Shadow mode acquisition triggering method and device based on driver behavior score

Abstract

The invention provides a shadow mode acquisition triggering method and device based on driver behavior scoring, and the method comprises the steps: receiving driver behavior data through a cloud server, and scoring the driver behavior data; comparing the driver behavior score with a score threshold, and judging whether the driver behavior has a training value or not; if yes, generating a shadow mode acquisition configuration file matched with the target vehicle and issuing the shadow mode acquisition configuration file to the target vehicle; the target vehicle collects the actual operation behavior of a driver, the current sensing data and the vehicle state in real time; simulating driving behavior prediction in the same scene based on the sensing data and the vehicle state; the actual operation behavior of the driver is compared with the driving behavior prediction, and if the deviation between the actual operation behavior of the driver and the driving behavior prediction is larger than a comparison threshold value, the actual operation behavior of the driver is transmitted back to the cloud server. Therefore, acquisition and uploading of low-quality or redundant data are greatly reduced, and the operation efficiency and economical efficiency of a data closed-loop system are improved.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving algorithm optimization technology, and in particular to a method and apparatus for triggering shadow pattern acquisition based on driver behavior scoring. Background Technology

[0002] Currently, autonomous driving systems rely on large amounts of real-world road data for training and model iteration. Among these methods, "shadow mode" is a technique that compares the actions of a human driver with the predictions of the autonomous driving algorithm during normal vehicle operation. This is used to identify algorithmic flaws and optimize model performance.

[0003] However, existing technologies generally suffer from the following problems: all vehicles continuously collect data without differentiation, resulting in a large amount of low-value information being uploaded to the cloud, causing serious redundancy in storage and computing resources; at the same time, the system lacks an effective evaluation mechanism for driver behavior, making it difficult to identify high-quality driving samples with "learnability" or "exemplary" characteristics; in addition, existing solutions lack dynamic feedback control capabilities from the cloud to the vehicle, making it impossible to flexibly adjust data collection strategies based on driver performance; this directly leads to a waste of bandwidth and computing resources—data generated by many invalid drivers contributes little to model training but still occupies a large amount of system resources, affecting the overall data closed-loop efficiency. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art, and proposes a shadow pattern acquisition triggering method and device based on driver behavior scoring.

[0005] In a first aspect, embodiments of the present invention provide a shadow pattern acquisition triggering method based on driver behavior scoring, comprising:

[0006] The cloud server receives driver behavior data and scores the driver behavior data to obtain a driver behavior score;

[0007] The driver behavior score is compared with a scoring threshold to determine whether the driver behavior has training value.

[0008] If it has training value, generate a shadow pattern acquisition configuration file that matches the target vehicle and send it to the target vehicle;

[0009] The target vehicle collects the driver's actual operating behavior, current perception data, and vehicle status in real time.

[0010] Based on the perceived data and vehicle status, driving behavior predictions are simulated under the same scenario.

[0011] If the driver's actual operating behavior is compared with the predicted driving behavior, and the deviation between the driver's actual operating behavior and the predicted driving behavior is greater than the comparison threshold, then the driver's actual operating behavior is sent back to the cloud server.

[0012] Optionally, before scoring the driver behavior data to obtain a driver behavior score, the method includes: constructing a driving behavior scoring model; the construction of the driving behavior scoring model includes:

[0013] Rating data is extracted from the original database of all drivers, the rating data is processed, and multiple rating dimension features are extracted.

[0014] The original values ​​of each of the scoring dimension features are standardized and normalized to obtain dimension feature values, and weights are assigned to each of the scoring dimension features to obtain feature weights.

[0015] A driving behavior scoring model is constructed based on the dimensional feature values ​​and the feature weights.

[0016] Optionally, scoring the driver behavior data to obtain a driver behavior score includes: inputting the driver behavior data into the driving behavior scoring model and outputting the driver behavior score.

[0017] Optionally, comparing the driver's actual operating behavior with the predicted driving behavior includes:

[0018] Set a comparison threshold and compare the driver's actual operating behavior with the predicted driving behavior frame by frame;

[0019] If the deviation between the driver's actual operating behavior and the predicted driving behavior is greater than the comparison threshold, then it is determined that the driver's actual operating behavior is significantly inconsistent with the predicted driving behavior.

[0020] Optionally, after the cloud server receives the driver behavior data, it further includes:

[0021] The driver behavior data, the driver's actual operating behavior, and the current perception data and vehicle status are written into the ring buffer of the target vehicle in real time.

[0022] If the driver's actual operating behavior is significantly inconsistent with the predicted driving behavior, the time point of the corresponding event is recorded, the information of the event is written into a file, and all relevant data in the circular buffer is saved to disk.

[0023] After the file is written to disk, an efficient compression algorithm is used to compress the file.

[0024] Optionally, after compressing the file using an efficient compression algorithm, the method further includes:

[0025] Add the compressed file to the upload task queue.

[0026] Each file in the upload task queue is sliced, and each sliced ​​file is uploaded to the cloud server sequentially according to the communication protocol;

[0027] Remove the uploaded slice from the upload task queue.

[0028] Optionally, comparing the driver's actual operating behavior with the predicted driving behavior includes: comparing at least one of the following: actual steering wheel angle with predicted steering wheel angle, actual throttle and brake depth with predicted control commands, actual lane change behavior with predicted lane change, and actual driving trajectory with predicted path.

[0029] Secondly, embodiments of the present invention provide a shadow pattern acquisition triggering device based on driver behavior scoring, comprising:

[0030] The cloud-based configuration management module is used to receive driver behavior data, score the driver behavior data to obtain a driver behavior score, and compare the driver behavior score with a scoring threshold to determine whether the driver behavior has training value.

[0031] The vehicle-side shadow mode trigger capture module is used to generate a shadow mode acquisition configuration file that matches the target vehicle and send it to the target vehicle; the target vehicle collects the driver's actual operation behavior, current perception data, and vehicle status in real time; based on the perception data and vehicle status, it simulates driving behavior prediction under the same scenario; it compares the driver's actual operation behavior with the driving behavior prediction, and if the deviation between the driver's actual operation behavior and the driving behavior prediction is greater than the comparison threshold, it sends the driver's actual operation behavior back to the cloud server.

[0032] Thirdly, the present invention provides an electronic device including a processor and a memory, the memory storing computer-readable instructions that, when executed by the processor, perform the steps of the method provided in the first aspect above.

[0033] Fourthly, the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the method provided in the first aspect above.

[0034] Fifthly, the present invention provides a computer program product comprising a computer program that, when executed by a processor, performs the steps of the method provided in the first aspect above.

[0035] As can be seen from the above technical solutions, the present invention has the following advantages:

[0036] The present invention provides a shadow pattern acquisition triggering method based on driver behavior scoring. This method receives driver behavior data from a cloud server and scores the data to obtain a driver behavior score. The driver behavior score is compared with a scoring threshold to determine if the driver behavior has training value. If it does, a shadow pattern acquisition configuration file matching the target vehicle is generated and sent to the target vehicle. The target vehicle collects the driver's actual operating behavior, current perception data, and vehicle status in real time. Based on the perception data and vehicle status, a driving behavior prediction under the same scenario is simulated. The actual driver operating behavior is compared with the driving behavior prediction. If the deviation between the actual driver operating behavior and the driving behavior prediction is greater than the comparison threshold, the actual driver operating behavior is transmitted back to the cloud server. By performing multi-dimensional modeling and scoring of the driver's historical driving behavior, high-quality samples with "exemplary" characteristics are automatically identified, solving the problems of traditional data acquisition methods that cannot distinguish between good and bad driving behavior and easily introduce low-quality samples. Simultaneously, by pre-screening driver behavior, the collection and uploading of low-quality or redundant data is significantly reduced, significantly saving communication bandwidth, storage space, and edge computing resources, and improving the operating efficiency and economy of the data closed-loop system. Attached Figure Description

[0037] Figure 1 This is a flowchart illustrating a first embodiment of the shadow pattern acquisition and triggering method based on driver behavior scoring according to the present invention.

[0038] Figure 2 This is a flowchart illustrating a second embodiment of the shadow pattern acquisition triggering method based on driver behavior scoring according to the present invention.

[0039] Figure 3 This is a flowchart illustrating a third embodiment of the shadow pattern acquisition triggering method based on driver behavior scoring according to the present invention.

[0040] Figure 4 This is a structural block diagram of an embodiment of a shadow pattern acquisition triggering device based on driver behavior scoring according to the present invention;

[0041] Figure 5 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0042] To enable those skilled in the art to better understand the technical solutions of the present invention, exemplary embodiments of the present invention are described below in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0043] Where there is no conflict, the various embodiments of the present invention and the features thereof may be combined with each other.

[0044] As used herein, the term “and / or” includes any and all combinations of one or more related enumerated entries.

[0045] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used herein, the singular forms “a” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that when the terms “comprising” and / or “made of” are used in this specification, the presence of the stated feature, integral, step, operation, element, and / or component is specified, but the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof is not excluded. Terms such as “connected” or “linked” are not limited to physical or mechanical connections but can include electrical connections, whether direct or indirect.

[0046] Unless otherwise specified, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having the meaning consistent with their meaning in the context of the relevant art and the invention, and will not be interpreted as having an idealized or overly formal meaning unless expressly so defined herein.

[0047] In the technical solution of this invention, the collection, storage, use, processing, transmission, provision, and disclosure of user personal information all comply with relevant laws and regulations and do not violate public order and good morals. The use of user data in this technical solution follows relevant national laws and regulations (e.g., the "Information Security Technology - Personal Information Security Specification"). For example: appropriate measures are taken for personal information access control; restrictions are imposed on the display of personal information; the purpose of using personal information does not exceed the scope of direct or reasonable association; and explicit identity targeting is eliminated when using personal information to avoid precisely locating a specific individual.

[0048] In the related art, (the related art is analyzed, and the technical problem solved by this application is pointed out)

[0049] To address at least one of the technical problems existing in the aforementioned related technologies, the present invention provides a shadow pattern acquisition triggering method based on driver behavior scoring. Figure 1 This is a flowchart illustrating a shadow pattern acquisition and triggering method based on driver behavior scoring according to the present invention; see reference. Figure 1 As shown, the method includes:

[0050] In step S101, the cloud server receives driver behavior data and scores the driver behavior data to obtain a driver behavior score.

[0051] In this embodiment of the application, the vehicle sends the collected driver behavior data to the cloud server. After receiving the driver behavior data, the cloud server does not store or analyze it, but first uses a driving behavior scoring model to score the driver behavior data and obtain a driver behavior score.

[0052] Step S102: Compare the driver behavior score with the scoring threshold to determine whether the driver behavior has training value.

[0053] In this embodiment of the application, the driver behavior score obtained in step S101 is compared with a preset scoring threshold to distinguish whether the driver is a good driver. If so, it indicates that the driver behavior has training value. Optionally, when the driver behavior score is higher than the scoring threshold, the driver behavior on the shore section has training value.

[0054] Step S103: If the data has training value, generate a shadow pattern acquisition configuration file that matches the target vehicle and send it to the target vehicle.

[0055] In this embodiment of the application, if the driver's behavior is determined to have training value, the cloud server will generate a shadow pattern collection configuration file that matches the target vehicle and send it to the target vehicle to start real-time and continuous collection of the actual operation behavior of the driver who has passed the score.

[0056] Step S104: The target vehicle collects the driver's actual operating behavior, current perception data, and vehicle status in real time.

[0057] In this embodiment of the application, the target vehicle not only collects the driver's actual operating behavior in real time, but also collects the vehicle's current perception data and vehicle status data, so as to carry out subsequent steps.

[0058] Step S105: Based on the perceived data and vehicle status, simulate and predict driving behavior under the same scenario.

[0059] In this embodiment of the application, the collected perception data and vehicle status are input into the intelligent driving algorithm model to simulate the driving behavior prediction that the system considers optimal under the same scenario.

[0060] Step S106: Compare the driver's actual operating behavior with the predicted driving behavior. If the deviation between the driver's actual operating behavior and the predicted driving behavior is greater than the comparison threshold, then the driver's actual operating behavior is sent back to the cloud server.

[0061] In this embodiment of the application, the driving behavior prediction obtained in step S105 and the actual driving behavior collected in step S104 are compared frame by frame to obtain the deviation between the actual driving behavior and the driving behavior prediction. The deviation is then compared with a preset comparison threshold to determine whether the actual driving behavior and the driving behavior prediction are inconsistent. If so, the actual driving behavior is sent back to the cloud server for model optimization.

[0062] The present invention provides a shadow pattern acquisition triggering method based on driver behavior scoring. This method receives driver behavior data from a cloud server and scores the data to obtain a driver behavior score. The driver behavior score is compared with a scoring threshold to determine if the driver behavior has training value. If it does, a shadow pattern acquisition configuration file matching the target vehicle is generated and sent to the target vehicle. The target vehicle collects the driver's actual operating behavior, current perception data, and vehicle status in real time. Based on the perception data and vehicle status, a driving behavior prediction under the same scenario is simulated. The actual driver operating behavior is compared with the driving behavior prediction. If the deviation between the actual driver operating behavior and the driving behavior prediction is greater than the comparison threshold, the actual driver operating behavior is transmitted back to the cloud server. By performing multi-dimensional modeling and scoring of the driver's historical driving behavior, high-quality samples with "exemplary" characteristics are automatically identified, solving the problems of traditional data acquisition methods that cannot distinguish between good and bad driving behavior and easily introduce low-quality samples. Simultaneously, by pre-screening driver behavior, the collection and uploading of low-quality or redundant data is significantly reduced, significantly saving communication bandwidth, storage space, and edge computing resources, and improving the operating efficiency and economy of the data closed-loop system.

[0063] This invention also provides a method for triggering shadow pattern acquisition based on driver behavior scoring. Please refer to [link to relevant documentation]. Figure 2 , Figure 2 This is a flowchart illustrating the steps of a shadow pattern acquisition and triggering method based on driver behavior scoring according to the present invention; the steps include:

[0064] Step S201: Construct a driving behavior scoring model.

[0065] In the embodiments of this application, please refer to Figure 3 , Figure 2 This is a flowchart illustrating a method for constructing a driving behavior scoring model according to the present invention; the method includes:

[0066] Step S301: Extract rating data from the original database of all drivers, process the rating data, and extract multiple rating dimension features.

[0067] In a specific implementation, the processing of the scoring data may optionally include data cleaning, normalization, and statistical processing. The processed data can be used to extract five categories of scores as degree features as shown in the table below.

[0068]

[0069] Step S302: Standardize and normalize the original values ​​of each of the scoring dimension features to obtain dimension feature values, and assign weights to each of the scoring dimension features to obtain feature weights.

[0070] In the specific implementation, in order to unify the differences in the units of measurement among different features, the following method is used to construct the score:

[0071] 1. Standardized processing

[0072] The original values ​​of each feature are Z-score normalized to ensure they fall within the [0,1] interval. The formula for calculating the dimensional feature values ​​is as follows:

[0073] Z i =(x i -μ i ) / σ i；

[0074] Among them, Z i x is the standardized value of the i-th feature (i.e., the Z-score); i The original data value; μ i σ is the mean (average value) of the dataset for the i-th feature; i Let be the standard deviation of the dataset for the i-th feature.

[0075] 2. Weighting method

[0076] Based on expert experience, traffic behavior research experts assign weights w to each feature. i ;

[0077] Step S303: Construct a driving behavior scoring model based on the dimensional feature values ​​and the feature weights.

[0078] The final driver behavior scoring model is constructed using the aforementioned dimensional feature values ​​and feature weights as follows:

[0079]

[0080] Where score is the driver's behavior score, w i Z is the weight of the i-th feature. i Let be the value of the i-th feature.

[0081] In step S202, the cloud server receives driver behavior data and scores the driver behavior data to obtain a driver behavior score.

[0082] In practice, driver behavior data is input into the aforementioned driver behavior scoring model, which then outputs a driver behavior score.

[0083] Step S203: Compare the driver behavior score with the scoring threshold to determine whether the driver behavior has training value.

[0084] In its implementation, the system presets a scoring threshold to distinguish between "good driving behavior" and "low-quality driving behavior." When a driver's behavior score is higher than this threshold, the driver is judged as a "good driving behavior" driver, whose driving behavior is referential and learnable; conversely, if the score is lower, the driver is considered a "low-quality driving behavior" driver and has no training value. Optionally, the system presets a scoring threshold of 0.8.

[0085] Step S204: If the model has training value, generate a shadow pattern acquisition configuration file that matches the target vehicle and send it to the target vehicle.

[0086] In this embodiment, for drivers who meet the scoring criteria, the cloud system automatically generates a shadow mode data collection configuration file that matches their vehicle. Optionally, the configuration content includes: data collection content (such as images, radar, etc.), data collection time window settings (retaining 5 seconds of data before and after triggering), etc.

[0087] In the specific implementation, the configuration file is encapsulated in a structured format (JSON) and encrypted before being sent to the target vehicle via a standard communication protocol (MQTT). During transmission, the system uses an authentication mechanism (such as PKI public key infrastructure) to ensure the security and legitimacy of the communication link.

[0088] If a driver's behavior score is below a set threshold, the system will not send any shadow mode data collection configuration file to the corresponding vehicle, and the vehicle will not initiate any related data collection logic. This effectively prevents low-quality driving behavior samples from entering the data loop system, improving the overall effectiveness of data collection and resource utilization efficiency.

[0089] Step S205: The target vehicle collects the driver's actual operating behavior, current perception data, and vehicle status in real time.

[0090] In this embodiment, after the vehicle is powered on, the vehicle-side system receives and parses the configuration file sent from the cloud and deploys it to the vehicle. At this time, a shadow mode is activated to determine in real time whether data collection is triggered; the intelligent driving algorithm model (perception fusion + path planning + control decision model) is initialized and runs silently in the background without participating in actual control; optionally, the actual driver operation behavior is collected, including: steering wheel angle, accelerator / brake pedal opening, gear changes, turn signal usage status, etc.; vehicle status includes vehicle speed, acceleration, lateral / longitudinal G-value, lane position, etc.; perception data includes obstacle recognition results, traffic sign recognition, road line detection, vehicle positioning information, etc. In addition, the timestamp at that time is recorded simultaneously for subsequent behavior comparison and data alignment.

[0091] Step S206: Based on the perceived data and vehicle status, simulate and predict driving behavior under the same scenario.

[0092] In this embodiment, the shadow mode invokes an intelligent driving algorithm model to simulate the "optimal" driving behavior prediction for the scenario, based on the same input perception data and vehicle state. Optionally, the driving behavior prediction includes: expected steering wheel angle, expected acceleration / deceleration strategy, expected lane change or overtaking timing, expected following distance, and response actions, etc.

[0093] Step S207: Cache the vehicle-side data.

[0094] In this embodiment, all data that needs to be collected, such as driver behavior data, driver's actual operating behavior, current perception data, and vehicle status, are written into the ring buffer of the target vehicle in real time to ensure that data fragments before and after the trigger are not lost.

[0095] Step S208: Compare the driver's actual operating behavior with the predicted driving behavior. If the deviation between the driver's actual operating behavior and the predicted driving behavior is greater than the comparison threshold, then the driver's actual operating behavior is sent back to the cloud server.

[0096] In this embodiment, the driver's actual operating behavior is compared frame-by-frame with the driving behavior prediction from the shadow mode. Optionally, the comparison includes: actual steering wheel angle vs. predicted steering wheel angle, actual accelerator / brake depth vs. predicted control command, actual lane change behavior vs. predicted lane change, actual driving trajectory vs. predicted path, etc., and comparison thresholds are set (e.g., steering wheel angle difference > 5°, accelerator depth difference > 20%, brake depth difference > 20%, expected path deviation > 0.5m, etc.). It is determined whether the driver's actual operating behavior and the driving behavior prediction constitute a "significant inconsistency." If the deviation between the driver's actual operating behavior and the driving behavior prediction is greater than the comparison threshold, then the driver's actual operating behavior is determined to be significantly inconsistent with the driving behavior prediction. If there is a significant inconsistency, data feedback is triggered.

[0097] If the driver's actual driving behavior is significantly inconsistent with the predicted driving behavior, it is considered a potentially valuable learning event. The system immediately records the time of the corresponding event, writes the event information (trigger reason, timestamp, vehicle ID, etc.) into a file, and saves all relevant data in the circular buffer to disk. After disk saving is complete, an efficient compression algorithm (e.g., ZIP, H.264 video encoding) is used to compress the file to reduce the upload size.

[0098] After compression is complete, the system adds the compressed file to the upload task queue. The vehicle-side uses a Public Key Infrastructure (PKI) authentication mechanism to establish a secure connection with the cloud. Each file in the upload task queue is sliced, ensuring that each data packet contains a unique identifier (UUID), a serial number, and a checksum (SHA-256). Each sliced ​​file is then uploaded to the cloud server sequentially according to the HTTP communication protocol.

[0099] Step S209: After successful upload, remove the cached data from the vehicle.

[0100] In this embodiment of the application, the uploaded slice is removed from the upload task queue to prevent duplicate uploads.

[0101] Compared to the commonly used "event-triggered" data acquisition methods in existing technologies, this invention proposes a more intelligent shadow mode triggering mechanism. By introducing a driver behavior scoring model, it performs real-time evaluation of the driver in the cloud and dynamically controls the start and stop of the vehicle-side shadow mode system based on the scoring results, thereby achieving the goal of "accurate data acquisition and efficient closed-loop processing." This mechanism not only improves the effectiveness of data acquisition but also significantly reduces the cost of data transmission and processing, demonstrating greater technological advancement and commercial application prospects.

[0102] This invention also provides a shadow pattern acquisition triggering device based on driver behavior scoring. Please refer to [link to relevant documentation]. Figure 4 , Figure 4 This is a structural block diagram of a shadow pattern acquisition triggering device based on driver behavior scoring according to the present invention. The device includes:

[0103] The cloud-based configuration management module 401 is used to receive driver behavior data, score the driver behavior data to obtain a driver behavior score, and compare the driver behavior score with a scoring threshold to determine whether the driver behavior has training value.

[0104] The vehicle-side shadow mode triggering and capture module 402 is used to generate a shadow mode acquisition configuration file that matches the target vehicle and send it to the target vehicle; the target vehicle collects the driver's actual operation behavior, current perception data, and vehicle status in real time; based on the perception data and vehicle status, it simulates driving behavior prediction under the same scenario; it compares the driver's actual operation behavior with the driving behavior prediction, and if the deviation between the driver's actual operation behavior and the driving behavior prediction is greater than the comparison threshold, it sends the driver's actual operation behavior back to the cloud server.

[0105] In this embodiment, the cloud configuration management module 401 includes:

[0106] The data receiving module is used to receive driver behavior data.

[0107] The driver behavior scoring module is used to score the driver behavior data to obtain a driver behavior score.

[0108] The behavior score comparison module is used to compare the driver's behavior score with a scoring threshold to determine whether the driver's behavior has training value.

[0109] In this embodiment of the application, the vehicle-side shadow mode trigger capture module 402 includes:

[0110] The configuration file distribution module is used to generate a shadow mode acquisition configuration file that matches the target vehicle and distribute it to the target vehicle.

[0111] The data acquisition module is used to collect the driver's actual operating behavior, current perception data, and vehicle status in real time.

[0112] The driving behavior prediction module is used to simulate and predict driving behavior under the same scenario based on the perception data and vehicle status.

[0113] The prediction and comparison module is used to compare the driver's actual operating behavior with the predicted driving behavior. If the deviation between the driver's actual operating behavior and the predicted driving behavior is greater than a comparison threshold, the driver's actual operating behavior is sent back to the cloud server. Specifically, a comparison threshold is set, and the driver's actual operating behavior and the predicted driving behavior are compared frame by frame. If the deviation between the driver's actual operating behavior and the predicted driving behavior is greater than the comparison threshold, it is determined that the driver's actual operating behavior and the predicted driving behavior are significantly inconsistent.

[0114] In this embodiment of the application, a driving behavior scoring model construction module is also included, which is used to extract scoring data from the original database of all drivers, process the scoring data, and extract multiple scoring dimension features; standardize and normalize the original values ​​of each scoring dimension feature to obtain dimension feature values, assign weights to each scoring dimension feature to obtain feature weights; and construct a driving behavior scoring model based on the dimension feature values ​​and the feature weights.

[0115] In this embodiment, a vehicle-side data caching module 403 is also included. This module writes all the data to be collected, including driver behavior data, actual driver operation behavior, current perception data, and vehicle status, into the circular buffer of the target vehicle in real time, ensuring that data segments before and after the trigger are not lost. If the actual driver operation behavior is significantly inconsistent with the predicted driving behavior, it is considered a potentially valuable learning event. The system immediately records the time point of the corresponding event, writes the event information (trigger reason, timestamp, vehicle ID, etc.) into a file, and saves all relevant data in the circular buffer to disk. After disk storage is completed, an efficient compression algorithm (e.g., ZIP, H.264 video encoding) is used to compress the file to reduce the upload size.

[0116] In this embodiment, a vehicle-side data encryption and upload module 404 is also included. After compression is completed, the system adds the compressed file to the upload task queue, and the vehicle-side uses a Public Key Infrastructure (PKI) authentication mechanism to establish a secure connection with the cloud. Each file in the upload task queue is sliced, so that each data packet contains: a unique identifier (UUID), a serial number, and a checksum (SHA-256); and each sliced ​​file is uploaded to the cloud server sequentially according to the communication protocol (HTTP); after successful upload, the slice is removed from the upload task queue to prevent duplicate uploads.

[0117] Based on the same inventive concept, embodiments of the present invention also provide an electronic device. Figure 5 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Figure 5As shown, an embodiment of the present invention provides an electronic device including: one or more processors 101, a memory 102, and one or more I / O interfaces 103. The memory 102 stores one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement any of the shadow pattern acquisition triggering methods based on driver behavior scoring in the above embodiments; the one or more I / O interfaces 103 are connected between the processor and the memory, configured to enable information interaction between the processor and the memory.

[0118] The processor 101 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 102 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read / write interface) 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, including but not limited to a data bus (Bus).

[0119] In some embodiments, the processor 101, memory 102, and I / O interface 103 are interconnected via bus 104, and thus connected to other components of the computing device.

[0120] In some embodiments, the one or more processors 101 include a field-programmable gate array.

[0121] This invention also provides a computer-readable medium. The computer-readable medium stores a computer program, which, when executed by a processor, implements the steps of any of the shadow pattern acquisition triggering methods based on driver behavior scoring described above. The computer-readable storage medium can be volatile or non-volatile.

[0122] This invention also provides a computer program product, including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code. When the computer-readable code is run in the processor of an electronic device, the processor in the electronic device executes the above-described shadow pattern acquisition triggering method based on driver behavior scoring.

[0123] Those skilled in the art will understand that all or some of the steps, systems, and apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software can be distributed on a computer-readable storage medium, which may include computer storage media (or non-transitory media) and communication media (or transient media).

[0124] As is known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable program instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technologies, portable compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, it is known to those skilled in the art that communication media typically contain computer-readable program instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0125] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0126] The computer program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of the invention.

[0127] The computer program product described herein can be implemented specifically through hardware, software, or a combination thereof. In one alternative embodiment, the computer program product is specifically embodied in a computer storage medium; in another alternative embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.

[0128] Various aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0129] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0130] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0131] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0132] Example embodiments have been disclosed herein, and while specific terminology has been used, it is for illustrative purposes only and should be construed as such, and is not intended to be limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of the invention as set forth in the appended claims.

Claims

1. A shadow pattern acquisition triggering method based on driver behavior scoring, characterized in that, include: The cloud server receives driver behavior data and scores the driver behavior data to obtain a driver behavior score; The driver behavior score is compared with a scoring threshold to determine whether the driver behavior has training value. If it has training value, generate a shadow pattern acquisition configuration file that matches the target vehicle and send it to the target vehicle; The target vehicle collects the driver's actual operating behavior, current perception data, and vehicle status in real time. Based on the perceived data and vehicle status, driving behavior predictions are simulated under the same scenario. If the driver's actual operating behavior is compared with the predicted driving behavior, and the deviation between the driver's actual operating behavior and the predicted driving behavior is greater than the comparison threshold, then the driver's actual operating behavior is sent back to the cloud server.

2. The shadow pattern acquisition triggering method based on driver behavior scoring according to claim 1, characterized in that, Before scoring the driver behavior data to obtain a driver behavior score, the process includes: constructing a driving behavior scoring model; the construction of the driving behavior scoring model includes: Rating data is extracted from the original database of all drivers, the rating data is processed, and multiple rating dimension features are extracted. The original values ​​of each of the scoring dimension features are standardized and normalized to obtain dimension feature values, and weights are assigned to each of the scoring dimension features to obtain feature weights. A driving behavior scoring model is constructed based on the dimensional feature values ​​and the feature weights.

3. The shadow pattern acquisition triggering method based on driver behavior scoring according to claim 2, characterized in that, The step of scoring the driver behavior data to obtain a driver behavior score includes: inputting the driver behavior data into the driving behavior scoring model and outputting the driver behavior score.

4. The shadow pattern acquisition triggering method based on driver behavior scoring according to claim 1, characterized in that, The comparison between the driver's actual operating behavior and the predicted driving behavior includes: Set a comparison threshold and compare the driver's actual operating behavior with the predicted driving behavior frame by frame; If the deviation between the driver's actual operating behavior and the predicted driving behavior is greater than the comparison threshold, then it is determined that the driver's actual operating behavior is significantly inconsistent with the predicted driving behavior.

5. The shadow pattern acquisition triggering method based on driver behavior scoring according to claim 4, characterized in that, After the cloud server receives the driver behavior data, it also includes: The driver behavior data, the driver's actual operating behavior, and the current perception data and vehicle status are written into the ring buffer of the target vehicle in real time. If the driver's actual operating behavior is significantly inconsistent with the predicted driving behavior, the time point of the corresponding event is recorded, the information of the event is written into a file, and all relevant data in the circular buffer is saved to disk. After the file is written to disk, a compression algorithm is used to compress the file.

6. The shadow pattern acquisition triggering method based on driver behavior scoring according to claim 5, characterized in that, After compressing the file using an efficient compression algorithm, the process also includes: Add the compressed file to the upload task queue. Each file in the upload task queue is sliced, and each sliced ​​file is uploaded to the cloud server sequentially according to the communication protocol; Remove the uploaded slice from the upload task queue.

7. The shadow pattern acquisition triggering method based on driver behavior scoring according to claim 1, characterized in that, The comparison between the driver's actual operating behavior and the driving behavior prediction includes: comparing at least one of the following: actual steering wheel angle with predicted steering wheel angle, actual throttle and brake depth with predicted control commands, actual lane change behavior with predicted lane change, and actual driving trajectory with predicted path.

8. A shadow pattern acquisition triggering device based on driver behavior scoring, characterized in that, include: The cloud-based configuration management module is used to receive driver behavior data and score the driver behavior data to obtain a driver behavior score. The driver behavior score is compared with a scoring threshold to determine whether the driver behavior has training value. The vehicle-side shadow mode trigger capture module is used to generate a shadow mode acquisition configuration file that matches the target vehicle and send it to the target vehicle. The target vehicle collects the driver's actual operating behavior, current perception data, and vehicle status in real time. Based on the perceived data and vehicle status, driving behavior predictions are simulated under the same scenario. If the driver's actual operating behavior is compared with the predicted driving behavior, and the deviation between the driver's actual operating behavior and the predicted driving behavior is greater than the comparison threshold, then the driver's actual operating behavior is sent back to the cloud server.

9. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1 to 7.

10. A computer-readable medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images