A data evaluation and transmission method and system based on a vehicle-road cloud cooperation environment

Abstract

This invention relates to the field of intelligent transportation technology and discloses a data evaluation and transmission method and system based on a vehicle-road-cloud collaborative environment. The data evaluation and transmission method includes: responding to receiving a perception task published by a service requesting entity, sending a data perception request to a target vehicle based on the road information required by the perception task; responding to receiving perception data returned by the target vehicle within a preset window period, performing authenticity verification and value evaluation on the perception data according to the task type of the perception task to obtain the value of the perception data; and determining the target perception data from all perception data based on the perceived data value and sending the target perception data to the service requesting entity. Through the above scheme, this invention can flexibly perform authenticity verification and value evaluation on perception data according to the task type of the perception task, ensuring the accuracy and reliability of the selected perception data, reducing data transmission latency, and improving the real-time performance of data transmission.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation technology, and in particular to a data evaluation and transmission method and system based on a vehicle-road-cloud collaborative environment. Background Technology

[0002] With the development of intelligent connected vehicles and the smart transportation industry, vehicle-road-cloud integrated collaborative technology has gradually emerged. Through communication technologies such as V2X and 5G, this technology enables full-domain interconnection and collaborative control of intelligent connected vehicles, roadside sensing devices, and cloud-based management platforms. It can effectively fill the blind spots in single-vehicle perception, achieve low-latency collaborative decision-making, and optimize overall traffic flow.

[0003] In the current vehicle-road-cloud integrated collaborative technology, roadside units (RSUs) mainly collect road information (such as traffic light status, blind spot obstacles, etc.) and upload it to the cloud platform for use by connected vehicles or third-party platforms (such as traffic management platforms) in order to build a global traffic network, promote traffic management, and optimize vehicle driving decisions.

[0004] In the aforementioned vehicle-road-cloud collaborative environment, vehicles primarily act as data consumers, acquiring the necessary roadside data through a paid subscription model. The data they collect is not proactively provided to the cloud platform for use by other vehicles. Furthermore, due to the difficulty in achieving full coverage of roadside equipment, issues such as missing data in some road sections and obstructed views arise, which are detrimental to the construction of the overall traffic network and the collaborative optimization of vehicle driving decisions.

[0005] To encourage vehicles to upload their real-time traffic data to the cloud for sharing and to address the aforementioned issues, vehicle perception data trading technology based on the vehicle-road-cloud collaborative environment has emerged. In the current vehicle-road-cloud collaborative trading environment, the transmission of vehicle perception data is usually achieved based on blockchain technology. Before data trading, the value of the data is usually evaluated through complex game theory auction mechanisms or black-box deep learning mechanisms. The evaluation process is complex and computationally intensive, resulting in a large delay and poor real-time performance in the entire process of transmitting the evaluated data to the data consumer. Summary of the Invention

[0006] In view of this, the present invention proposes a data evaluation and transmission method and system based on a vehicle-road-cloud collaborative environment, which can solve the problems of large latency and poor real-time performance in the traditional data evaluation and transmission process based on a vehicle-road-cloud collaborative environment.

[0007] On one hand, embodiments of the present invention provide a data evaluation and transmission method based on a vehicle-road-cloud cooperative environment, which includes: In response to receiving a perception task published by the service requesting entity, a data perception request is sent to all target vehicles based on the road information that the perception task needs to perceive. In response to receiving the perception data returned by the target vehicle within a preset window period, the perception data is verified for authenticity and evaluated for value according to the task type of the perception task, so as to obtain the value of the perception data. Based on the perceived data value, target perceived data is determined from all perceived data, and the target perceived data is sent to the service request subject.

[0008] In some implementations, the perceived data is verified for authenticity and its value is assessed based on the task type of the perception task, and the value of the perceived data includes: In response to the fact that the task type of the sensing task is a continuous sensing task, the time delay factor of the sensing data is determined based on the time of receiving the sensing data and the time of collecting the sensing data; The timeliness of the perceived data is determined based on the time delay factor. In response to the timeliness determination being passed, the authenticity of the perceived data is verified. In response to the successful verification of authenticity, the perceived data is evaluated for value based on the time delay factor, combined with the quality and scarcity of the perceived data, to obtain the value of the perceived data.

[0009] In some implementations, the perceived data is valued based on the time delay factor, combined with the quality and scarcity of the perceived data, to obtain the value of the perceived data, including: The quality score of the sensed data is determined based on the hardware configuration information of the data acquisition device. Based on the scarcity of the perceived data, a scarcity score for the perceived data is determined; The value of the perceived data is determined based on the quality score, the scarcity score, and the time delay factor.

[0010] In some implementations, determining the value of the sensed data based on its quality score, scarcity score, and latency factor includes: Determine the quality coefficient based on the data perception request corresponding to the perceived data; Based on the perception scenario corresponding to the perception data, determine the scarcity coefficient; The value of the perceived data is obtained by weighting and summing the quality score and the quality coefficient, the scarcity score and the scarcity coefficient, and then multiplying the sum by the time delay factor.

[0011] In some implementations, determining the quality coefficient based on the data sensing request corresponding to the sensed data includes: Based on the data perception request corresponding to the perceived data, determine the comprehensive score of the data perception request; The quality coefficient is determined based on the comprehensive score of the data perception request, the pre-set theoretical score value, and the quality coefficient limit value.

[0012] In some implementations, the quality coefficient is determined based on the overall score of the data perception request, a pre-set theoretical score value, and a quality coefficient limit value, including: Based on the dynamic adjustment model of the quality coefficient, and combining the comprehensive score of the data perception request, the pre-set theoretical score value, and the quality coefficient limit value, the quality coefficient is determined, wherein the dynamic adjustment model of the quality coefficient is:

[0013] in, Indicates the quality coefficient. This represents the overall score of the data-aware request. This represents the minimum theoretical score. This represents the theoretical maximum value of the scoring system. This represents the lower limit of the quality coefficient limit. This indicates the upper limit of the quality coefficient limit.

[0014] In some implementations, the perceived data is valued based on the task type of the perceived task, and the value of the perceived data includes: In response to the fact that the task type of the perception task is an isolated perception task, the authenticity of the perception data is verified. In response to the successful verification of authenticity, the perceived data is evaluated for value based on its quality, scarcity, and latency to obtain the value of the perceived data.

[0015] In some implementations, the perceived data is valued based on its quality, scarcity, and latency, and the value of the perceived data includes: The quality score of the sensed data is determined based on the hardware configuration information of the data acquisition device. Based on the scarcity of the perceived data, a scarcity score for the perceived data is determined; The time delay factor of the sensing data is determined based on the time of receipt of the sensing data and the time of acquisition of the sensing data; The value of the perceived data is determined based on the quality score, the scarcity score, and the time delay factor.

[0016] In some implementations, the verification of the authenticity of the sensed data includes: Perform static environmental feature consistency verification on the perceived data; In response to the successful verification of static environmental feature consistency, a dynamic target motion rationality verification is performed on the perceived data; In response to the successful verification of the rationality of the dynamic target motion, a multi-source field-of-view consistency verification is performed on the perceived data.

[0017] On the other hand, the present invention also provides a data evaluation and transmission system based on a vehicle-road-cloud cooperative environment, which is configured to perform the steps of the data evaluation and transmission method based on a vehicle-road-cloud cooperative environment as described in the above embodiments.

[0018] The present invention has at least the following beneficial effects: This invention provides a data evaluation and transmission method based on a vehicle-road-cloud collaborative environment. The technical solution of this application sends data perception requests to all target vehicles based on the road information required by the perception task. Within a preset window period, the perceived data returned by the target vehicles is verified for authenticity and its value is evaluated according to the task type. Based on the value of the perceived data, target perceived data is selected from all perceived data and sent to the service requesting entity. This method flexibly verifies and evaluates the perceived data according to the task type, ensuring the accuracy and reliability of the selected perceived data, reducing data transmission latency, and improving data transmission real-time performance. For isolated perception tasks with low latency requirements, all perceived data received within the preset window period can be used for authenticity verification and value evaluation to select accurate and reliable perceived data. For perception tasks with high latency requirements, a latency factor is calculated before data verification to eliminate some high-latency data, thereby reducing the amount of data involved in the authenticity verification and value evaluation process. This reduces data transmission latency, improves data transmission real-time performance, and ensures the accuracy and reliability of the selected perceived data. By verifying the authenticity of data before data value assessment, data that fails the verification is eliminated, thereby improving the reliability of data transmission. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained based on these drawings without creative effort.

[0020] Figure 1A flowchart illustrating a data evaluation and transmission method based on a vehicle-road-cloud collaborative environment, provided as an embodiment of the present invention; Figure 2 A flowchart illustrating the authenticity verification of sensing data in a data evaluation and transmission method based on a vehicle-road-cloud collaborative environment provided in an embodiment of the present invention. Figure 3 The flowchart illustrates the process of verifying the authenticity and evaluating the value of perceived data in the data evaluation and transmission method based on a vehicle-road-cloud collaborative environment provided in this embodiment of the invention. Figure 4 A flowchart illustrating the value assessment of perceived data in the data assessment and transmission method based on a vehicle-road-cloud collaborative environment provided in an embodiment of the present invention; Figure 5 A schematic diagram of the structure of a data evaluation and transmission system based on a vehicle-road-cloud collaborative environment provided in an embodiment of the present invention; Figure 6 A flowchart of a data evaluation and transmission system based on a vehicle-road-cloud collaborative environment is provided for an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to specific examples and the accompanying drawings.

[0022] It should be noted that all uses of "first" and "second" in the embodiments of the present invention are for the purpose of distinguishing two entities or parameters with the same name but different names. It is clear that "first" and "second" are only for the convenience of expression and should not be construed as limiting the embodiments of the present invention. Subsequent embodiments will not explain this in detail.

[0023] The present invention will now be described in detail with reference to the embodiments and accompanying drawings.

[0024] The first aspect of this invention provides a data evaluation and transmission method based on a vehicle-road-cloud cooperative environment, such as... Figure 1 As shown, the data evaluation and transmission method based on the vehicle-road-cloud collaborative environment may include steps S100 to S120.

[0025] S100: In response to receiving the perception task published by the service requesting subject, send a data perception request to all target vehicles according to the road information that the perception task needs to perceive.

[0026] In this embodiment of the invention, steps S100-S120 can be executed by a cloud platform. The cloud platform can be an electronic device such as a desktop computer, laptop computer, or server. The service requesting entity can be a vehicle, the cloud platform itself, or a third-party platform (e.g., a traffic management platform) or other entities with sensing needs.

[0027] In this embodiment of the invention, a vehicle database can be pre-established. Vehicles register with the cloud platform based on their basic information, such as license plate number, vehicle model, color, and hardware configuration. After successful registration, the cloud platform assigns a unique identifier to each vehicle for use in subsequent data transmission. The cloud platform also stores the vehicle's basic information and unique identifier in the vehicle database. All vehicles in the database can issue and respond to perception tasks.

[0028] A perception task is issued by the service requesting entity based on its own needs and does not involve specific data details. A data perception request, on the other hand, is directed to the perception entity and is used to convert the perception task into a specific data-level request (i.e., a data perception request). The data perception request includes information such as the perception task requirements and the location range to be perceived. It is used to guide the target vehicle in data collection and transmission.

[0029] The road information that needs to be perceived includes the location range of the road, the target vehicle is a vehicle located in the vehicle database, and the real-time location or driving path of the target vehicle overlaps with the location range of the road.

[0030] In this embodiment of the invention, the cloud platform can receive perception tasks sent by vehicles or third-party platforms, determine the target vehicle based on the road information that the perception task needs to perceive, and send data perception requests to all target vehicles.

[0031] For example, if a vehicle wants to obtain information about events affecting its driving path (such as congestion, accidents, or construction) to optimize its route, it sends a corresponding perception task to the cloud platform. Upon receiving the perception task, the cloud platform generates a corresponding data perception request based on the task and sends it to vehicles and / or roadside equipment traveling along the vehicle's path to obtain event-level perception data (e.g., event type (congestion, accident, or construction) and event location). As another example, if a vehicle wants to obtain perception data about its blind spots, it sends a corresponding perception task to the cloud platform. Upon receiving the task, the cloud platform generates a corresponding data perception request based on the task and sends it to vehicles and / or roadside equipment whose visual range covers the blind spot to obtain raw / semi-raw data about the blind spot (e.g., point cloud fragments, ROI region images). For example, if a vehicle lacking high-precision sensing capabilities wants to obtain information about surrounding obstacles (e.g., the location and speed of nearby vehicles and pedestrians), it sends a corresponding high-precision sensing task to the cloud platform. Upon receiving the task, the cloud platform generates a corresponding data sensing request and sends it to vehicles and / or roadside equipment with high-precision sensing capabilities to obtain the obstacle information for the vehicle lacking high-precision sensing capabilities. Similarly, if a traffic management platform wants to obtain sensing data from a traffic accident scene, it sends a corresponding sensing task to the cloud platform. Upon receiving the task, the cloud platform generates a corresponding data sensing request and sends it to vehicles located near the accident scene to obtain the accident scene sensing data.

[0032] In this embodiment of the invention, the cloud platform can perform integrity identification on the traffic road data sent by the roadside equipment. If it is identified that the data of a certain road segment is missing or a certain area is obscured, it is determined to be incomplete. Then, a perception task is issued for the missing / obscured road segment, and based on the perception task, a corresponding data perception task is generated and sent to the perception subject.

[0033] For example, if the cloud platform identifies a blind spot in a road area (e.g., an intersection) based on traffic data uploaded by a roadside device, it generates a perception task, converts the perception task into a data perception request, and sends the data perception request to vehicles located near the road area.

[0034] S110. In response to receiving the perception data returned by the target vehicle within a preset window period, the perception data is verified for authenticity and evaluated for value according to the task type of the perception task, and the value of the perception data is obtained.

[0035] The preset window period is a pre-defined time period, such as 1 minute, 5 minutes, 10 minutes, or 20 minutes. Sensing data received within this preset window period is considered valid and can be verified for authenticity and evaluated for value. Sensing data received outside the preset window period is discarded to reduce subsequent computation, decrease data transmission latency, and improve the real-time performance of data transmission.

[0036] In this embodiment, after receiving a sensing task, the task type of the sensing task can be determined first, and then the authenticity verification and value assessment of the sensing data can be performed based on the determined task type.

[0037] Perception tasks can be categorized into isolated perception tasks and continuous perception tasks. Isolated perception tasks do not have continuous frame constraints or require sub-second return times. Data transmission latency is not as high, and the requesting entity only needs to obtain accurate and reliable perception data within a reasonable window. Examples include temporary road condition retrieval by third-party traffic management platforms, random checks of blind spot data in urban construction zones, and data collection for rare scenarios during non-emergency periods. Continuous perception tasks require a continuous return of perception data in consecutive frames. For example, a low-spec or limited-view passenger vehicle navigating from point A to point B continuously requires perception data for "virtual blind spot correction / advanced driver assistance" during the journey. Sub-second return times are required at every moment along the A→B path. Once data times out, it not only loses its usability but may also pose safety risks; therefore, extremely low data transmission latency is required.

[0038] Authenticity verification verifies whether data is forged or false, thereby improving data reliability. Value assessment calculates the value of data through multiple dimensions, providing a basis for rational data transactions.

[0039] In this embodiment of the invention, if the sensing task is an isolated sensing task, then all received sensing data are directly subjected to authenticity verification and value assessment. If the sensing task is a continuous sensing task, then the latency factor of the sensing data returned by the sensing task is first calculated. If the latency factor is less than a threshold, authenticity verification and value assessment are performed. If the latency factor is not less than the threshold, the data is directly deleted without further authenticity verification and value assessment. This not only removes some worthless data but also reduces the amount of data to be subsequently verified and assessed, thereby further reducing data transmission latency and improving the real-time performance of data transmission.

[0040] S120. Based on the perceived data value, determine the target perceived data from all perceived data and send the target perceived data to the service request subject.

[0041] In this embodiment, target perceived data can be selected from all perceived data that has passed authenticity verification based on the perceived data value (e.g., the highest data value), and then sent to the service requesting entity for use. Transactions are then conducted with both the perceiving entity and the service requesting entity.

[0042] In this embodiment of the invention, the value of the target perception data can be used to determine the fees to be deducted and the rewards to be issued, thereby deducting fees from the service requesting entity and issuing rewards to the target vehicle that returned the target perception data. Both the fees and rewards can be in the form of points or currency, and the specific form can be determined based on the actual application scenario.

[0043] In this embodiment of the invention, fees can be selectively deducted from the requesting entity based on its type. For example, if the requesting entity is a vehicle, fees are deducted from the vehicle; if the requesting entity is a cloud platform, fees are deducted from the cloud platform's data transaction fund pool.

[0044] In the technical solution of this invention, data perception requests are sent to all target vehicles based on the road information required for the perception task. The perceived data received from the target vehicles within a preset window period is then verified for authenticity and evaluated for value according to the task type. Based on the perceived data value, target perceived data is selected from all perceived data and sent to the service requesting entity. This allows for flexible verification and evaluation of perceived data based on the task type, ensuring the accuracy and reliability of the selected data, reducing data transmission latency, and improving real-time data transmission. For isolated perception tasks with low latency requirements, all perceived data received within the preset window period can be used for verification and evaluation to select accurate and reliable data. For perception tasks with high latency requirements, a latency factor is calculated before data verification to eliminate some high-latency data, thereby reducing the amount of data involved in the verification and evaluation process. This reduces data transmission latency, improves real-time data transmission, and ensures the accuracy and reliability of the selected perceived data.

[0045] In some embodiments, such as Figure 2 As shown, the authenticity of the perceived data can be verified through S200~S220.

[0046] S200. Perform static environmental feature consistency verification on the perceived data.

[0047] In this embodiment of the invention, the static environmental feature consistency verification is performed on the sensing data. If the verification passes, step S210 is executed; if the verification fails, the sensing data is intercepted.

[0048] In this embodiment of the invention, when performing static environmental feature consistency verification on the perceived data, the environment matching error can be determined based on the location information of the static environmental features in the perceived data and the location information of the static environmental features in the map; the static environmental feature consistency verification result can be determined based on the environment matching error and the preset matching distance threshold.

[0049] Specifically, a set of static environmental features can be extracted from the perceived data first. Based on the vehicle's reported location information, a set of real environmental features corresponding to static environmental features is extracted from the high-definition map. Subsequently, a coordinate registration algorithm was used to calculate the set of static environmental features. and the set of real-world environmental features Environmental matching error between The expression for the coordinate registration algorithm is as follows:

[0050] in, The number of feature points to be matched. Feature points in the static environment feature set Feature points in the real environment feature set The Euclidean distance between them.

[0051] Subsequently, based on environmental matching error and matching distance threshold Perform a static environment characteristic consistency determination. If Exceeding the matching distance threshold (For example, 0.5 meters), then the vehicle location report is determined to be distorted or the data source is a fake environment, and the static environment feature consistency check fails, resulting in direct interception. If Not exceeding the matching distance threshold If the static environment feature consistency check passes, then the dynamic target motion rationality check will be performed.

[0052] S210. Verify the rationality of dynamic target motion based on the perceived data.

[0053] In this embodiment of the invention, the dynamic target motion rationality verification can be performed on the sensing data. If the dynamic target motion rationality verification passes, step S220 is executed. If the dynamic target motion rationality verification fails, the sensing data is intercepted.

[0054] In this embodiment of the invention, when verifying the rationality of dynamic target motion in the sensing data, the dynamic target in the sensing data can be verified for instantaneous motion state, and in response to the instantaneous motion state verification being passed, the dynamic target in the sensing data can be verified for kinematic residuals.

[0055] Instantaneous motion state can include instantaneous velocity and / or instantaneous acceleration. By checking whether the instantaneous velocity and / or instantaneous acceleration of the dynamic target meet the set conditions, it can be verified whether the instantaneous motion state of the dynamic target has changed abruptly.

[0056] In one specific embodiment, when performing instantaneous motion state verification on a dynamic target in the sensing data, the instantaneous velocity and instantaneous acceleration of the dynamic target can be determined based on the position information of the dynamic target at the current moment and the position information of the dynamic target at historical moments; and the instantaneous motion state verification result can be determined based on the instantaneous velocity and instantaneous acceleration.

[0057] Specifically, an instantaneous acceleration threshold can be set based on the type of dynamic target, and the speed limit of the road segment where the vehicle is located can be obtained to set a speed threshold based on that speed limit. The instantaneous speed of the dynamic target is compared with the speed threshold, and the instantaneous acceleration of the dynamic target is compared with the instantaneous acceleration threshold. If the instantaneous speed of the dynamic target is greater than the speed threshold, or if the instantaneous acceleration of the dynamic target is greater than the instantaneous acceleration threshold, it is determined that the instantaneous motion state of the dynamic target has changed abruptly, the instantaneous motion state verification fails, the data is intercepted and marked as tampered data or ghost data, etc.; otherwise, the instantaneous motion state verification passes.

[0058] For instantaneous acceleration thresholds, different physical limits can be set based on different types of dynamic targets. In some specific embodiments, to prevent false positives, the instantaneous acceleration threshold can be set slightly higher than the physical limit. For example, the instantaneous acceleration threshold for pedestrians can be set to 3 m / s² to 5 m / s². The instantaneous acceleration threshold for bicycles can be set to 3 m / s² to 4 m / s². The instantaneous acceleration threshold for electric bicycles can be set to 4 m / s² to 6 m / s². The instantaneous acceleration threshold for passenger vehicles (e.g., sedans, SUVs, taxis) can be set to 10 m / s² to 12 m / s². The instantaneous acceleration threshold for heavy trucks / commercial vehicles can be set to 4 m / s² to 6 m / s².

[0059] For the speed limit of the road segment where the vehicle is located, the speed limit can be obtained from a high-precision map as a speed threshold. In some specific embodiments, to prevent misjudgment, the speed threshold can also be set slightly higher than the speed limit of the road segment where the vehicle is located. For example, a certain error tolerance floating ratio (such as 120% or 150% of the speed limit) can be added to the speed limit of the road segment where the vehicle is located to serve as the speed threshold.

[0060] In this embodiment of the invention, kinematic residual verification can verify whether the spatial position of a dynamic target is reasonable. In one specific embodiment, when performing kinematic residual verification on a dynamic target in the sensing data, the theoretical position information of the dynamic target at the current moment can be predicted based on the historical trajectory of the dynamic target; the kinematic residual of the dynamic target can be determined based on the position information of the dynamic target at the current moment and the theoretical position information; and the kinematic residual verification result can be determined based on the kinematic residual and the motion boundary threshold.

[0061] Specifically, the historical trajectory of the dynamic target can be acquired. This historical trajectory consists of the target's effective position at a historical moment. A motion state prediction model (e.g., Kalman filter algorithm) is used to predict the target's theoretical position at the current moment. Based on this theoretical position information and motion information such as the target's maximum acceleration or maximum angular velocity, a motion boundary threshold is determined. Subsequently, based on the target's current position and theoretical position information, the kinematic residual is calculated and compared with the motion boundary threshold to determine the kinematic residual verification result. If the kinematic residual is greater than the motion boundary threshold, the target's spatial position is unreasonable, the kinematic residual verification fails, and the data is intercepted. If the kinematic residual is not greater than the motion boundary threshold, the target's spatial position is reasonable, the kinematic residual verification passes, and a multi-source field-of-view consistency verification is further performed.

[0062] When determining the motion boundary threshold of a dynamic target, the theoretical position information can be used as the center point to determine the legal motion space of the dynamic target. Based on this legal motion space, the motion boundary threshold of the dynamic target can be determined. Specifically, based on the theoretical position information of the dynamic target, combined with the physical kinematic limits (such as maximum acceleration, maximum turning angular velocity, etc.) of the target's category and its actual motion velocity, the maximum three-dimensional spatial boundary that the target may reach within a preset time interval can be calculated. The radius of this spatial boundary is defined as the motion boundary threshold.

[0063] For motion boundary threshold This can be understood as a dynamically calculated spatial distance, rather than a fixed constant. That is, within a given time interval... Within, the dynamic target moves at its current speed, even with maximum acceleration. The deviation between the theoretical position and the actual position during a rapid speed or sudden braking. (i.e., kinematic residuals) cannot exceed .if Exceeded If so, it means that the perception data containing the dynamic target is fake or abnormal, that is, untrusted data.

[0064] S220. Perform multi-source field-of-view consistency verification on the perceived data.

[0065] In this embodiment of the invention, the perception data that has passed the dynamic target motion rationality verification can be subjected to multi-source field-of-view consistency verification. If the multi-source field-of-view consistency verification passes, the data transaction process can be carried out. If the multi-source field-of-view consistency verification fails, the perception data is intercepted.

[0066] In this embodiment of the invention, when performing multi-source field-of-view consistency verification on the sensing data, second sensing data that has spatiotemporal overlap with the sensing data can be obtained; based on the detection box information for the same dynamic target in the sensing data and the second sensing data, multi-source field-of-view consistency verification is performed.

[0067] Spatiotemporal overlap refers to second sensing data collected by a data acquisition device (e.g., roadside sensing device or other vehicle) at the same time and location as the sensing data collection.

[0068] In one specific embodiment, when performing multi-source field-of-view consistency verification based on the detection box information of the same dynamic target in the perception data and the second perception data, the first detection box information of the dynamic target can be obtained from the perception data, and the first detection box information can be projected onto the coordinate system where the second perception data is located to obtain the projected detection box information of the dynamic target; the multi-source field-of-view consistency verification is performed based on the degree of overlap between the projected detection box information of the dynamic target and the second detection box information of the dynamic target in the second perception data.

[0069] Specifically, firstly, searching for vehicles The uploaded sensing data has a second data source with spatiotemporal overlap (such as roadside sensing devices). The second sensing data was collected. Then, Reported dynamic targets bounding box Transformed to using geometric projection transformation On the image plane, the projection frame is obtained. Then calculate the projection frame. and actual detected bounding boxes of dynamic targets Crossover ratio between :

[0070] like Less than the preset overlap threshold ( ),and If it is in normal working condition, then determine The uploaded data is fake, and the data is marked and blocked.

[0071] In this embodiment of the invention, the multi-layered cascaded verification mechanism verifies the perception data returned by the vehicle based on the data perception request. Only perception data that passes all verifications is recognized as a high-confidence valid asset and allowed to proceed to the subsequent value assessment. Data that fails verification will be intercepted and marked as fraudulent or low-quality data, and will not be executed in subsequent steps. This improves the reliability of data transmission, the credibility of transaction data, and protects the rights and interests of data buyers and the fairness of data market transactions.

[0072] In some embodiments, such as Figure 3 As shown, the authenticity verification and value assessment of the perceived data can be performed through steps S300~S360.

[0073] S300, Determine the task type of the sensing task corresponding to the sensing data.

[0074] If the task type of the perception task is a continuous perception task, execute steps S310~S340. If the task type of the perception task is an isolated perception task, execute steps S350~S360.

[0075] S310. Determine the time delay factor of the sensing data based on the time of receiving the sensing data and the time of collecting the sensing data.

[0076] In this embodiment, the latency factor is related to the time of receiving the sensed data and the time of acquiring the sensed data. The latency factor can be determined based on the difference between the time of receiving the sensed data and the time of acquiring the sensed data. A larger difference indicates a larger latency and a smaller latency factor, while a smaller difference indicates a smaller latency and a larger latency factor.

[0077] In some specific embodiments, the time delay factor of the sensed data can be calculated based on the following formula. .

[0078]

[0079] in, Indicates the delay score. This represents the time difference between when the vehicle's sensors generate sensing data and when the cloud platform receives the sensing data. This indicates the upper limit of the lifecycle of the perceived data. The upper limit of the lifecycle of perceived data varies depending on the task type. The upper limit of the lifecycle of perceived data for isolated perception tasks is greater than that for continuous perception tasks. (For example, the upper limit of the lifecycle for isolated perception tasks (such as road watering, construction alerts, traffic congestion, and accident reporting) is 1 to 10 minutes, e.g., 3 minutes, 5 minutes, 8 minutes, etc.). The upper limit of the lifecycle for continuous perception tasks (such as virtual blind spot filling for autonomous driving, remote driving takeover, and automatic parking blind spot filling) is 1 second, e.g., 200 ms, 500 ms, etc.). It should be noted that these illustrative examples of the lifecycle upper limit are not intended to limit the invention. In practical applications, other timeframes can be flexibly set according to the actual situation.

[0080] In this embodiment, based on the above formula, the time delay factor of the sensed data can be accurately evaluated when... Greater than If the perceived data is not used, its value drops to zero, thus ensuring the real-time nature of the perceived data.

[0081] S320. Determine the timeliness of the sensed data based on the time delay factor.

[0082] If the timeliness determination passes, proceed to step S330; if the timeliness determination fails, delete the perceived data.

[0083] Specifically, the time delay factor can be compared with a pre-set time delay threshold (e.g., 1). If the time delay factor is less than the time delay threshold, the timeliness determination passes; otherwise, the timeliness determination fails.

[0084] S330, Verify the authenticity of the perceived data.

[0085] The timeliness of the perceived data is verified for authenticity. If the verification passes, step S340 is executed; otherwise, the perceived data is deleted.

[0086] S340. Based on the time delay factor, combined with the quality and scarcity of the sensed data, the value of the sensed data is assessed to obtain the value of the sensed data.

[0087] In this embodiment, the value of the sensing data can be determined by comprehensively evaluating the sensing data based on the time delay factor, the quality of the sensing data, and the scarcity of the sensing data.

[0088] S350, Verify the authenticity of the perceived data.

[0089] In this embodiment, the authenticity of the sensing data returned based on the isolated sensing task is verified. If the authenticity verification passes, step S360 is executed; if the authenticity verification fails, the sensing data is deleted.

[0090] S360. Based on the quality, scarcity, and latency of the perceived data, the value of the perceived data is assessed to obtain the value of the perceived data.

[0091] In this embodiment, for the sensing data returned by the isolated sensing task that has passed the authenticity verification, the value of the sensing data is evaluated based on its quality, scarcity, and latency. Specifically, the process of evaluating the value of the sensing data based on its quality, scarcity, and latency is as follows: A quality score is determined based on the hardware configuration information of the sensing data acquisition device; a scarcity score is determined based on the scarcity of the sensing data; a latency factor is determined based on the time the sensing data is received and the time it is acquired; and the value of the sensing data is determined based on the quality score, scarcity score, and latency factor.

[0092] The technical solution of this invention can flexibly perform authenticity verification and value assessment on sensing data according to the task type of the sensing task, thereby ensuring the accuracy and reliability of the selected sensing data, reducing data transmission latency, and improving data transmission real-time performance. When the sensing task is a continuous sensing task, a latency factor is determined based on the time the sensing data is received and the time the data is collected. The timeliness of the sensing data is then judged based on this latency factor. Sensing data that passes the timeliness judgment is verified for authenticity, while data that fails is deleted. If the authenticity verification passes, the value of the sensing data is assessed based on the pre-calculated latency factor, combined with the quality and scarcity of the sensing data, to obtain the value of the sensing data. When the sensing task is an isolated sensing task, the authenticity of the sensing data returned from the isolated sensing task is verified. If the authenticity verification passes, the value of the sensing data is assessed based on its quality, scarcity, and latency, to obtain the value of the sensing data. Therefore, when facing isolated sensing tasks with low latency requirements, all sensing data received within a preset window period can be used for authenticity verification and value assessment to select accurate and reliable sensing data. When facing perception tasks with high latency requirements, a latency factor is calculated before data verification to eliminate some data with high latency. This reduces the amount of data required for data authenticity verification and value assessment, thereby reducing data transmission latency, improving data transmission real-time performance, and ensuring the accuracy and reliability of the selected perception data.

[0093] In some embodiments, the quality score of the sensed data can be determined based on the hardware configuration information of the sensed data acquisition device to determine the quality of the sensed data; the scarcity score of the sensed data can be determined based on the scarcity of the sensed data to determine the scarcity of the sensed data; and the value of the sensed data can be determined based on the quality score, the scarcity score, and the time delay factor.

[0094] In this embodiment, the quality score of the sensed data can be determined based on the hardware configuration information of the data acquisition device. Higher hardware configuration results in a higher quality score for the sensed data, and correspondingly, a higher value for the sensed data.

[0095] In some specific embodiments, the higher the hardware configuration information of the acquisition device, the higher the quality score of the perceived data, and the quality score ranges from 0.1 to 1.

[0096] In one specific embodiment, the quality score of the perceived data can also be determined based on the mapping relationship in Table 1.

[0097] Table 1

[0098] In this embodiment, the scarcity score of the perceived data can also be determined based on the scarcity of the perceived data. The rarer the perceived data, the higher the scarcity score, and correspondingly, the higher the value of the perceived data.

[0099] In one specific embodiment, the scarcity score of the perceived data can be determined based on the number of perceived data returned in response to the same data perception request. The fewer the number of perceived data returned in response to the same data perception request, the scarcer the perceived data, and the higher the scarcity score.

[0100] In one specific embodiment, the scarcity score of the perceived data can be calculated based on the following formula. :

[0101] in, This indicates the number of sensing data returned for the same data sensing request.

[0102] In a specific application scenario, a data requester initiates a visual blind spot filling request, and the cloud platform issues a data perception request based on this request. For this data perception request, if only one vehicle on the entire road segment can clearly see the area through a side-mounted camera, then the perception data returned by that vehicle... This indicates that perception data is extremely scarce; if two vehicles with similar configurations can provide data for this blind spot, then the perception data returned by the two vehicles will be... This indicates that the perception data is moderately scarce; if the area requiring blind spot coverage is located at a busy open intersection, where more than 10 vehicles can provide overlapping perception information, then... This indicates that the perception data returned by any vehicle is general data.

[0103] In one embodiment, the sum of the quality score and the scarcity score can be multiplied by the latency factor to obtain the value of the perceived data.

[0104] This invention provides a technical solution for determining the value of perceived data based on the hardware configuration information of the data acquisition device, the scarcity score, and the latency score. This solution reduces the complexity of data transaction calculations, eliminates the need for data iteration and model inference, and allows for data value calculation and transaction settlement to be completed instantly upon receiving the perceived data. It achieves high real-time data sharing and trading. By using quality evaluation based on hardware configuration information and scarcity evaluation considering supply and demand, the rationality, adaptability, and interpretability of the entire scoring mechanism are improved. The introduction of a latency attenuation factor not only penalizes outdated data in the final pricing but also directly blocks the upload of timed-out invalid data at the edge, thereby filtering out useless information, significantly reducing redundant computing load on the cloud platform, minimizing network communication bandwidth waste, improving communication speed between the sensing entity and the cloud platform, and further enhancing the real-time performance of data transactions.

[0105] In some embodiments, a quality coefficient related to quality and a scarcity coefficient related to scarcity can also be introduced to determine the value of the perceived data. Specifically, a quality coefficient is determined based on the data perception request corresponding to the perceived data; a scarcity coefficient is determined based on the perception scenario corresponding to the perceived data; a weighted sum of the quality score and quality coefficient, and the scarcity score and scarcity coefficient are calculated, and the sum is multiplied by a time delay factor to obtain the value of the perceived data.

[0106] In this embodiment, the quality coefficient (W1) can be determined based on the accuracy requirements, granularity requirements, and security sensitivity requirements of the data perception request corresponding to the returned perception data.

[0107] In one specific embodiment, the data perception request can define dedicated fields to store accuracy requirements, granularity requirements, and security sensitivity requirements respectively. Thus, after receiving the perception data, the quality coefficient can be determined based on the accuracy requirements, granularity requirements, and security sensitivity requirements of the data perception request corresponding to the perception data.

[0108] In one specific embodiment, the quality coefficient can be determined based on the sum of accuracy requirements, granularity requirements, and safety sensitivity requirements.

[0109] In one specific embodiment, the quality coefficient can also be determined based on the sum of accuracy requirements, granularity requirements, and safety sensitivity requirements, combined with other weights.

[0110] In the embodiments of the present invention, there are multiple ways to determine the quality coefficient. It should be understood that the above are only any two of the multiple ways to determine the quality coefficient and are not intended to limit the present invention.

[0111] In this embodiment, the scarcity coefficient represents the degree of importance the scoring mechanism places on spatial coverage and blind spot filling needs. Therefore, the scarcity coefficient (W2) can be determined based on whether the sensing scene corresponding to the sensing data is scarce. If the sensing scene corresponding to the sensing data is relatively scarce, the scarcity coefficient is higher; conversely, the scarcity coefficient is lower.

[0112] In one specific embodiment, when the perception data to be acquired is the perception data of a high-incidence area of ​​visual blind spots, an area with frequent accidents, or a certain area under extreme weather conditions, it is determined that the perception scene corresponding to the perception data is relatively scarce, and thus the value of the scarcity coefficient will be relatively high.

[0113] For example, if the scenario in which the vehicle acquires perception data is during a rainstorm, the roadside cameras will be relatively blurry due to the rainstorm, resulting in a scarcity of close-range data provided by vehicles on the road. Therefore, the scarcity coefficient of the perception data provided by the vehicle can be set to 0.8.

[0114] For example, if a vehicle equipped with only a regular camera is located in the blind spot, the platform will increase the scarcity coefficient of the perception data it provides, for example, to 0.7, in order to encourage the vehicle to drive into the blind spot and provide perception data.

[0115] In this embodiment, the scarcity coefficient can be dynamically adjusted according to the scarcity of the perception scenario, thereby strengthening the emphasis on data space coverage and blind spot filling needs, encouraging more vehicles to participate in perception data sharing and trading, and improving the adaptability of the entire scoring mechanism to various scenarios.

[0116] In this embodiment, the quality score and quality coefficient of the perceived data, as well as the scarcity score and scarcity coefficient of the perceived data, can be weighted and summed. Based on the weighted summation result and the delay score, the value of the perceived data can be determined.

[0117] In some specific embodiments, the value of perceived data can be determined based on a data value assessment model, combining the quality score, scarcity score, and latency score of the perceived data, as well as the quality coefficient and scarcity coefficient. The data value assessment model is as follows:

[0118] in, This indicates the value of perceived data. This represents the quality score of the perceived data. Indicates the quality coefficient. This represents the scarcity score of perceived data. Represents the scarcity coefficient. This represents the time delay factor.

[0119] like Figure 4 As shown, the quality score of the perceived data can be calculated. Scarcity score and delay decay factor Then, a quality coefficient that can be adjusted by weights is introduced. and deficiency coefficient And respectively, the quality scores of the perceived data were determined. and quality coefficient Perform weighted calculations and obtain a scarcity score for the perceived data. and scarcity coefficient Perform weighted calculations and add the two weighted calculation results together to obtain the preliminary value of the perceived data. (Then the initial value of the perceived data and the time delay decay factor will be determined.) Multiply the values ​​to obtain the perceived data, and use this value to determine the final price of the perceived data.

[0120] This invention further improves the adaptability and interpretability of the entire scoring mechanism by evaluating the quality of hardware configuration information and considering the scarcity of supply and demand. It matches the value of data with the dynamically changing environment, as well as the accuracy, granularity, and security sensitivity requirements of the sensing task. This improves the matching degree between the selected target sensing data and the sensing needs published by the service requesting subject, and also improves the matching degree between the target sensing data and the dynamically changing environment. In this way, the best-fit sensing data can be selected based on the needs of the service requesting subject and the environment in which the service requesting subject is located.

[0121] In some embodiments, the step of determining the quality coefficient based on the data perception request corresponding to the perceived data may include determining the comprehensive score of the data perception request based on the data perception request corresponding to the perceived data; and determining the quality coefficient based on the comprehensive score of the data perception request, a pre-set theoretical score, and a quality coefficient limit value.

[0122] Specifically, the theoretical score and the quality coefficient limit are used to ensure that different types of tasks can obtain corresponding benefits. In this embodiment, the comprehensive score of the data perception request can be determined first based on the accuracy requirements, granularity requirements, and security sensitivity requirements. Then, the quality coefficient W1 is determined based on the comprehensive score of the data perception request, the pre-set theoretical score, and the quality coefficient limit.

[0123] In this embodiment, the specific scores for accuracy requirements, granularity requirements, and security sensitivity requirements are determined according to the specific type of data perception request. The comprehensive score of the data perception request is obtained based on the sum of the scores of the three.

[0124] In one specific embodiment, the accuracy requirement score is set based on the task's tolerance for spatial positioning. For example, when the task's spatial positioning error is in the meter range (e.g., congestion reporting), the accuracy requirement can be set to a lower score (e.g., 1 point). When the task's spatial positioning error is in the centimeter range (e.g., remote takeover, road pothole location), the accuracy requirement can be set to a higher score (e.g., 5 points). The accuracy requirement score setting includes, but is not limited to, the examples above, and can also be set to different values ​​based on the actual application scenario.

[0125] In one specific embodiment, the score for granularity requirement is set based on the task's requirement for data detail. For example, when the task is an event-level task (e.g., traffic congestion reporting, accident reporting, road construction, etc.), the granularity requirement can be set to a lower score (e.g., 1 point). When the task is a feature-level task (e.g., remote driving takeover, small obstacle detection, etc.), the granularity requirement can be set to a higher score (e.g., 5 points).

[0126] In one specific embodiment, the score for the safety sensitivity requirement is set based on whether a mission failure would result in a direct collision risk. For example, a higher score (e.g., 5 points) is set when the mission is related to life safety, but a lower score (e.g., 1 point) can be set when the mission is not related to safety.

[0127] The following table 2 illustrates the score settings for the accuracy requirements, granularity requirements, and safety sensitivity requirements described in this embodiment. It should be understood that the following embodiments are only used to explain the present invention and are not intended to limit the present invention.

[0128] Table 2

[0129] In this embodiment, the quality coefficient W1 represents the degree of importance the scoring mechanism places on data accuracy and physical reliability, and the range of the quality coefficient is: In other words, when the cloud platform issues a high-precision data perception request (e.g., remote driving takeover, high-precision map updates, or small obstacle recognition), the perception data returned by the sensing entity will be appropriately increased during the comprehensive scoring calculation. The value of .

[0130] In one specific embodiment, data perception requests can be categorized into multiple levels, from low to high, based on the different perception tasks they encompass. Correspondingly, the theoretical scoring values ​​and quality coefficient limits also differ. The lower the level of the data perception request, the lower the corresponding theoretical scoring value and quality coefficient limit, and vice versa.

[0131] In one specific embodiment, the theoretical scoring value includes a theoretical maximum value and a theoretical minimum value, and the quality coefficient limit value includes an upper limit and a lower limit of the quality coefficient limit value, thereby ensuring that the perceived data returned by any level of data perception request can obtain benefits.

[0132] In one specific embodiment, the quality coefficient can be determined based on a dynamic adjustment model for the quality coefficient, combining the comprehensive score of the data perception request, a pre-set theoretical score value, and a quality coefficient limit value. The dynamic adjustment model for the quality coefficient is as follows:

[0133] in, Indicates the quality coefficient. This represents the overall score of the data-aware request. This represents the theoretical minimum value of the scoring. This represents the theoretical maximum value of the score. This represents the lower limit of the quality coefficient limit. This indicates the upper limit of the quality coefficient limit.

[0134] In this embodiment, the theoretical minimum score can be set to 1~5, for example, 2, 3, 4, etc., the theoretical maximum score can be set to 15~18, for example, 15, 16, 17, etc., the lower limit of the quality coefficient limit can be set to 0.1~0.3, for example, 0.2, and the upper limit of the quality coefficient limit can be set to 0.9 or above, for example, 0.9, 0.95, etc., thereby ensuring that low-level tasks can also have basic benefits, and the calculated result of the quality coefficient is always <1.

[0135] The process of determining the quality coefficient is illustrated below through specific embodiments. It should be understood that the following examples are only used to explain the present invention and are not intended to limit the present invention.

[0136] In this embodiment, Set to 15. Set to 3. Set to 0.2. The overall score for each task and the determined quality coefficient are set to 0.95, as shown in Table 3.

[0137] Table 3

[0138] As can be seen from Table 3, the scoring mechanism of this invention provides certain rewards to vehicles with various configurations participating in the task, in order to encourage all vehicles to actively participate in the sharing of perception data. Furthermore, it provides higher rewards to high-configuration vehicles that can undertake various high-precision perception tasks, thereby encouraging high-configuration vehicles to participate in high-precision perception tasks.

[0139] Based on the same inventive concept, according to another aspect of the present invention, embodiments of the present invention also provide a data evaluation and transmission system based on a vehicle-road-cloud collaborative environment, the data evaluation and transmission system being configured to perform the steps of the data evaluation and transmission method for a vehicle-road-cloud collaborative environment as described in the above embodiments.

[0140] In some embodiments, such as Figure 5 As shown, the data evaluation and transmission system based on the vehicle-road-cloud cooperative environment includes a cloud platform, vehicles, and roadside equipment. The data evaluation and transmission process based on this vehicle-road-cloud cooperative environment data evaluation and transmission system is as follows: Figure 6 As shown.

[0141] The cloud platform acquires sensor hardware configuration information (such as LiDAR beams and camera resolution), computing power level, and current perception algorithm version for all vehicles connected to it, and continuously monitors for perception tasks. If a perception task request is detected, a data perception request is sent; otherwise, the platform continues monitoring.

[0142] Vehicles that meet the perception request criteria respond to the perception task, perform data perception and preprocessing, and then upload the preprocessed perception data to the cloud platform. Preprocessing includes operations such as de-identification and lightweighting.

[0143] The cloud platform is based on a data value assessment model ( The system scores the perceived data to obtain a comprehensive score for each data point, and then selects the data that meets certain criteria (e.g., the highest comprehensive score). The selected data is then sent to the task publisher. Settlement is then automatically completed via a smart contract.

[0144] The embodiments of this invention reduce the complexity of data transaction calculations through the above-described scheme. Data value calculation and transaction settlement can be completed instantly upon receiving the perceived data, eliminating the need for data iteration and model inference. This achieves highly real-time data sharing and transactions. By using quality evaluation based on hardware configuration information and scarcity evaluation considering supply and demand, the adaptability and interpretability of the entire scoring mechanism are improved. The introduction of a latency decay factor not only penalizes outdated data in the final pricing but also directly blocks the upload of timed-out invalid data at the edge. This filters out useless information, significantly reduces the redundant computing load on the cloud platform, minimizes network communication bandwidth waste, improves the communication speed between the sensing entity and the cloud platform, and further enhances the real-time performance of data transactions.

[0145] Finally, it should be noted that those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium for the program can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc. The above computer program embodiments can achieve the same or similar effects as any of the corresponding foregoing method embodiments.

[0146] Those skilled in the art will also understand that the various exemplary logic blocks, modules, circuits, and algorithm steps described in conjunction with the disclosure herein can be implemented as electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability between hardware and software, the functionality of various illustrative components, blocks, modules, circuits, and steps has been generally described. Whether this functionality is implemented as software or as hardware depends on the specific application and the design constraints imposed on the system as a whole. Those skilled in the art can implement the functionality in various ways for each specific application, but such implementation decisions should not be construed as departing from the scope of the embodiments disclosed herein.

[0147] The above are exemplary embodiments disclosed in this invention. However, it should be noted that various changes and modifications can be made without departing from the scope of the embodiments of this invention as defined by the claims. The functions, steps, and / or actions of the methods according to the disclosed embodiments described herein do not need to be performed in any particular order. The sequence numbers of the disclosed embodiments of this invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. Furthermore, although the elements disclosed in the embodiments of this invention may be described or claimed individually, they may be understood as multiple unless explicitly limited to a singular number.

[0148] It should be understood that, as used herein, the singular form “a” is intended to include the plural form as well, unless the context clearly supports an exception. It should also be understood that, as used herein, “and / or” refers to any and all possible combinations of one or more of the associated listed items.

[0149] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the invention (including the claims) is limited to these examples. Within the framework of the invention, technical features of the above embodiments or different embodiments can be combined, and many other variations of different aspects of the invention exist, which are not provided in the details for the sake of brevity. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the invention should be included within the protection scope of the invention.

Claims

1. A data evaluation and transmission method based on a vehicle-road-cloud collaborative environment, characterized in that, include: In response to receiving a perception task published by the service requesting entity, a data perception request is sent to all target vehicles based on the road information that the perception task needs to perceive. In response to receiving the perception data returned by the target vehicle within a preset window period, the perception data is verified for authenticity and evaluated for value according to the task type of the perception task, so as to obtain the value of the perception data. Based on the perceived data value, target perceived data is determined from all perceived data, and the target perceived data is sent to the service request subject.

2. The method according to claim 1, characterized in that, Based on the task type of the perception task, the perception data is verified for authenticity and its value is evaluated. The value of the perception data includes: In response to the fact that the task type of the sensing task is a continuous sensing task, the time delay factor of the sensing data is determined based on the time of receiving the sensing data and the time of collecting the sensing data; The timeliness of the perceived data is determined based on the time delay factor. In response to the timeliness determination being passed, the authenticity of the perceived data is verified. In response to the successful verification of authenticity, the perceived data is evaluated for value based on the time delay factor, combined with the quality and scarcity of the perceived data, to obtain the value of the perceived data.

3. The method according to claim 2, characterized in that, Based on the time delay factor, and considering the quality and scarcity of the sensed data, a value assessment is performed on the sensed data to determine its value, which includes: The quality score of the sensed data is determined based on the hardware configuration information of the data acquisition device. Based on the scarcity of the perceived data, a scarcity score for the perceived data is determined; The value of the perceived data is determined based on the quality score, the scarcity score, and the time delay factor.

4. The method according to claim 3, characterized in that, Determining the value of the perceived data based on the quality score, the scarcity score, and the latency factor includes: Determine the quality coefficient based on the data perception request corresponding to the perceived data; Based on the perception scenario corresponding to the perception data, determine the scarcity coefficient; The value of the perceived data is obtained by weighting and summing the quality score and the quality coefficient, the scarcity score and the scarcity coefficient, and then multiplying the sum by the time delay factor.

5. The method according to claim 4, characterized in that, Based on the data perception request corresponding to the perceived data, the quality coefficient is determined as follows: Based on the data perception request corresponding to the perceived data, determine the comprehensive score of the data perception request; The quality coefficient is determined based on the comprehensive score of the data perception request, the pre-set theoretical score value, and the quality coefficient limit value.

6. The method according to claim 5, characterized in that, The quality coefficient is determined based on the comprehensive score of the data perception request, the pre-set theoretical score value, and the quality coefficient limit value, including: Based on the dynamic adjustment model of the quality coefficient, and combining the comprehensive score of the data perception request, the pre-set theoretical score value, and the quality coefficient limit value, the quality coefficient is determined, wherein the dynamic adjustment model of the quality coefficient is: in, Indicates the quality coefficient. This represents the overall score of the data-aware request. This represents the minimum theoretical score. This represents the theoretical maximum value of the scoring system. This represents the lower limit of the quality coefficient limit. This indicates the upper limit of the quality coefficient limit.

7. The method according to claim 1, characterized in that, Based on the task type of the perception task, the perception data is verified for authenticity and its value is evaluated. The value of the perception data includes: In response to the fact that the task type of the perception task is an isolated perception task, the authenticity of the perception data is verified. In response to the successful verification of authenticity, the perceived data is evaluated for value based on its quality, scarcity, and latency to obtain the value of the perceived data.

8. The method according to claim 7, characterized in that, The perceived data is valued based on its quality, scarcity, and latency, and the resulting value includes: The quality score of the sensed data is determined based on the hardware configuration information of the data acquisition device. Based on the scarcity of the perceived data, a scarcity score for the perceived data is determined; The time delay factor of the sensing data is determined based on the time of receipt of the sensing data and the time of acquisition of the sensing data; The value of the perceived data is determined based on the quality score, the scarcity score, and the time delay factor.

9. The method according to claim 2 or 7, characterized in that, The verification of the authenticity of the perceived data includes: Perform static environmental feature consistency verification on the perceived data; In response to the successful verification of static environmental feature consistency, a dynamic target motion rationality verification is performed on the perceived data; In response to the successful verification of the rationality of the dynamic target motion, a multi-source field-of-view consistency verification is performed on the perceived data.

10. A data evaluation and transmission system based on a vehicle-road-cloud collaborative environment, characterized in that, The data evaluation and transmission system is configured to perform the steps of the method as described in any one of claims 1 to 9.

Images