Edge node-based vehicle-road cloud data hierarchical forwarding method, system and device

Abstract

This application provides a method, system, and device for hierarchical forwarding of vehicle-road-cloud data based on edge nodes, relating to the field of intelligent transportation technology. The method includes: real-time collection of vehicle-road-cloud traffic data through an integrated vehicle-road-cloud system, and deployment of N edge computing nodes on the roadside; feature extraction and data analysis of the vehicle-road-cloud traffic data based on the N edge computing nodes; hierarchical processing of the vehicle-road-cloud traffic data according to the analysis results; construction of a vehicle-road-cloud data forwarding strategy library; differential analysis of the hierarchical vehicle-road-cloud data based on the library; determination of the target data hierarchical forwarding strategy; and hierarchical forwarding control of the vehicle-road-cloud traffic data. This application solves the technical problems of low data processing efficiency caused by excessively high data transmission latency, uneven network bandwidth allocation, and inefficient computation in existing technologies. By deploying edge computing nodes on the roadside, it improves the processing efficiency and real-time performance of traffic data.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation technology, specifically to a method, system, and device for hierarchical forwarding of vehicle-road-cloud data based on edge nodes. Background Technology

[0002] Existing vehicle-road-cloud (V2X) systems typically rely on a centralized cloud computing architecture. Traffic data collected by vehicles and roadside equipment needs to be transmitted to a remote cloud for processing. In high-traffic, high-frequency application scenarios, this can easily lead to significant data transmission latency, especially during periods of limited network bandwidth or peak traffic hours. With the continuous increase in traffic data volume, the traditional centralized processing model results in excessive computing pressure on the cloud, not only delaying data response speed but also reducing the overall efficiency of the system. Furthermore, low-priority data consumes a large amount of bandwidth in the network, affecting the transmission of high-priority emergency data, further exacerbating the waste of network resources and the imbalance in data transmission.

[0003] In summary, existing technologies suffer from low data processing efficiency due to excessively high data transmission latency, uneven network bandwidth allocation, and inefficient computational processing. Summary of the Invention

[0004] The purpose of this application is to provide a method, system, and device for hierarchical forwarding of vehicle-road-cloud data based on edge nodes, in order to solve the technical problems of low data processing efficiency caused by excessively high data transmission latency, uneven network bandwidth allocation, and inefficient computing in the prior art.

[0005] To achieve the above objectives, this application provides a method, system, and device for hierarchical forwarding of vehicle-road-cloud data based on edge nodes.

[0006] Firstly, this application provides a method for hierarchical forwarding of vehicle-road-cloud data based on edge nodes. This method is implemented through a vehicle-road-cloud data hierarchical forwarding system based on edge nodes. The method includes: real-time collection of vehicle-road-cloud traffic data through an integrated vehicle-road-cloud system; deployment of N edge computing nodes on the roadside; mapping and transmitting the vehicle-road-cloud traffic data to the N edge computing nodes; feature extraction and data analysis of the vehicle-road-cloud traffic data based on the N edge computing nodes to obtain vehicle-road-cloud data analysis results; hierarchical processing of the vehicle-road-cloud traffic data according to the vehicle-road-cloud data analysis results to obtain hierarchical vehicle-road-cloud data; construction of a vehicle-road-cloud data forwarding strategy library; differential analysis of the hierarchical vehicle-road-cloud data based on the vehicle-road-cloud data forwarding strategy library to determine target data hierarchical forwarding strategies; and hierarchical forwarding control of the vehicle-road-cloud traffic data through the target data hierarchical forwarding strategies.

[0007] Optionally, a three-dimensional model is generated based on the roadside traffic layout information to produce a roadside traffic three-dimensional model; key parts are identified and edge nodes are divided in the roadside traffic three-dimensional model to obtain N roadside traffic edge nodes; edge device selection and deployment analysis are performed on the N roadside traffic edge nodes in sequence to construct N edge computing nodes.

[0008] Optionally, the key parts of the roadside traffic 3D model are identified according to the traffic system analysis objectives to obtain multiple sets of key roadside traffic parts; edge node partitioning rules are obtained, including spatial distribution proximity and computational load balancing; and edge nodes are partitioned based on the edge node partitioning rules to obtain N roadside traffic edge nodes.

[0009] Optionally, the feature extraction module and the data analysis module are invoked through the N edge computing nodes; the feature extraction module is used to preprocess and extract features from the vehicle-road-cloud traffic data to obtain a vehicle-road-cloud traffic association feature set; and the data analysis module is used to perform data task analysis on the vehicle-road-cloud traffic association feature set to obtain the vehicle-road-cloud data analysis results.

[0010] Optionally, the traffic system analysis target is parsed to obtain a data analysis task list; a traffic vehicle-road cloud historical dataset is collected, and each analysis task in the data analysis task list is matched and associated with the traffic vehicle-road cloud historical dataset to obtain a vehicle-road cloud task-associated dataset; the vehicle-road cloud task-associated dataset is trained and integrated for analysis tasks to construct a data analysis module, and the data analysis module is stored in the N edge computing nodes.

[0011] Optionally, a data grading dimension is constructed, which includes urgency, real-time requirements, and data value density; the vehicle-road-cloud data analysis results are evaluated in a multi-dimensional manner according to the data grading dimension to obtain vehicle-road-cloud data grading results; and the vehicle-road-cloud traffic data is processed in a graded manner based on the vehicle-road-cloud data grading results to obtain vehicle-road-cloud graded data.

[0012] Optionally, based on the vehicle-road-cloud data forwarding strategy library, the data at each level in the vehicle-road-cloud hierarchical data is matched and analyzed to obtain a data hierarchical matching strategy; real-time network conditions and system load are monitored, and the data hierarchical matching strategy is dynamically adjusted based on the real-time network conditions and system load to determine the target data hierarchical forwarding strategy.

[0013] Secondly, this application also provides a vehicle-road-cloud data hierarchical forwarding system based on edge nodes, used to execute the vehicle-road-cloud data hierarchical forwarding method based on edge nodes as described in the first aspect. The vehicle-road-cloud data hierarchical forwarding system based on edge nodes includes: a data acquisition and transmission module, used to acquire vehicle-road-cloud traffic data in real time through a vehicle-road-cloud integrated system, and deploy N edge computing nodes on the roadside to map and transmit the vehicle-road-cloud traffic data to the N edge computing nodes; a feature extraction and data analysis module, used to perform feature extraction and data analysis on the vehicle-road-cloud traffic data based on the N edge computing nodes to obtain vehicle-road-cloud data analysis results; a data hierarchical processing module, used to perform hierarchical processing on the vehicle-road-cloud traffic data according to the vehicle-road-cloud data analysis results to obtain vehicle-road-cloud hierarchical data; and a data differentiation analysis module, used to construct a vehicle-road-cloud data forwarding strategy library, perform differentiation analysis on the vehicle-road-cloud hierarchical data based on the vehicle-road-cloud data forwarding strategy library, determine a target data hierarchical forwarding strategy, and control the hierarchical forwarding of the vehicle-road-cloud traffic data through the target data hierarchical forwarding strategy.

[0014] Thirdly, this application also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the edge node-based vehicle-road-cloud data hierarchical forwarding method described in any one of the first aspects above.

[0015] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0016] A vehicle-road-cloud integrated system collects real-time traffic data from vehicles, roads, and the cloud. N edge computing nodes are deployed on the roadside, and the traffic data is mapped and transmitted to these N edge computing nodes. Feature extraction and data analysis are performed on the traffic data based on these N edge computing nodes to obtain vehicle-road-cloud data analysis results. The traffic data is then graded according to these analysis results to obtain tiered vehicle-road-cloud data. A vehicle-road-cloud data forwarding strategy library is constructed, and differentiated analysis is performed on the tiered vehicle-road-cloud data based on this library to determine target data tiered forwarding strategies. These target data tiered forwarding strategies are then used to control the tiered forwarding of the traffic data. In other words, by deploying multiple edge computing nodes on the roadside, data processing tasks are distributed to edge nodes close to the data source, reducing data transmission latency. The collected traffic data is tiered and classified according to priority, ensuring that high-priority data receives priority processing and forwarding. The forwarding strategy is dynamically adjusted based on the current network status, improving the data transmission efficiency, computing power, response speed, and network resource utilization of the vehicle-road-cloud system.

[0017] The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this application 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 merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating the hierarchical forwarding method for vehicle-road-cloud data based on edge nodes in this application.

[0020] Figure 2 This is a schematic diagram of the structure of the vehicle-road-cloud data hierarchical forwarding system based on edge nodes in this application.

[0021] Figure 3 This is a schematic diagram of the structure of an exemplary electronic device of this application.

[0022] Explanation of reference numerals in the attached figures: 11 data acquisition and transmission module, 12 feature extraction and data analysis module, 13 data classification processing module, 14 data differentiation analysis module, 300 bus, 301 receiver, 302 processor, 303 transmitter, 304 memory, and 305 bus interface. Detailed Implementation

[0023] This application provides a method, system, and device for hierarchical forwarding of vehicle-road-cloud data based on edge nodes. This addresses the technical problems of low data processing efficiency in existing technologies caused by excessively high data transmission latency, uneven network bandwidth allocation, and inefficient computation. By deploying multiple edge computing nodes on the roadside, data processing tasks are distributed to edge nodes close to the data source, reducing data transmission latency. The collected traffic data is hierarchically processed and classified according to priority, ensuring that high-priority data receives priority processing and forwarding. The forwarding strategy is dynamically adjusted based on the current network status, improving the data transmission efficiency, computing power, response speed, and network resource utilization of the vehicle-road-cloud system.

[0024] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0025] Example 1, please refer to the appendix. Figure 1 This application provides a hierarchical forwarding method for vehicle-road-cloud data based on edge nodes. The method is applied to a hierarchical forwarding system for vehicle-road-cloud data based on edge nodes, and specifically includes the following steps:

[0026] S100: Real-time vehicle-road-cloud traffic data is collected through the vehicle-road-cloud integrated system, and N edge computing nodes are deployed on the roadside to map and transmit the vehicle-road-cloud traffic data to the N edge computing nodes.

[0027] Furthermore, S100 of this application includes: performing three-dimensional modeling based on roadside traffic layout information to generate a roadside traffic three-dimensional model; identifying key parts and dividing edge nodes in the roadside traffic three-dimensional model to obtain N roadside traffic edge nodes; and sequentially performing edge device selection and deployment analysis on the N roadside traffic edge nodes to construct N edge computing nodes.

[0028] Furthermore, this application also includes the following steps: identifying key parts of the roadside traffic 3D model according to the traffic system analysis objectives to obtain multiple sets of key roadside traffic parts; obtaining edge node partitioning rules, the edge node partitioning rules including spatial distribution proximity and computational load balancing; partitioning the multiple sets of key roadside traffic parts into edge nodes based on the edge node partitioning rules to obtain N roadside traffic edge nodes.

[0029] Specifically, this involves obtaining detailed information about roadside traffic layout, including road length, width, intersections, traffic lights, and the location of road signs. Roadside traffic layout information is the geographic and spatial information of roads and traffic facilities, including data on all entities affecting traffic management and their spatial relationships, such as traffic flow, road surface conditions, intersection locations, traffic light layouts, and traffic signs.

[0030] Using roadside traffic layout information, 3D modeling software is used to model the roadside environment. Two-dimensional geographic information is converted into data in a three-dimensional coordinate system, constructing a spatially recognizable road network, traffic facilities, and surrounding environment. The generated 3D roadside traffic model includes not only roads and traffic signs but also elements such as terrain and buildings. 3D modeling involves creating a model of all physical features of the road and its surrounding environment, such as roads, buildings, and traffic facilities, according to a three-dimensional coordinate system. 3D modeling helps to intuitively display all elements in space and accurately simulate the real environment.

[0031] Traffic system analysis aims to fulfill core requirements and objectives of traffic management systems, such as improving traffic flow, reducing traffic accidents, and optimizing traffic signal control. Based on these objectives, it identifies critical components for system optimization, such as high-accident-rate road sections, high-traffic intersections, or areas with traffic bottlenecks. By analyzing historical traffic data, such as accident statistics and traffic flow monitoring data, combined with 3D models, several important components are identified. These components are typically key areas for improving traffic efficiency or enhancing traffic safety.

[0032] Based on the distribution of key traffic locations, the rules for dividing edge nodes are defined. Spatial proximity requires that areas that are close to each other and have high correlation in traffic conditions be grouped into the same edge node to reduce data transmission latency across nodes. For example, multiple intersections or lane areas with high traffic volume can be grouped into the same edge node. Computational load balancing requires that the computational tasks and load on each edge node be balanced as much as possible to avoid overloading any node and affecting computational efficiency. Therefore, the division must consider not only the spatial distribution of the areas but also the computational capacity of each node. Spatial proximity ensures that traffic data from adjacent areas can be processed centrally to reduce data transmission latency. Computational load balancing ensures that each edge node is allocated an appropriate amount of computational tasks, preventing some nodes from experiencing a decrease in processing capacity due to overload.

[0033] Based on the edge node partitioning rules, multiple key traffic areas are divided to determine which areas will be handled by which edge node. Based on traffic data analysis, geospatial distribution, and load assessment, the final determination is made which edge computing node should handle each area, resulting in N roadside traffic edge nodes, each responsible for a specific roadside traffic area or key location.

[0034] For each edge computing node, appropriate computing devices, transmission devices, and related hardware resources are selected based on its location and the tasks it undertakes. Device selection must consider factors such as computing requirements, power consumption, durability, and network connectivity. For example, high-traffic areas may require high-performance computing devices to meet rapid data processing needs, while low-load areas can opt for low-power devices. After selection, a detailed deployment analysis is conducted to ensure that the device installation locations, cabling design, and network connections can effectively support real-time data transmission and processing, ultimately building an efficient edge computing node network.

[0035] Edge computing nodes are located near critical transportation infrastructure, significantly reducing data transmission latency and improving real-time data processing speed. By using edge node partitioning rules, a balanced computing load is ensured for each edge node, avoiding overload issues on any single node and improving overall system efficiency.

[0036] S200: Based on the N edge computing nodes, feature extraction and data analysis are performed on the vehicle-road-cloud traffic data to obtain the vehicle-road-cloud data analysis results.

[0037] Furthermore, S200 of this application includes: calling the feature extraction module and the data analysis module through the N edge computing nodes; using the feature extraction module to preprocess and extract features from the vehicle-road-cloud traffic data to obtain a vehicle-road-cloud traffic association feature set; and performing data task analysis on the vehicle-road-cloud traffic association feature set based on the data analysis module to obtain vehicle-road-cloud data analysis results.

[0038] Furthermore, this application also includes the following steps: parsing the data tasks of the traffic system analysis target to obtain a data analysis task list; collecting historical traffic vehicle-road-cloud datasets, matching and associating each analysis task in the data analysis task list with the historical traffic vehicle-road-cloud datasets to obtain a vehicle-road-cloud task-associated dataset; training and integrating the analysis tasks on the vehicle-road-cloud task-associated datasets respectively to construct a data analysis module, and storing the data analysis module on the N edge computing nodes.

[0039] Specifically, based on the traffic system analysis objectives, such as improving traffic flow or reducing traffic accidents, the first step is to define the analysis tasks. These objectives are then translated into concrete, actionable data tasks. For example, if the traffic system analysis objective is to optimize traffic light timing, the tasks might include: analyzing traffic flow at intersections, predicting peak traffic periods, and calculating lane occupancy rates. In this way, the complex objectives of the traffic system are broken down into specific, quantifiable analysis tasks, forming a detailed list of data analysis tasks. This list contains all relevant data analysis tasks, detailing the individual analytical work required to achieve the traffic system analysis objectives.

[0040] Historical traffic data sets from the vehicle-road cloud platform are collected, including past traffic flow, road conditions, weather conditions, and accident records. Based on the previously obtained list of data analysis tasks, each analysis task is matched with the historical dataset. For example, for the traffic flow analysis task, historical data related to traffic flow, such as traffic flow data at different times and traffic light cycles, are matched. For the accident analysis task, accident occurrence data, such as accident location, time, and vehicle speed, are associated. Through this matching, vehicle-road cloud task-related datasets are created specifically for each analysis task, containing all the historical data required to perform the task.

[0041] For each vehicle-road-cloud task-related dataset, data training and analysis tasks are performed. Taking traffic flow analysis as an example, regression analysis is used to train the model on historical traffic flow data, allowing it to learn traffic flow patterns under specific time periods and weather conditions. These training results are integrated into a data analysis module, enabling automatic real-time data analysis. This data analysis module is stored on multiple edge computing nodes distributed along traffic roads, directly processing real-time traffic data. Each edge node is equipped with an analysis module suitable for its region, performing tasks such as real-time traffic flow calculation, traffic congestion prediction, and traffic light timing adjustment. This reduces latency and improves the response speed of the traffic system.

[0042] The goal of building the data analysis module is to utilize historical datasets from the vehicle-road cloud platform to analyze traffic data using appropriate machine learning or deep learning algorithms, and to provide accurate predictions or decision support. This module includes multiple functional layers, including an input layer, feature extraction layer, training layer, and prediction layer. Input data types: Each vehicle-road cloud task-related dataset (such as traffic flow, accident data, etc.) will be preprocessed to form input data suitable for the algorithm, including traffic flow data, vehicle speed data, weather data, accident data, and time factors. All input data needs to be standardized to normalize data from different dimensions to the same unit range, avoiding excessively large or small values ​​for certain features that could affect model training. For time-series data, such as traffic flow and vehicle speed, time-series features need to be extracted, such as daily and hourly trends, and even the impact of weather. A sliding window approach can be used to capture data changes over a period of time. For some categorical data, such as weather type and accident location, categorical variables are converted into numerical variables.

[0043] After feature extraction, an appropriate machine learning or deep learning model is selected for training. Model selection needs to consider the specific task objective, such as traffic flow prediction or accident risk prediction. A neural network model is chosen to predict numerical results, such as predicting traffic flow over a future period. The vehicle-road cloud historical dataset is divided into training, validation, and test sets. The training set is used to train the model, the validation set is used to tune hyperparameters, and the test set is used to evaluate the model's final performance. The model architecture is initialized based on the selected algorithm. Processed traffic data is input into the model for training. The model parameters are tuned using backpropagation. The validation set is used to fine-tune the model's hyperparameters to avoid overfitting. The model is evaluated using the test set, the prediction error is calculated, and the training process is further optimized based on the results. After training, the format of the model's output varies depending on the task type. For traffic flow prediction, the model output is the traffic flow at each intersection in the future period; for accident risk assessment, the model output is the probability of an accident occurring at a certain time or location; for signal optimization, the model output is the signal timing scheme for each intersection, or scheduling suggestions.

[0044] Multiple trained models are integrated into a complete data analysis module based on actual needs. Different models are switched according to different task types; for example, a regression model is used for traffic flow prediction, while a classification model is used for accident prediction. The module's output is dynamically adjusted according to different task requirements. In other words, multiple individual models are integrated into a unified framework. Each module can automatically select the appropriate model for calculation based on the current data input and provide the corresponding prediction results.

[0045] Once the data analytics module is trained, it will be deployed to edge computing nodes. The purpose of deploying edge computing nodes is to reduce latency in data transmission to the remote cloud and enable faster local data processing. Each edge computing node stores analytics modules for a specific region, capable of processing traffic data collected from sensors or vehicles in real time.

[0046] The constructed data analysis modules are stored across multiple edge computing nodes distributed along traffic roads to directly process real-time traffic data. Each edge node is equipped with an analysis module suitable for its region, performing tasks such as real-time traffic flow calculation, traffic congestion prediction, and traffic light timing adjustment, reducing latency and improving the responsiveness of the traffic system.

[0047] N edge computing nodes will invoke the feature extraction module and the data analysis module respectively, based on their configured task type and the data to be processed. Each node selects the appropriate module for data processing according to its geographical location and traffic environment. Through this distributed architecture, multiple edge nodes can process traffic data from different areas simultaneously and in parallel, improving the overall processing efficiency of the system. The feature extraction module is responsible for extracting features useful for prediction, classification, or optimization from the raw traffic data. The data analysis module is the core component for in-depth analysis of traffic data based on the extracted features.

[0048] The feature extraction module preprocesses and extracts features from the received vehicle-road-cloud traffic data, removing noisy data, filling in missing values, and converting the data into a unified format. The goal of feature extraction is to transform the raw data into a structured feature set that better represents the key factors in the traffic system, allowing the data analysis module to further process it. Specifically, traffic data may contain missing or outlier values. These outliers are first processed, for example, by using interpolation to fill in missing values ​​or deleting unreasonable data points. Different types of data may have inconsistent dimensions, so they need to be standardized to the same range; common methods include Z-score standardization and min-max standardization. For time-series data, the sliding window method is used to extract data features from multiple time periods, such as the average, maximum, and minimum traffic flow over the past hour. If geospatial information, such as intersections and lanes, is involved, spatial features such as lane occupancy rate, road density, and congestion index are extracted. The extracted vehicle-road-cloud traffic association feature set will contain multi-dimensional feature information.

[0049] The data analysis module performs data task analysis based on the vehicle-road-cloud traffic association feature set. Depending on the task, the module employs different algorithms. For example, if the task is to predict future traffic flow, the module uses regression analysis for training and prediction, with the vehicle-road-cloud traffic association feature set as input and the predicted traffic flow as output. If the task is to predict traffic accidents, the module uses classification algorithms to analyze past accident data and combines this with current traffic characteristics to predict future accident risks, with features such as vehicle speed, traffic flow, and weather as input and the probability of an accident occurring within a certain time period as output. If the task is to optimize traffic light timing, the module uses reinforcement learning or optimization algorithms to output the optimal traffic light timing scheme based on real-time traffic flow and vehicle speed data.

[0050] The data analysis module combines historical and real-time data to perform corresponding analysis tasks, ultimately producing vehicle-road-cloud data analysis results, including prediction results, classification results, and optimization suggestions. By calling the feature extraction and data analysis modules in real time on edge computing nodes, it can quickly respond to traffic changes and provide timely predictions of traffic flow and accident risks. Deploying edge computing nodes on the roadside for data processing not only reduces the latency of data transmission to the remote cloud but also improves the system's response speed, ensuring real-time traffic management.

[0051] S300: Based on the analysis results of the vehicle-road-cloud data, perform hierarchical processing on the vehicle-road-cloud traffic data to obtain vehicle-road-cloud hierarchical data.

[0052] Furthermore, S300 of this application includes: constructing a data classification dimension, wherein the data classification dimension includes urgency, real-time requirements, and data value density; performing a multi-dimensional classification evaluation on the vehicle-road-cloud data analysis results according to the data classification dimension to obtain vehicle-road-cloud data classification results; and performing classification processing on the vehicle-road-cloud traffic data based on the vehicle-road-cloud data classification results to obtain vehicle-road-cloud classified data.

[0053] Specifically, data grading dimensions are defined based on the needs and objectives of the transportation system. These dimensions include urgency, real-time requirements, and data value density. Urgency refers to the urgency of the data or its importance to traffic safety and management. For example, traffic accident data may have a very high urgency level, while traffic flow data may have a relatively low urgency level. Real-time requirements determine whether the data needs immediate processing and response. For example, real-time traffic flow data typically requires rapid processing, while historical traffic flow data can be processed later. Data value density reflects the meaning and importance of the data. Data with higher decision-making value, such as accident warning data, has a higher value density; while some data, such as routine vehicle speed data, may have a lower value.

[0054] After determining the data grading dimensions, the next step is to conduct a multi-dimensional grading evaluation of the vehicle-road-cloud data analysis results. Based on the defined dimensions, the traffic data analysis results are evaluated, and their scores or priorities are calculated for each dimension. Each data point is comprehensively scored based on its included characteristics, determining its priority in the processing flow. The urgency of the data is scored based on factors such as accident severity and traffic fluctuations. Accident data scores higher, while traffic prediction scores lower. The timeliness requirement of the analyzed data is also considered. Real-time traffic flow and accident data require rapid processing and receive higher scores. The data's value density is assessed based on its impact on traffic decisions. Emergency accident data and abnormal traffic flow data score higher, while regular vehicle speed data scores lower. The resulting vehicle-road-cloud data grading results represent the grading and priority of each data point across different dimensions.

[0055] Based on the vehicle-road-cloud data classification results, each data point will be assigned a different priority, and the processing method will be determined according to the priority. High-priority data, such as accident warning data, must be processed and responded to immediately, triggering emergency response mechanisms, such as optimizing traffic light timing and sending warning messages to vehicles; medium-priority data, such as traffic flow monitoring data, requires rapid analysis and processing, but can be slightly delayed. For example, traffic flow data can be used to adjust traffic light timing to ensure smooth traffic; low-priority data, such as vehicle speed data, has a smaller impact on real-time decision-making, and can be appropriately delayed for batch processing or long-term trend analysis.

[0056] A tiered data processing strategy ensures that high-priority data is processed promptly, thereby improving system response speed and efficiency, especially during peak traffic hours or emergencies, ensuring traffic safety and smooth flow. Through multi-dimensional evaluation, high-priority and urgent data are prioritized, avoiding delays in data transmission and processing. Low-priority data can be processed later, preventing system overload and allowing for better allocation of computing resources.

[0057] S400: Construct a vehicle-road-cloud data forwarding strategy library, perform differential analysis on the vehicle-road-cloud graded data based on the vehicle-road-cloud data forwarding strategy library, determine the target data graded forwarding strategy, and perform graded forwarding control on the vehicle-road-cloud traffic data through the target data graded forwarding strategy.

[0058] Furthermore, S400 of this application includes: performing matching analysis on data at each level in the vehicle-road-cloud hierarchical data based on the vehicle-road-cloud data forwarding strategy library to obtain a data hierarchical matching strategy; monitoring real-time network conditions and system load conditions, and dynamically adjusting the data hierarchical matching strategy based on the real-time network conditions and system load conditions to determine the target data hierarchical forwarding strategy.

[0059] Specifically, based on the characteristics of the data and business needs, a vehicle-road-cloud data forwarding strategy library is constructed, containing rules and strategies for handling different types and levels of data. The library defines specific rules for forwarding data based on different data characteristics. Different processing rules are set based on data priority, such as urgency and real-time requirements. Bandwidth allocation for each type of data is determined based on the current network resource allocation strategy. For example, high-priority data may receive more bandwidth resources, while low-priority data may be temporarily postponed when bandwidth is limited. Data forwarding paths are defined based on network topology and node capabilities. For data with high real-time requirements, closer edge computing nodes may be selected for processing to ensure minimal latency. To ensure reliable data transmission, the library includes rules for redundant transmission. For example, in unstable network conditions, important data may be transmitted through multiple paths to ensure no data loss.

[0060] Based on the classification results of each data point, a matching analysis is performed in the vehicle-road-cloud data forwarding strategy library to determine the optimal forwarding strategy for each data point. The priority of each data point is evaluated based on the multi-dimensional data classification results. For example, traffic accident data has a high urgency level, therefore it needs to be matched with a high-priority forwarding strategy in the strategy library. Based on the data classification results, the most suitable forwarding rules are selected from the vehicle-road-cloud data forwarding strategy library, including priority matching, latency matching, and bandwidth matching. Based on the matching analysis, a forwarding strategy is generated for each data point and dynamically adjusted as needed. For example, if the real-time requirements of traffic flow data at a certain moment are high, its forwarding strategy is automatically adjusted to high priority, low latency, and high bandwidth.

[0061] To cope with dynamically changing network conditions, the vehicle-road-cloud system needs to monitor network bandwidth, network latency, edge computing node load, and node availability in real time. It monitors real-time bandwidth utilization to ensure reasonable allocation of network resources. When bandwidth is insufficient, it adjusts data forwarding strategies to avoid delays in high-priority data transmission. It monitors network transmission latency, especially for high-priority data. If latency is too high, it searches for low-latency transmission paths. It monitors the computing load and storage capacity of each edge computing node to prevent overloading of any node. When the load is too high, it dynamically adjusts data forwarding paths, selecting nodes with lower loads for data processing. It monitors the online status of each node in the network to ensure data can be transmitted through healthy nodes. If a node becomes unavailable, it selects a backup node.

[0062] Based on real-time network conditions and system load, the vehicle-road-cloud system dynamically adjusts its data forwarding strategy. This includes adjusting transmission priorities, forwarding paths and bandwidth allocation, dynamically allocating computing resources, and optimizing redundant transmissions. The resulting hierarchical forwarding strategy ensures that high-priority data is processed first, while simultaneously improving overall network resource utilization. The hierarchical forwarding strategy is a final rule determined after dynamic adjustments to the data forwarding strategy based on real-time network conditions and system load. It ensures that high-priority and urgent data are transmitted first when network load is high or conditions are unstable, while the processing of low-priority data can be postponed.

[0063] Based on the established target data hierarchical forwarding strategy, the vehicle-road-cloud system controls the data in real time to ensure that data of different priorities are properly forwarded and processed in the system. This ensures that data is transmitted and processed in priority order, avoids delays in high-priority data, and improves the overall network resource utilization of the system.

[0064] In summary, the edge node-based vehicle-road-cloud data hierarchical forwarding method provided in this application has the following technical effects:

[0065] A vehicle-road-cloud integrated system collects real-time traffic data from vehicles, roads, and the cloud. N edge computing nodes are deployed on the roadside, and the traffic data is mapped and transmitted to these N edge computing nodes. Feature extraction and data analysis are performed on the traffic data based on these N edge computing nodes to obtain vehicle-road-cloud data analysis results. The traffic data is then graded according to these analysis results to obtain tiered vehicle-road-cloud data. A vehicle-road-cloud data forwarding strategy library is constructed, and differentiated analysis is performed on the tiered vehicle-road-cloud data based on this library to determine target data tiered forwarding strategies. These target data tiered forwarding strategies are then used to control the tiered forwarding of the traffic data. In other words, by deploying multiple edge computing nodes on the roadside, data processing tasks are distributed to edge nodes close to the data source, reducing data transmission latency. The collected traffic data is tiered and classified according to priority, ensuring that high-priority data receives priority processing and forwarding. The forwarding strategy is dynamically adjusted based on the current network status, improving the data transmission efficiency, computing power, response speed, and network resource utilization of the vehicle-road-cloud system.

[0066] Example 2: Based on the same inventive concept as the edge node-based vehicle-road-cloud data hierarchical forwarding method in Example 1, this application also provides an edge node-based vehicle-road-cloud data hierarchical forwarding system. Please refer to the appendix. Figure 2 The edge node-based vehicle-road-cloud data hierarchical forwarding system includes:

[0067] The data acquisition and transmission module 11 is used to collect vehicle-road-cloud traffic data in real time through the vehicle-road-cloud integrated system, and deploy N edge computing nodes on the roadside to map and transmit the vehicle-road-cloud traffic data to the N edge computing nodes; the feature extraction and data analysis module 12 is used to perform feature extraction and data analysis on the vehicle-road-cloud traffic data based on the N edge computing nodes to obtain vehicle-road-cloud data analysis results; the data classification processing module 13 is used to classify the vehicle-road-cloud traffic data according to the vehicle-road-cloud data analysis results to obtain vehicle-road-cloud classified data; the data differentiation analysis module 14 is used to construct a vehicle-road-cloud data forwarding strategy library, perform differentiation analysis on the vehicle-road-cloud classified data based on the vehicle-road-cloud data forwarding strategy library, determine the target data classification forwarding strategy, and control the vehicle-road-cloud traffic data through classification forwarding using the target data classification forwarding strategy.

[0068] Furthermore, the data acquisition and transmission module 11 in the vehicle-road-cloud data hierarchical forwarding system based on edge nodes is also used for: performing three-dimensional modeling based on roadside traffic layout information to generate a roadside traffic three-dimensional model; identifying key parts and dividing edge nodes in the roadside traffic three-dimensional model to obtain N roadside traffic edge nodes; and sequentially performing edge device selection and deployment analysis on the N roadside traffic edge nodes to construct N edge computing nodes.

[0069] Furthermore, the data acquisition and transmission module 11 in the vehicle-road-cloud data hierarchical forwarding system based on edge nodes is also used to: identify key parts of the roadside traffic three-dimensional model according to the traffic system analysis objectives, and obtain multiple sets of key roadside traffic parts; obtain edge node partitioning rules, which include spatial distribution proximity and computational load balancing; and partition the multiple sets of key roadside traffic parts based on the edge node partitioning rules to obtain N roadside traffic edge nodes.

[0070] Furthermore, the feature extraction and data analysis module 12 in the edge node-based vehicle-road-cloud data hierarchical forwarding system is also used to: call the feature extraction module and the data analysis module through the N edge computing nodes; preprocess and extract features from the vehicle-road-cloud traffic data using the feature extraction module to obtain a vehicle-road-cloud traffic association feature set; and perform data task analysis on the vehicle-road-cloud traffic association feature set based on the data analysis module to obtain the vehicle-road-cloud data analysis results.

[0071] Furthermore, the feature extraction and data analysis module 12 in the edge node-based vehicle-road-cloud data hierarchical forwarding system is also used for: parsing the data task of the traffic system analysis target to obtain a data analysis task list; collecting historical traffic vehicle-road-cloud datasets, matching and associating each analysis task in the data analysis task list with the historical traffic vehicle-road-cloud datasets to obtain a vehicle-road-cloud task-associated dataset; training and integrating the analysis tasks on the vehicle-road-cloud task-associated datasets respectively to construct a data analysis module, and storing the data analysis module in the N edge computing nodes.

[0072] Furthermore, the data classification processing module 13 in the edge node-based vehicle-road-cloud data classification and forwarding system is also used to: construct data classification dimensions, including urgency, real-time requirements, and data value density; perform multi-dimensional classification evaluation on the vehicle-road-cloud data analysis results according to the data classification dimensions to obtain vehicle-road-cloud data classification results; and perform classification processing on the vehicle-road-cloud traffic data based on the vehicle-road-cloud data classification results to obtain vehicle-road-cloud classified data.

[0073] Furthermore, the data differentiation analysis module 14 in the vehicle-road-cloud data hierarchical forwarding system based on edge nodes is also used to: perform matching analysis on the data at each level in the vehicle-road-cloud hierarchical data based on the vehicle-road-cloud data forwarding strategy library to obtain a data hierarchical matching strategy; monitor real-time network conditions and system load conditions, and dynamically adjust the data hierarchical matching strategy based on the real-time network conditions and system load conditions to determine the target data hierarchical forwarding strategy.

[0074] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The edge node-based vehicle-road-cloud data hierarchical forwarding method and specific examples in the aforementioned embodiment one are also applicable to the edge node-based vehicle-road-cloud data hierarchical forwarding system of this embodiment. Through the foregoing detailed description of the edge node-based vehicle-road-cloud data hierarchical forwarding method, those skilled in the art can clearly understand the edge node-based vehicle-road-cloud data hierarchical forwarding system of this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0075] Example 3: Based on the same inventive concept as the edge node-based vehicle-road-cloud data hierarchical forwarding method in Example 1, this application also provides an electronic device, including: at least one processor; a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of any of the edge node-based vehicle-road-cloud data hierarchical forwarding methods in Example 1.

[0076] Appendix Figure 3 This is a schematic diagram of the structure of an exemplary electronic device of this application. Figure 3 In this document, the bus architecture is represented by bus 300. Bus 300 may include any number of interconnected buses and bridges, and bus 300 connects various circuits including one or more processors represented by processor 302 and memory represented by memory 304. Bus 300 may also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. Bus interface 305 provides an interface between bus 300 and receiver 301 and transmitter 303. Receiver 301 and transmitter 303 may be the same element, i.e., a transceiver, providing a unit for communicating with various other devices over a transmission medium. Processor 302 is responsible for managing bus 300 and general processing, while memory 304 can be used to store data used by processor 302 during operation.

[0077] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0078] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A hierarchical forwarding method for vehicle-road-cloud data based on edge nodes, characterized in that, include: The vehicle-road-cloud integrated system collects vehicle-road-cloud traffic data in real time, and deploys N edge computing nodes on the roadside to map and transmit the vehicle-road-cloud traffic data to the N edge computing nodes. Based on the N edge computing nodes, feature extraction and data analysis are performed on the vehicle-road-cloud traffic data to obtain vehicle-road-cloud data analysis results. Based on the analysis results of the vehicle-road-cloud data, the vehicle-road-cloud traffic data is classified and processed to obtain vehicle-road-cloud classified data. A vehicle-road-cloud data forwarding strategy library is constructed. Based on the vehicle-road-cloud data forwarding strategy library, differential analysis is performed on the vehicle-road-cloud graded data to determine the target data graded forwarding strategy. The vehicle-road-cloud traffic data is then controlled by graded forwarding through the target data graded forwarding strategy.

2. The hierarchical forwarding method for vehicle-road-cloud data based on edge nodes as described in claim 1, characterized in that, Deploy N edge computing nodes on the roadside, including: Three-dimensional modeling is performed based on roadside traffic layout information to generate a three-dimensional roadside traffic model. The key parts of the roadside traffic 3D model are identified and the edge nodes are divided to obtain N roadside traffic edge nodes. The edge device selection and deployment analysis are performed sequentially on the N roadside traffic edge nodes to construct N edge computing nodes.

3. The hierarchical forwarding method for vehicle-road-cloud data based on edge nodes as described in claim 2, characterized in that, The key parts of the roadside traffic 3D model are identified and edge nodes are divided to obtain N roadside traffic edge nodes, including: According to the traffic system analysis objectives, the key parts of the roadside traffic 3D model are identified to obtain multiple sets of key roadside traffic parts. Obtain edge node partitioning rules, which include spatial distribution proximity and computational load balancing. Based on the edge node partitioning rules, the multiple sets of key roadside traffic locations are partitioned into edge nodes to obtain N roadside traffic edge nodes.

4. The hierarchical forwarding method for vehicle-road-cloud data based on edge nodes as described in claim 3, characterized in that, The results of the vehicle-road-cloud data analysis include: The feature extraction module and data analysis module are invoked through the N edge computing nodes; The feature extraction module is used to preprocess and extract features from the vehicle-road-cloud traffic data to obtain a vehicle-road-cloud traffic-related feature set. Based on the data analysis module, data task analysis is performed on the vehicle-road-cloud traffic association feature set to obtain vehicle-road-cloud data analysis results.

5. The method for hierarchical forwarding of vehicle-road-cloud data based on edge nodes as described in claim 4, characterized in that, The data analysis module is constructed, including: Data task parsing is performed on the traffic system analysis objectives to obtain a data analysis task list; Collect historical datasets of traffic vehicle-road cloud, match and associate each analysis task in the data analysis task list with the historical datasets of traffic vehicle-road cloud, and obtain the vehicle-road cloud task-associated dataset. The vehicle-road-cloud task-related datasets are analyzed, trained, and integrated to construct a data analysis module, which is then stored in the N edge computing nodes.

6. The hierarchical forwarding method for vehicle-road-cloud data based on edge nodes as described in claim 1, characterized in that, Obtain vehicle-road-cloud graded data, including: Construct data grading dimensions, which include urgency, real-time requirements, and data value density; The vehicle-road-cloud data analysis results are evaluated in a multi-dimensional manner according to the data classification dimensions to obtain the vehicle-road-cloud data classification results. Based on the classification results of the vehicle-road-cloud data, the vehicle-road-cloud traffic data is classified to obtain vehicle-road-cloud classified data.

7. The hierarchical forwarding method for vehicle-road-cloud data based on edge nodes as described in claim 1, characterized in that, Determine the target data tiered forwarding strategy, including: Based on the vehicle-road-cloud data forwarding strategy library, the data at each level in the vehicle-road-cloud hierarchical data is matched and analyzed to obtain the data hierarchical matching strategy. Monitor real-time network conditions and system load, and dynamically adjust the data classification matching strategy based on the real-time network conditions and system load to determine the target data classification forwarding strategy.

8. A vehicle-road-cloud data hierarchical forwarding system based on edge nodes, characterized in that, The step of implementing the edge node-based vehicle-road-cloud data hierarchical forwarding method according to any one of claims 1 to 7, wherein the edge node-based vehicle-road-cloud data hierarchical forwarding system comprises: The data acquisition and transmission module is used to collect vehicle-road-cloud traffic data in real time through the vehicle-road-cloud integrated system, and deploy N edge computing nodes on the roadside to map and transmit the vehicle-road-cloud traffic data to the N edge computing nodes; The feature extraction and data analysis module is used to perform feature extraction and data analysis on the vehicle-road-cloud traffic data based on the N edge computing nodes, and obtain the vehicle-road-cloud data analysis results. The data classification processing module is used to classify the vehicle-road-cloud traffic data according to the analysis results of the vehicle-road-cloud data to obtain vehicle-road-cloud classified data. The data differentiation analysis module is used to construct a vehicle-road-cloud data forwarding strategy library, perform differentiation analysis on the vehicle-road-cloud graded data based on the vehicle-road-cloud data forwarding strategy library, determine the target data graded forwarding strategy, and perform graded forwarding control on the vehicle-road-cloud traffic data through the target data graded forwarding strategy.

9. An electronic device, characterized in that, include: At least one processor; A memory that is communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the steps of the edge node-based vehicle-road-cloud data hierarchical forwarding method according to any one of claims 1 to 7.

Images