ensure that the system can be quickly restored to normal operation when a fault occurs;

[0066] Provides a manual recovery function, allowing technicians to manually repair faults based on fault reports, and provides recovery guides or tools to help technicians quickly locate and resolve problems;

[0067] When data is lost due to system failure, data backup is used for recovery and data recovery software or tools are provided to help technicians recover data from damaged storage media.

[0068] (5) Encryption and security management.

[0069] Sensitive data is encrypted and stored to ensure data security. Advanced encryption algorithms and technologies are used to ensure the security of data during transmission and storage. A security authentication mechanism is implemented to ensure that only authenticated users or devices can access the data. Authentication logs are recorded to track and audit authentication status.

[0070] (6) Communication interface management.

[0071] Provide efficient data transmission interface to ensure that data can be transmitted to the autonomous driving system or other related systems in real time and accurately, and support multiple data transmission protocols and formats to meet the needs of different systems;

[0072] Remote monitoring and management functions are realized through the communication interface, allowing technicians to remotely view system status, configuration parameters, etc., and provide remote upgrade and update functions to promptly repair system vulnerabilities and update software versions.

[0073] like Figure 2 As shown, the sensors in this implementation (such as millimeter-wave radar, camera, lidar and map positioning sensor, etc.) are used to identify the external environment data of the vehicle and transmit it to the intelligent driving domain controller; the intelligent driving domain controller receives the sensor information, performs perception fusion, completes intelligent driving path planning and issues control instructions, and pre-buries the trigger mechanism to save and back up the perception results, planning results, control instructions and original perception data in collision and specific scenarios, and send them to the autonomous driving data recording storage.

[0074] The communication interface of the autonomous driving data recording storage is connected to the gateway of the Zhijia Cloud Platform through TBOX (Internet of Vehicles system, used to connect the vehicle to the Internet), and is connected to the PKI platform through TBOX. The PKI platform of this implementation is a public key infrastructure. As a key management platform, its core functions include certificate management, key management, identity authentication, etc. It can provide security services such as data encryption and decryption, digital signature, identity authentication, etc., and is an important part of network security communication.

[0075] In this implementation, the Zhijia Cloud Platform serves as a cloud-based data management platform for intelligent driving and is responsible for data storage and processing services such as data storage, data management, and data analysis.

[0076] The autonomous driving data recording storage in this implementation has a high data throughput and can record a large amount of sensor data during vehicle driving, such as camera data, radar data, lidar data, etc., in real time to ensure the real-time and integrity of the data.

[0077] In this implementation, preferably, a distributed storage architecture can be used to disperse and store data on multiple storage nodes to improve the overall data reading and writing capabilities; RAID (Redundant Array of Independent Disks) technology can be used to provide higher data throughput and data reliability by combining multiple physical disks.

[0078] The autonomous driving data recording storage in this implementation has data redundancy and fault tolerance functions, adopts a data redundant storage strategy, and uses multiple independent storage units in the storage medium to store data. Even if some storage units fail, the key data can be completely preserved, thereby enhancing data reliability and fault tolerance.

[0079] The autonomous driving data recording storage in this implementation can realize intelligent data management, intelligently identify and prioritize the storage of key data, define data upload task priorities, scheduling, and temporary shutdown under high load conditions, reduce unnecessary data storage, and optimize the use of storage space. For example, it can be implemented in the following ways:

[0080] (1) Intelligent identification and priority preservation of key data. Through machine learning technology, models are trained to identify important features and patterns in data. Using these models, key data, such as core business indicators and user behavior data, can be intelligently identified. According to the business needs of autonomous driving, it is clear which data is critical and which data is secondary or unnecessary, and data classification and labeling rules are formulated so that the importance of data can be quickly identified when it is generated or received. Before the data is written to the storage, filters are used to screen the data according to key data standards, and key data is saved first. For secondary or unnecessary data, it can be chosen not to save or save it to a lower-cost storage medium.

[0081] (2) Define the priority of data upload tasks. Evaluate the importance of each data upload task based on factors such as the business value, urgency, and timeliness of the data, set a priority label for each data upload task, such as high, medium, and low, and arrange the execution order of data upload tasks based on the priority label to ensure that high-priority tasks can obtain resources and processing time first.

[0082] (3) Scheduling and temporary shutdown under high-load conditions. Dynamically adjust the scheduling strategy of data upload tasks according to system load and resource usage. When resources are sufficient, speed up the processing of tasks; when resources are tight, appropriately slow down the processing speed or suspend secondary tasks; monitor the system load in real time, including CPU usage, memory occupancy, disk I / O, etc. When the load reaches the preset threshold, trigger the early warning mechanism. Under high-load conditions, in order to protect system stability and data security, some or all data upload tasks can be temporarily closed, and the execution of these tasks can be gradually resumed after the system load is reduced.

[0083] (4) Reduce unnecessary data storage and optimize storage space usage. Perform deduplication processing on stored data to eliminate duplicate data, use data compression technology to reduce the storage space occupied by data, formulate data lifecycle management strategies based on the business value and timeliness of data, and delete or archive expired or no longer needed data.

[0084] The autonomous driving data recording storage in this implementation adopts advanced encryption technology to protect data from unauthorized access and ensure data security and privacy protection.

[0085] In this implementation, a symmetric encryption algorithm can be selected, such as an advanced symmetric encryption algorithm such as AES (Advanced Encryption Standard) to ensure the security of data during transmission and storage. AES provides three key lengths of 128 bits, 192 bits and 256 bits, which has high security. Alternatively, an asymmetric encryption algorithm such as RSA (Rivest-Shamir-Adleman) or ECC (Elliptic Curve Cryptography) can be used for data signing and key exchange to ensure data integrity and confidentiality.

[0086] In this implementation, the focus is on encryption during data transmission, using encryption protocols such as TLS (Transport Layer Security) or IPsec (Internet Protocol Security) to encrypt the original data to ensure that the data is not intercepted or tampered with during transmission. It is understandable that storage encryption can also be performed, that is, the data stored in the autonomous driving data recording storage is encrypted and stored using encryption algorithms such as AES to ensure that even if the data is stolen, it cannot be easily decrypted by unauthorized users.

[0087] The autonomous driving data recording storage in this implementation has the function of remote data transmission, supports real-time or on-demand remote data transmission, facilitates remote monitoring and accident analysis, and improves accident response speed and processing efficiency. The autonomous driving data recording storage in this implementation has high environmental adaptability, and the design takes into account stable operation under harsh environmental conditions such as extreme temperature and vibration, ensuring reliable data recording under various conditions.

[0088] The autonomous driving data recording storage in this implementation adopts a standardized and compatible design, follows industry standards, and ensures the compatibility and interchangeability of data between autonomous driving vehicles of different brands and models. The autonomous driving data recording storage in this implementation not only meets the high requirements of autonomous driving vehicles for data recording, but also provides a solid foundation for vehicle safety, accident analysis, and continuous improvement of autonomous driving technology.

[0089] The autonomous driving data recording storage device of this implementation method can upload data to the intelligent driving cloud platform. The specific implementation method is as follows:

[0090] The autonomous driving data recording storage should receive raw data and Json strings from the intelligent driving domain controller through the network protocol. The raw data and Json strings should be sent or received in pairs; the intelligent driving cloud platform receives metadata and structured and unstructured raw data sent from the autonomous driving data recording storage through the https protocol.

[0091] For example:

[0092] Raw data transmission

[0093] Request URL:

[0094] POST{}

[0095] The Url will be provisioned depending on the actual deploymentenvironment.

[0096] HTTP / 1.1

[0097] Method:POST

[0098] Host: follow server-side design

[0099] URL: follow server-side design

[0100] Content-type:application / octet-stream

[0101] Http status code:

[0102]

[0103]

[0104] In view of the autonomous driving data recording storage device designed above, the present implementation also provides a data reading method, which is implemented by an event reading device, and the event reading device is connected to the communication interface of the autonomous driving data recording storage device, including the following process:

[0105] S1: The event reading device requests to read the time range of the event data of the autonomous driving data recording system stored in the vehicle. If the return value is invalid or a negative response is returned, the event reading device must terminate the process; the return instruction needs to be followed by the time range of the event data of the autonomous driving data recording system of the vehicle;

[0106] S2: The event reading device requests the data ECU to enter the extended mode. If the return value is invalid or a negative response is returned, the event reading device must terminate the process;

[0107] S3: The event reading device requests to enter the secure access mode. If the return value is invalid or a negative response is returned, the event reading device shall terminate the process;

[0108] S4: The event reading device drives the ECU (i.e., the control and management unit corresponding to the autonomous driving data recording memory) to dump the event data to a specified folder and compress it into a ZIP package. If the return value is invalid or a negative response is returned, the event reading device must terminate the process;

[0109] S5: The event reading device requests to query the compression progress. If the return value is invalid or a negative response is returned, the event reading device must terminate the process;

[0110] S6: The event reading device requests to upload event data. If the return value is invalid or a negative response is returned, the event reading device shall terminate the process;

[0111] S7: The event reading device requests to start data transmission. If the return value is invalid or a negative response is returned, the event reading device must terminate the process.

[0112] S7: The event reading device requests to exit data transmission. If the return value is invalid or a negative response is returned, the event reading device shall terminate the process;

[0113] S7: The event reading device requests to verify whether the transmitted file is complete. The request instruction issued by the event reading device needs to add the file characteristics (CRC32) received by the reading device to read the event data. The return instruction received by the event reading device must follow the verification result; if the return value is invalid or returns a negative response, the event reading device must terminate the process. After the CRC32 verification is successful, the autonomous driving data recording storage device automatically deletes the compressed package.

[0114] In this implementation, the calculation of CRC32 is based on the compression package. CRC32, or Cyclic Redundancy Check 32-bit, is an algorithm based on binary polynomials. It regards the input data as a large integer and performs a modulo division operation with a fixed generating polynomial. The remainder obtained is the CRC checksum. This checksum is a 32-bit integer used to verify the integrity of the data.

[0115] Through the above-mentioned reading method, the autonomous driving data recording storage device can be uploaded to the cloud and can also be read through the event reading device. Even if the cloud data has not been fully transmitted back, the data stored before the fault can be read through the event reading device, ensuring the safe, stable storage and readability of the autonomous driving data, so that the data can be effectively used for subsequent data analysis.

[0116] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An autonomous driving data recording storage device, It is characterized in that It includes: data input interface, data processing unit, storage medium, power management module, communication interface and control and management unit; The data input interface is used to receive raw data from a vehicle sensor, the data processing unit is used to pre-process the data transmitted from the data input interface, and the storage medium is used to store the pre-processed data; The power management module is used to manage power supply and provide power supply for a set time when the vehicle power is interrupted, and the communication interface is used to transmit data to an external device; The control and management unit is respectively connected to the data input interface, the data processing unit, the storage medium, the power management module, and the communication interface, and is used to control and manage the various components of the autonomous driving data recording storage.

2. The automatic driving data recording storage device according to claim 1, characterized in that: The raw data of the vehicle sensors include: raw data of cameras, radars, lidars and GPS.

3. The automatic driving data recording storage device according to claim 1, wherein: The preprocessing at least includes: data compression, format conversion and data cleaning.

4. The automatic driving data recording storage device according to claim 1, wherein: The communication interface includes: a wired communication interface and a wireless communication interface, and the external device includes a smart driving cloud platform or an external data reading device.

5. The automatic driving data recording storage device according to claim 1, wherein: It also includes an encryption and security module, which is used to encrypt data before transmitting the data through the communication interface.

6. The automatic driving data recording storage device according to claim 5, characterized in that: The data input interface, the data processing unit, the storage medium, the power management module, the communication interface, the data encryption and security module and the control and management unit are all arranged in an environmental protection housing.

7. The automatic driving data recording storage device according to any one of claims 1 to 6, characterized in that: The communication interface is connected to the intelligent driving cloud platform through a gateway via the Internet of Vehicles. The autonomous driving data recording storage receives raw data and Json strings from the intelligent driving domain controller through the data input interface. The raw data and Json strings should be sent or received in pairs. The intelligent driving cloud platform receives metadata and structured and unstructured raw data sent from the communication interface through the https protocol. The communication interface is connected to the PKI platform through the Internet of Vehicles.

8. The automatic driving data recording storage device according to any one of claims 1 to 6, characterized in that: The data input interface is connected to the intelligent driving domain controller, and the intelligent driving domain controller is connected to various sensors of the vehicle. The intelligent driving domain controller receives sensor information, performs perception fusion, completes intelligent driving path planning, and issues control instructions, and pre-buries a trigger mechanism to save and back up the perception results, planning results, control instructions and original perception data in collision and specific scenarios, and send them to the autonomous driving data recording storage device.

9. A method for reading data from an autonomous driving data recording memory according to any one of claims 1 to 8, characterized in that: The external event reading device is connected to the communication interface, including the following process: The event reading device requests to read the time range of the stored autonomous driving data recording system event data. If the return value is invalid or a negative response is returned, the event reading device terminates the process; if the return value is valid, the time range of the event data is obtained; The event reading device requests the ECU storing the data to enter the extended mode. If the return value is invalid or a negative response is returned, the event reading device terminates the process; The event reading device requests to enter the secure access mode. If the return value is invalid or a negative response is returned, the event reading device terminates the process; The event reading device requests the driver ECU to dump the event data to the specified folder and compress it into a compressed package. If the return value is invalid or a negative response is returned, the event reading device terminates the process; The event reading device queries the compression progress. If the return value is invalid or a negative response is returned, the event reading device terminates the process; The event reading device requests to upload event data. If the return value is invalid or a negative response is returned, the event reading device terminates the process; The event reading device requests to start data transmission. If the return value is invalid or a negative response is returned, the event reading device terminates the process; After data transmission is completed or when data transmission needs to be interrupted, the event reading device requests to exit data transmission. If the return value is invalid or a negative response is returned, the event reading device terminates the process; The event reading device requests to verify whether the transferred file is complete. If the return value is invalid or a negative response is returned, the event reading device terminates the process; If the verification is successful, the autonomous driving data recording storage device will automatically delete the compressed package.

10. The method for reading data from the automatic driving data recording memory according to claim 9, characterized in that: When the event reading device requests to verify whether the transmitted file is complete, the request instruction issued by the event reading device includes the file characteristics received by the event reading device when reading the event data. The return instruction received by the event reading device follows the verification result, and the file characteristics are calculated based on the compressed package.

Images