Cross-regional vehicle-mounted network switching method and system based on digital twinning and medium

Abstract

The invention discloses a cross-regional vehicle-mounted network switching method and system based on digital twinning, and a medium. The method comprises the following steps: constructing a cloud side end digital twinning network architecture; acquiring vehicle state data and traffic environment sensing data through the terminal sensing layer; carrying out data aggregation on the vehicle state data and the traffic environment perception data through edge computing nodes to obtain a local digital twinborn model; meteorological data, operator base station data and a local digital twinborn model are integrated through a cloud digital twinborn center to obtain a global digital twinborn model, and the traffic congestion condition and the network congestion condition of each road network region in a future time period are predicted according to the global digital twinborn model. Determining a network switching strategy according to the traffic congestion condition and the network congestion condition; and performing a network switching decision on the target vehicle through the edge computing node. According to the method, the timeliness and accuracy of cross-regional vehicle-mounted network switching are improved, so that the stability and reliability of the vehicle-mounted network are improved, and the method can be applied to the technical field of the Internet of Vehicles.

Description

Technical Field

[0001] The present invention relates to the field of vehicle networking technology, and in particular to a method, system, and medium for switching cross-regional vehicle networks based on digital twins. Background Art

[0002] Currently, the network switching process when a vehicle travels across regions is as follows: the on-board terminal continuously monitors the signal strength of the current base station, and triggers a switching evaluation when the RSRP (reference signal received power) is lower than the threshold; the on-board terminal first disconnects from the original base station, and then establishes a connection with the new base station.

[0003] However, existing cross-regional vehicle network switching solutions have the following disadvantages:

[0004] 1) Communication Interruption: During the handover process, the connection to the original base station must be terminated before the connection to the new one is established. This "disconnect first, then handover" process results in a brief interruption in communication. This can still affect call quality to a certain extent, especially in real-time communications or applications that require high data continuity.

[0005] 2) Risk of dropped calls: During the handover process, interference or unstable signal quality may cause handover failure, resulting in dropped calls. This is particularly noticeable in high-speed scenarios such as highways.

[0006] 3) Ping-Pong Effect: When a vehicle-mounted terminal is at the intersection of two base station coverage areas, it may switch back and forth between the two base stations due to fluctuations in signal strength. This not only increases the burden on the switching system but also further increases the possibility of dropped calls.

[0007] 4) Resource Occupancy and Increased Load: The handover process requires additional channel and signaling resources, increasing the network load. Frequent handovers can also impact network stability and reliability.

[0008] The above problems need to be solved urgently. Summary of the Invention

[0009] The purpose of the present invention is to solve one of the technical problems existing in the prior art to at least a certain extent.

[0010] To this end, an object of an embodiment of the present invention is to provide a cross-regional vehicle network switching method based on digital twins, which improves the timeliness and accuracy of cross-regional vehicle network switching, thereby improving the stability and reliability of the vehicle network.

[0011] Another object of an embodiment of the present invention is to provide a cross-regional vehicle network switching system based on digital twins.

[0012] In order to achieve the above technical objectives, the technical solutions adopted by the embodiments of the present invention include:

[0013] In a first aspect, an embodiment of the present invention provides a cross-regional vehicle network switching method based on digital twins, comprising the following steps:

[0014] A cloud-edge digital twin network architecture is constructed based on a terminal perception layer, edge computing nodes, and a cloud-based digital twin center; vehicle status data and traffic environment perception data of a target vehicle are obtained through the terminal perception layer, and the vehicle status data and traffic environment perception data are transmitted to the edge computing nodes in the corresponding area;

[0015] Aggregating the vehicle status data and the traffic environment perception data through the edge computing node to obtain a local digital twin model, and uploading the local digital twin model to the cloud digital twin center;

[0016] The cloud-based digital twin center integrates meteorological data, operator base station data, and the local digital twin model to obtain a global digital twin model, predicts traffic congestion and network congestion conditions in each road network area in the future period based on the global digital twin model, determines a network switching strategy based on the traffic congestion and network congestion conditions, and sends the network switching strategy to the edge computing node;

[0017] The edge computing node makes a network switching decision for the target vehicle according to the network switching strategy, and controls the target vehicle to switch to a cross-regional vehicle network according to the decision result.

[0018] Furthermore, in one embodiment of the present invention, the terminal perception layer includes an on-board terminal installed on the target vehicle and a roadside unit installed on both sides of the road. The edge computing node is set in each road network area, and the cloud-based digital twin center is connected to the meteorological monitoring system and the operator service system.

[0019] Furthermore, in one embodiment of the present invention, the traffic environment perception data includes vehicle-side environment perception data and roadside environment perception data. The acquiring of the vehicle status data and traffic environment perception data of the target vehicle through the terminal perception layer and the transmission of the vehicle status data and the traffic environment perception data to the edge computing node in the corresponding area specifically includes:

[0020] Collecting the vehicle status data, the vehicle-side environment perception data and the vehicle positioning information of the corresponding target vehicle through the vehicle-mounted terminal;

[0021] Determine the edge computing node in the corresponding area according to the vehicle positioning information, and transmit the vehicle status data and the vehicle-side environment perception data to the edge computing node in the corresponding area;

[0022] collecting the roadside environment perception data through the roadside unit;

[0023] The edge computing node in the corresponding area is determined according to the geographical location of the roadside unit, and the roadside environment perception data is transmitted to the edge computing node in the corresponding area.

[0024] Furthermore, in one embodiment of the present invention, the step of aggregating the vehicle status data and the traffic environment perception data by the edge computing node to obtain a local digital twin model specifically includes:

[0025] Acquire the vehicle status data and the traffic environment perception data in the corresponding area through the edge computing node;

[0026] Performing spatiotemporal alignment and coordinate system conversion on the vehicle state data and the traffic environment perception data, and extracting vehicle entities, pedestrian entities, traffic signal entities, and roadblock entities;

[0027] The vehicle entity, the pedestrian entity, the traffic signal entity, and the roadblock entity are abstracted into nodes, and the interaction relationship between each node is modeled through a graph neural network to obtain a spatiotemporal relationship graph;

[0028] Obtain the static road network topology of the corresponding area, fuse the spatiotemporal relationship map and the static road network topology, and obtain the local digital twin model of the corresponding area.

[0029] Furthermore, in one embodiment of the present invention, the cloud-based digital twin center integrates meteorological data, operator base station data, and the local digital twin model to obtain a global digital twin model, which specifically includes:

[0030] Obtaining the local digital twin models of each road network area through the cloud digital twin center, and aggregating the local digital twin models according to the global road network topology to obtain a road network digital twin model;

[0031] The meteorological data and the operator base station data are obtained through the cloud-based digital twin center, and the meteorological data, the operator base station data and the road network digital twin model are integrated to obtain the global digital twin model.

[0032] Furthermore, in one embodiment of the present invention, predicting the traffic congestion and network congestion conditions of each road network area in the future period based on the global digital twin model specifically includes:

[0033] Extracting traffic flow data, road condition data, network load data, and device connection status for each road network area based on the global digital twin model;

[0034] Inputting the traffic flow data and the road condition data into a pre-trained traffic congestion prediction model to obtain the traffic congestion situation of each road network area in the future period;

[0035] The network congestion situation of each road network area in the future period is predicted based on the traffic congestion situation, the network load data and the device connection status.

[0036] Furthermore, in one embodiment of the present invention, determining a network switching strategy based on the traffic congestion situation and the network congestion situation specifically includes:

[0037] Injecting the traffic congestion situation and the network congestion situation into the global digital twin model to obtain a digital twin simulation environment;

[0038] Based on the digital twin simulation environment, multiple regional agents and multiple base station agents are determined, and the network switching strategy for multi-network regional collaboration is obtained through multi-agent reinforcement learning based on the regional agents and the base station agents.

[0039] Furthermore, in one embodiment of the present invention, making a network switching decision for the target vehicle according to the network switching strategy, and controlling the target vehicle to perform cross-regional vehicle network switching according to the decision result, specifically includes:

[0040] Determine the target coverage range of each communication base station and the target road network area of each edge computing node according to the network switching strategy;

[0041] Determine the current communication base station and current edge computing node to which the target vehicle is currently connected, and determine, based on the vehicle status data, whether the target vehicle has entered the target coverage range of other communication base stations and whether it has entered the target road network area of other edge computing nodes;

[0042] When the target vehicle enters the target coverage range of another communication base station, a target communication link is established between the target vehicle and the other communication base station, and dual-link parallel transmission is performed based on the target communication link and the original communication link of the target vehicle until it is confirmed that the network performance of the target communication link meets expectations, and the target vehicle is switched to the target communication link;

[0043] When the target vehicle enters the target road network area of another edge computing node, the computing task of the target vehicle being executed on the current edge computing node is migrated to the other edge computing node for continued execution.

[0044] In a second aspect, an embodiment of the present invention provides a cross-regional vehicle network switching system based on digital twins, including a terminal perception layer, an edge computing node, and a cloud-based digital twin center. The terminal perception layer, the edge computing node, and the cloud-based digital twin center constitute a cloud-edge-end digital twin network architecture, wherein:

[0045] The terminal perception layer is configured to obtain vehicle status data and traffic environment perception data of the target vehicle, and transmit the vehicle status data and the traffic environment perception data to the edge computing node of the corresponding area;

[0046] The edge computing node is configured to perform data aggregation on the vehicle status data and the traffic environment perception data to obtain a local digital twin model, and upload the local digital twin model to the cloud digital twin center, and make a network switching decision for the target vehicle according to the network switching strategy issued by the cloud digital twin center, and control the target vehicle to perform cross-regional vehicle network switching according to the decision result;

[0047] The cloud-based digital twin center is configured to integrate meteorological data, operator base station data, and the local digital twin model to obtain a global digital twin model, and predict the traffic congestion and network congestion conditions of each road network area in the future period based on the global digital twin model. The network switching strategy is determined based on the traffic congestion and the network congestion conditions, and the network switching strategy is sent to the edge computing node.

[0048] In a third aspect, an embodiment of the present invention provides a cross-regional vehicle network switching device based on digital twins, including:

[0049] at least one processor;

[0050] at least one memory for storing at least one program;

[0051] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned cross-regional vehicle network switching method based on digital twins.

[0052] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, which stores a program executable by a processor. When the program is executed by the processor, it is used to execute the above-mentioned cross-regional vehicle network switching method based on digital twins.

[0053] The advantages and benefits of the present invention will be described in part in the following description and will become apparent from the following description or learned through practice of the present invention:

[0054] The embodiment of the present invention constructs a cloud-edge digital twin network architecture based on the terminal perception layer, edge computing nodes and cloud-based digital twin center, obtains the vehicle status data and traffic environment perception data of the target vehicle through the terminal perception layer, and transmits the vehicle status data and traffic environment perception data to the edge computing nodes in the corresponding area, aggregates the vehicle status data and traffic environment perception data through the edge computing nodes to obtain a local digital twin model, and uploads the local digital twin model to the cloud-based digital twin center, integrates meteorological data, operator base station data and the local digital twin model through the cloud-based digital twin center to obtain a global digital twin model, predicts the traffic congestion and network congestion conditions of each road network area in the future period according to the global digital twin model, determines the network switching strategy according to the traffic congestion and network congestion conditions, and sends the network switching strategy to the edge computing node, makes a network switching decision for the target vehicle according to the network switching strategy through the edge computing node, and controls the target vehicle to perform cross-regional vehicle network switching according to the decision result. The embodiment of the present invention integrates vehicle status data, traffic environment perception data, meteorological data and operator base station data based on the cloud-edge digital twin network architecture to build a global digital twin model, realizing high-precision road network mapping and multi-physical field coupling simulation. According to the global digital twin model, the traffic congestion and network congestion conditions of each road network area in the future period are predicted, so that the corresponding network switching strategy can be deployed in advance according to the predicted traffic congestion and network congestion conditions, and the network switching decision is made collaboratively through each edge computing node, so as to control the target vehicle to perform cross-regional vehicle network switching in advance before entering a weak network environment or network congestion, thereby improving the timeliness and accuracy of cross-regional vehicle network switching, thereby improving the stability and reliability of the vehicle network. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following introduction is made to the drawings required for use in the embodiments of the present invention. It should be understood that the drawings introduced below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative work.

[0056] Figure 1 A flowchart of the steps of a cross-regional vehicle network switching method based on digital twins provided in an embodiment of the present invention;

[0057] Figure 2 A structural block diagram of a cross-regional vehicle network switching system based on digital twins provided in an embodiment of the present invention;

[0058] Figure 3 A structural block diagram of a cross-regional vehicle network switching device based on digital twins provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention and are not to be construed as limiting the present invention. The step numbers in the following embodiments are provided for ease of explanation only and do not limit the order of the steps. The order of execution of the steps in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0060] In the description of the present invention, "a plurality" means two or more. The terms "first" and "second" are used solely to distinguish technical features and are not to be construed as indicating or implying relative importance, or as implicitly indicating the number of the indicated technical features, or as implicitly indicating the order of the indicated technical features. Furthermore, unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art.

[0061] Reference Figure 1 , an embodiment of the present invention provides a cross-regional vehicle network switching method based on digital twins, which specifically includes the following steps:

[0062] S101. Build a cloud-edge-device digital twin network architecture based on the terminal perception layer, edge computing nodes, and cloud-based digital twin center;

[0063] S102. Acquire vehicle status data and traffic environment perception data of the target vehicle through the terminal perception layer, and transmit the vehicle status data and traffic environment perception data to the edge computing node in the corresponding area;

[0064] S103: Aggregate vehicle status data and traffic environment perception data through edge computing nodes to obtain a local digital twin model, and upload the local digital twin model to a cloud digital twin center;

[0065] S104. Integrate meteorological data, operator base station data, and local digital twin models through the cloud digital twin center to obtain a global digital twin model. Use the global digital twin model to predict traffic congestion and network congestion conditions in each road network area in the future period. Determine a network switching strategy based on the traffic congestion and network congestion conditions, and send the network switching strategy to the edge computing node.

[0066] S105. Make a network switching decision for the target vehicle according to the network switching strategy through the edge computing node, and control the target vehicle to switch the cross-region vehicle network according to the decision result.

[0067] Specifically, the embodiment of the present invention integrates vehicle status data, traffic environment perception data, meteorological data and operator base station data based on the cloud-edge digital twin network architecture to build a global digital twin model, realizing high-precision road network mapping and multi-physical field coupling simulation. According to the global digital twin model, the traffic congestion and network congestion conditions of each road network area in the future period are predicted, so that the corresponding network switching strategy can be deployed in advance according to the predicted traffic congestion and network congestion conditions, and the network switching decision is made collaboratively through each edge computing node, so as to control the target vehicle to perform cross-regional vehicle network switching in advance before entering a weak network environment or network congestion, thereby improving the timeliness and accuracy of cross-regional vehicle network switching, thereby improving the stability and reliability of the vehicle network.

[0068] As a further optional implementation, the terminal perception layer includes an on-board terminal installed on the target vehicle and roadside units installed on both sides of the road. Edge computing nodes are set in each road network area, and the cloud-based digital twin center is connected to the meteorological monitoring system and operator service system.

[0069] Specifically, the embodiment of the present invention achieves global perception, collaborative decision-making, and dynamic optimization by building a three-level digital twin network architecture of "cloud-edge-end", as follows:

[0070] Cloud-based digital twin center: Deploy the CIM (city information) platform to integrate multi-source data such as transportation, meteorology, and operator base stations to build a digital twin foundation for the road network.

[0071] Edge computing nodes: Deploy MEC (distributed computing) servers at key locations such as highway service areas and tunnel entrances to achieve localized data processing and business diversion.

[0072] Terminal perception layer: The on-board terminal and roadside unit integrate communication modules, high-precision positioning units and multi-modal sensors to collect vehicle status and surrounding environment data in real time.

[0073] As an optional implementation, the traffic environment perception data includes vehicle-side environment perception data and roadside environment perception data. The vehicle status data and traffic environment perception data of the target vehicle are obtained through the terminal perception layer, and the vehicle status data and traffic environment perception data are transmitted to the edge computing node in the corresponding area, which specifically includes:

[0074] S1021. Collect vehicle status data, vehicle-side environment perception data, and vehicle positioning information of the corresponding target vehicle through the vehicle-mounted terminal;

[0075] S1022. Determine an edge computing node in a corresponding area based on the vehicle positioning information, and transmit the vehicle status data and vehicle-side environment perception data to the edge computing node in the corresponding area;

[0076] S1023. Collect roadside environment perception data through the roadside unit;

[0077] S1024. Determine the edge computing node in the corresponding area according to the geographical location of the roadside unit, and transmit the roadside environment perception data to the edge computing node in the corresponding area.

[0078] Specifically, the on-board terminal collects vehicle status data (speed, acceleration, steering angle), sensor data (camera, radar, lidar), communication parameters (RSRP, SINR), and positioning information (GPS / Beidou) in real time; performs preliminary cleaning, feature extraction, and compression on the raw data to reduce data redundancy; determines the edge computing node corresponding to the target vehicle's current road network area based on the positioning information, and transmits the vehicle status data and vehicle-side environmental perception data to the edge computing node. In addition, the on-board terminal is also used to perform low-latency local decision-making (such as emergency braking and lane keeping) and cache key data.

[0079] Specifically, the roadside unit (RSU) collects roadside environmental perception data, such as traffic light status at intersections, pedestrian detection, and road construction. Based on the RSU's geographic location, it determines the edge computing node corresponding to the road network area in which it is located and transmits this roadside environmental perception data to that edge computing node. Furthermore, the RSU is used to broadcast real-time road condition information (such as speed limits and accident locations) to vehicles, assist edge computing nodes in making handover decisions (such as pushing base station information in advance), and deploy a digital twin pre-handover system (RSU sends a list of neighboring cells in advance).

[0080] As an optional implementation method, the edge computing node aggregates the vehicle status data and traffic environment perception data to obtain a local digital twin model, which specifically includes:

[0081] S1031. Obtain vehicle status data and traffic environment perception data in the corresponding area through the edge computing node;

[0082] S1032: Performing spatiotemporal alignment and coordinate system conversion on the vehicle status data and the traffic environment perception data, and extracting vehicle entities, pedestrian entities, traffic signal entities, and roadblock entities;

[0083] S1033. Abstracting vehicle entities, pedestrian entities, traffic signal entities, and roadblock entities into nodes, and modeling the interaction relationships between the nodes using a graph neural network to obtain a spatiotemporal relationship graph;

[0084] S1034. Obtain the static road network topology of the corresponding area, fuse the spatiotemporal relationship map and the static road network topology, and obtain a local digital twin model of the corresponding area.

[0085] Specifically, edge computing nodes collect real-time data from multiple vehicles, roadside sensors (RSUs), and base stations within the corresponding road network area to build a local digital twin model. The specific process is as follows:

[0086] 1. Real-time data collection

[0087] Edge computing nodes use integrated data acquisition devices to obtain real-time vehicle status data (such as speed, acceleration, and driving direction) and traffic environment perception data (such as vehicle location, pedestrian location, traffic signal status, and roadblock location) within the corresponding area. The collected data is temporarily stored in the local storage of the edge computing node, awaiting further processing.

[0088] 2. Data processing and entity extraction

[0089] Spatiotemporal alignment: Because different data collection devices may have different sampling frequencies and timestamps, the collected vehicle status data and traffic environment perception data need to be spatiotemporally aligned. This can be achieved by using a time synchronization protocol (such as NTP) to ensure that each device maintains consistent time. Data collected by different devices can then be matched and aligned in time and space based on the data's timestamp and sampling frequency.

[0090] Coordinate system conversion: Different data acquisition devices may use different coordinate systems. To unify data representation, coordinate system conversion is required. Typically, all data is converted to a unified global coordinate system, such as a geographic coordinate system (WGS84) or a local Cartesian coordinate system. Data can be converted from the device coordinate system to the global coordinate system using a coordinate transformation matrix based on the device's installation location and posture information.

[0091] Entity extraction: Computer vision and machine learning algorithms are used to process data after spatiotemporal alignment and coordinate system conversion to extract vehicle entities, pedestrian entities, traffic signal entities, and roadblock entities. For example, object detection algorithms (such as YOLO and Faster R-CNN) are used to process camera images to identify objects such as vehicles, pedestrians, and traffic lights. Cluster analysis is performed on radar data to extract the location and shape of obstacles such as roadblocks.

[0092] 3. Graph Neural Network Modeling

[0093] Node abstraction: The extracted vehicle entities, pedestrian entities, traffic signal entities, and roadblock entities are abstracted into nodes in the graph neural network. Each node contains attribute information of the entity, such as location, speed, and status.

[0094] Edge Definition: Edges in a graph neural network are defined based on the interactions between entities. For example, vehicles may interact with each other, such as following and overtaking; vehicles and pedestrians may interact with each other, such as yielding and collision risk; and vehicles and traffic signals may interact with each other, such as following the signal. Edge weights can be defined based on the strength and frequency of interactions.

[0095] Graph Neural Network Training: Graph neural network models (such as Graph Convolutional Network (GCN) and Graph Attention Network (GAT)) are used to model the interactions between nodes. Node attribute information and edge connectivity are used as input. By training the graph neural network model, the spatiotemporal relationships between nodes are learned, resulting in a spatiotemporal relationship graph. During training, historical data can be used for supervised learning to optimize model parameters.

[0096] 4. Data fusion and digital twin model construction

[0097] Static road network topology acquisition: Using a geographic information system (GIS) or high-precision maps, obtain static road network topology information for the corresponding area, including road connectivity, lane information, and intersection information. Static road network topology information is typically represented as a graph, where nodes represent intersections and edges represent road segments.

[0098] Data fusion: The resulting spatiotemporal graph is integrated with the static road network topology. Nodes and edges in the spatiotemporal graph can be mapped onto the static road network topology, spatially matching the two. Dynamic information in the spatiotemporal graph (such as the motion of vehicles and pedestrians) is combined with static information in the static road network topology (such as road attributes) to form a unified model encompassing both dynamic and static information.

[0099] Local digital twin model generation: Based on the fused data, a local digital twin model of the corresponding area is generated. This model can reflect the traffic status within the corresponding area in real time, including vehicle trajectories, pedestrian movement paths, and traffic signal changes. Through analysis and prediction of the local digital twin model, decision support can be provided for applications such as traffic management and intelligent driving.

[0100] As an optional implementation method, a cloud-based digital twin center integrates meteorological data, operator base station data, and local digital twin models to obtain a global digital twin model, which specifically includes:

[0101] S1041. Obtain local digital twin models of each road network area through the cloud digital twin center, and aggregate the local digital twin models according to the global road network topology to obtain a road network digital twin model;

[0102] S1042. Obtain meteorological data and operator base station data through the cloud-based digital twin center, integrate the meteorological data, operator base station data, and the road network digital twin model to obtain a global digital twin model.

[0103] Specifically, the cloud-based digital twin center integrates data such as traffic flow, weather, and operator base stations to build a global digital twin model of the entire road network. The specific process is as follows:

[0104] 1. Aggregate local digital twin models into a road network digital twin model

[0105] 1) Data acquisition and preprocessing

[0106] Cloud center call: Obtain local digital twin models from each edge computing node through the cloud API interface. The models are usually stored in a standardized format (such as glTF, BIM or point cloud structure).

[0107] Data verification and alignment: Verify the spatiotemporal consistency of the model (such as timestamp matching and coordinate system 1), and fill in missing or abnormal data through interpolation or historical data to ensure the accuracy of subsequent aggregation.

[0108] 2) Model aggregation based on hierarchical structure

[0109] Hierarchical feature extraction: Using a hierarchical method similar to PointNet++, the local digital twin model is first divided into regions (such as road segments or grids) to extract key information such as local topological structure and traffic flow characteristics.

[0110] Parameter aggregation strategy:

[0111] (1) Federated weighted averaging: According to the amount or quality of data in the regional model (such as sensor coverage), different weights are assigned for parameter fusion (similar to the FedAvg algorithm).

[0112] (2) Attention mechanism optimization: Graph Attention Network (GAT) is used to identify key areas (such as transportation hubs) and dynamically adjust aggregation weights to improve global model accuracy.

[0113] 3) Static road network topology integration

[0114] GIS data mapping: Spatial matching of the aggregated dynamic model with the static road network (such as lane lines and traffic signs) in the high-precision map is performed, and geometric alignment is achieved using the coordinate transformation matrix.

[0115] Topology relationship verification: Graph Neural Network (GCN) is used to detect the logical consistency between the dynamic model and the static road network (such as whether there is a conflict in road connectivity) and automatically correct abnormal nodes.

[0116] 2. Multi-source data fusion to generate a global digital twin model

[0117] 1) Meteorological data fusion

[0118] Spatiotemporal interpolation processing: Discrete meteorological station data (such as rainfall and visibility) are used to generate a continuously distributed meteorological field through the Kriging interpolation algorithm, and are aligned with the time axis of the road network model.

[0119] Impact modeling: Use CNN to extract meteorological features (such as rainstorm areas) and simulate their impact on traffic flow (such as vehicle speed attenuation model and accident risk prediction) through a physical engine.

[0120] 2) Operator base station data integration

[0121] Network traffic density heat map generation: Analyze base station signaling data and combine it with a spatiotemporal clustering algorithm to generate a network traffic distribution heat map, identifying key monitoring areas (such as those around business districts).

[0122] Communication quality mapping: Overlay base station signal strength data with road network models to identify the impact of network blind spots on IoV communications and optimize edge node deployment.

[0123] 3) Multimodal data fusion

[0124] Feature-level fusion: The Transformer architecture is used to encode heterogeneous data such as meteorological, communication, and traffic flow into a unified feature vector to capture cross-domain correlations. A federated learning framework is used to achieve secure aggregation of cross-regional data, ensuring privacy protection while completing the spatiotemporal alignment of multi-source data.

[0125] Dynamic update mechanism: Using the federated learning framework, each local digital twin model regularly uploads incremental data to the cloud, triggering global model fine-tuning to ensure real-time performance.

[0126] As an optional implementation method, traffic congestion and network congestion conditions in each road network area in the future period are predicted based on the global digital twin model, which specifically includes:

[0127] S1043. Extracting traffic flow data, road condition data, network load data, and device connection status for each road network area based on the global digital twin model;

[0128] S1044: Input the traffic flow data and road condition data into a pre-trained traffic congestion prediction model to obtain traffic congestion conditions in each road network area in the future period;

[0129] S1045. Predict the network congestion situation of each road network area in the future period based on the traffic congestion situation, network load data, and device connection status.

[0130] Specifically, the cloud-based digital twin center predicts traffic congestion and network congestion in each road network area in the future based on the global digital twin model. The specific process is as follows:

[0131] 1. Global data extraction and preprocessing

[0132] 1) Data collection and fusion: Extract structured data such as traffic flow (such as vehicle volume and average speed), road conditions (such as road surface moisture and construction status), network load (bandwidth utilization and data transmission delay), and device status (camera / radar online rate and number of base station connections) for each road network area from the global digital twin model.

[0133] 2) Data cleaning and feature engineering: A box plot algorithm is used to detect abnormal traffic data (such as sudden zero traffic or extremely high peaks), and missing values are repaired using historical mean interpolation or adjacent area data compensation mechanisms. Feature encoding is implemented for road condition data, discrete events (such as construction and accidents) are converted into one-hot vectors, and continuous variables (such as visibility) are normalized using Z-scores.

[0134] 2. Traffic congestion prediction model deployment

[0135] 1) Dynamic prediction algorithm architecture: This algorithm uses an LSTM-GRU hybrid neural network as the core model. The input layer receives traffic flow (time series) and road conditions (static features) and outputs the congestion probability at the granularity level of the next hour. An attention mechanism is introduced to enhance the ability to identify key road sections, for example, by assigning higher weights to areas around schools or business districts.

[0136] 2) Real-time model update mechanism: A dynamic training set with a sliding time window (e.g., 30 days) is established, model parameters are updated daily through incremental learning, and online performance evaluation is performed using the F1-score and MAE metrics. Transfer learning technology is integrated to migrate the parameters of the trunk road network training model to newly built areas to address the cold start problem.

[0137] 3. Joint prediction of network congestion

[0138] 1) Multimodal Data Association Analysis: A causal inference graph model is constructed, using the traffic congestion index, 4G / 5G base station load rate, and edge computing node connection status as nodes. A Bayesian network is used to calculate the joint probability distribution. For example, when the congestion level in a certain area exceeds Level 3 and the base station load exceeds 80%, the network congestion warning threshold is triggered.

[0139] 2) Spatiotemporal Propagation Prediction: Graphs AGE graph neural networks are used to simulate the diffusion path of network congestion. Input dimensions include traffic-derived data (navigation app request volume, video surveillance bitrate) and network performance data (TCP retransmission rate, DNS response latency). The system outputs a 15-minute-level network health heat map, identifying base stations or edge nodes that require priority expansion.

[0140] As an optional implementation, a network switching strategy is determined based on traffic congestion and network congestion, which specifically includes:

[0141] S1046. Injecting traffic congestion and network congestion into the global digital twin model to obtain a digital twin simulation environment;

[0142] S1047. Determine multiple regional intelligent agents and multiple base station intelligent agents based on the digital twin simulation environment, and obtain a network switching strategy for multi-network regional collaboration through multi-agent reinforcement learning based on the regional intelligent agents and the base station intelligent agents.

[0143] Specifically, the cloud-based digital twin model determines the network switching strategy to deal with traffic congestion and network congestion based on multi-agent reinforcement learning. The specific process is as follows:

[0144] 1. Construction of digital twin simulation environment

[0145] 1) Data injection and dynamic modeling

[0146] Inject real-time traffic congestion index (such as average speed drop rate of road sections and queue length) and network congestion data (base station bandwidth utilization and TCP retransmission rate) into the global digital twin model through cloud APIs;

[0147] The SUMO microscopic traffic simulation engine is used to simulate vehicle trajectories, and the NS-3 network simulator is used to build a vehicle network communication link model to achieve bidirectional mapping between physical and virtual space.

[0148] 2) Multimodal data fusion

[0149] Establish a spatiotemporal correlation matrix, overlay the traffic congestion heat map with the 5G base station coverage map, and identify the spatial coupling relationship between congested areas and network bottlenecks (e.g., simultaneous traffic stagnation and base station overload around a business district);

[0150] GraphSAGE graph neural network is used to extract cross-domain features and generate dynamic twins that include the joint state of transportation and communication.

[0151] 2. Multi-agent system construction

[0152] 1) Agent definition and initialization

[0153] Regional agent: One agent is deployed in each road network area, and the observation space includes traffic dimension (average vehicle speed in the area, phase status of traffic lights) and communication dimension (edge node computing load, number of base station RRC connections).

[0154] Base station agent: One agent is deployed at each 5G base station, and its action space includes frequency band switching (2.6GHz / 4.9GHz) and MIMO beamforming angle adjustment.

[0155] 2) Environmental interaction mechanism design

[0156] Establish a hierarchical communication protocol: regional agents broadcast macro strategies through V2X, and base station agents feedback micro states based on the IEEE802.11p protocol.

[0157] Design joint reward function: R = α (1-D current / D threshold )+β·log(1+BW free / BW total ), where D current is the regional vehicle delay time, D threshold is the delay time threshold, BW free is the remaining bandwidth of the base station, BW total is the total bandwidth of the base station.

[0158] 3. Multi-agent reinforcement learning training

[0159] 1) Distributed training architecture

[0160] Adopting the MADDPG (Multi-Agent Deep Deterministic Policy Gradient) framework, the central critic network receives state information from the entire region, and each actor network makes independent decisions.

[0161] A federated learning mechanism is introduced, where each regional model uploads its gradient to the cloud for aggregation after local training, solving data privacy and communication overhead issues.

[0162] 2) Learning strategies

[0163] Phase 1: Train the frequency band switching strategy of the base station agent in a static road network scenario and freeze the network parameters after convergence.

[0164] Phase 2: Release the action space of regional intelligent agents, jointly optimize traffic light timing and base station resource scheduling, and focus on learning congestion-congestion concurrent scenarios through the PER (Prioritized Experience Replay) algorithm.

[0165] 4. Dynamic Strategy Optimization and Verification

[0166] 1) Online strategy deployment

[0167] The trained network switching strategy is run in a digital twin environment and sent to edge computing nodes and base station SDN controllers via the OPC UA protocol. An online learning mechanism with a sliding time window (30 minutes) is established, triggering model fine-tuning when the regional traffic flow mutation exceeds 20%.

[0168] 2) Multi-dimensional verification method

[0169] Traffic indicators: Compare regional traffic efficiency before and after strategy implementation (TPI index improvement ≥ 15%);

[0170] Communication metrics: measuring handover latency (target ≤ 50ms) and throughput fluctuation (target ≤ 10%);

[0171] Visual monitoring: Build a 3D monitoring interface using Three.js to display the agent’s decision path and network switching hotspots in real time.

[0172] As an optional implementation, a network switching decision is made for the target vehicle according to the network switching strategy, and the target vehicle is controlled to perform cross-regional vehicle network switching according to the decision result, which specifically includes:

[0173] S1051. Determine the target coverage range of each communication base station and the target road network area of each edge computing node according to the network switching strategy;

[0174] S1052. Determine the current communication base station and the current edge computing node to which the target vehicle is currently connected, and determine, based on the vehicle status data, whether the target vehicle has entered the target coverage range of other communication base stations and whether it has entered the target road network area of other edge computing nodes;

[0175] S1053. When the target vehicle enters the target coverage range of another communication base station, a target communication link is established between the target vehicle and the other communication base station, and dual-link parallel transmission is performed based on the target communication link and the original communication link of the target vehicle until it is confirmed that the network performance of the target communication link meets expectations, and the target vehicle is switched to the target communication link.

[0176] S1054: When the target vehicle enters the target road network area of another edge computing node, the computing task of the target vehicle currently being executed on the edge computing node is migrated to the other edge computing node for continued execution.

[0177] Specifically, each edge computing node of the embodiment of the present invention cooperates with the adjacent edge computing nodes to make cross-regional network switching decisions. First, the target coverage range of each communication base station and the target road network area of each edge computing node are re-determined according to the network switching strategy obtained in the above steps. The target coverage range is used to determine the target vehicle that needs to be switched to the corresponding communication base station, and the target road network area is used to determine the target vehicle that needs to be switched to the corresponding edge computing node; then, the current communication base station and the current edge computing node to which the target vehicle is currently connected are determined, and based on the real-time data of the target vehicle, it is determined whether it enters the target coverage range assigned to other communication base stations and whether it enters the target coverage range assigned to other edge computing nodes. The target road network area of the point; if the target vehicle enters the target coverage range of other communication base stations, it means that the target vehicle needs to switch to the other communication base station. At this time, a target communication link between the target vehicle and the other communication base stations is established, and dual-link parallel transmission is performed based on the target communication link and the original communication link of the target vehicle. The target vehicle is switched to the target communication link after it is confirmed that the network performance of the target communication link meets the expected standard; if the target vehicle enters the target road network area of other edge computing nodes, it means that the target vehicle needs to switch to the other edge computing point. At this time, the computing task of the target vehicle being executed on the current edge computing node is migrated to the other edge computing node through containerization technology for continued execution.

[0178] It can be recognized that the embodiment of the present invention does not start network switching after the target vehicle is in a weak network environment or the network is congested, but deploys the network switching strategy in advance in the cloud based on the prediction of traffic congestion and network congestion, and redistributes the coverage area of each communication base station and the road network area corresponding to each edge computing node. At the same time, network resources are allocated and optimized for the communication base stations and edge computing nodes. In this way, as long as the target vehicle enters the target coverage area assigned to other communication base stations, network switching can be performed in advance, and the link of the original communication base station can still be used for a period of time. The stability and reliability of the target vehicle's on-board network during network switching are guaranteed through dual-link parallel transmission.

[0179] The above describes the method steps of the embodiment of the present invention. It can be understood that the embodiment of the present invention integrates vehicle status data, traffic environment perception data, meteorological data and operator base station data based on the cloud-edge digital twin network architecture to build a global digital twin model, realizing high-precision road network mapping and multi-physical field coupling simulation, and predicting the traffic congestion and network congestion of each road network area in the future period according to the global digital twin model, so that the corresponding network switching strategy can be pre-deployed according to the predicted traffic congestion and network congestion, and the network switching decision can be made collaboratively through each edge computing node, so as to control the target vehicle to switch the cross-regional vehicle network in advance before entering a weak network environment or network congestion, thereby improving the timeliness and accuracy of the cross-regional vehicle network switching, thereby improving the stability and reliability of the vehicle network.

[0180] Reference Figure 2 The embodiment of the present invention provides a cross-regional vehicle network switching system based on digital twins, including a terminal perception layer, edge computing nodes, and a cloud-based digital twin center. The terminal perception layer, edge computing nodes, and cloud-based digital twin center constitute a cloud-edge-device digital twin network architecture, wherein:

[0181] The terminal perception layer is configured to obtain vehicle status data and traffic environment perception data of the target vehicle, and transmit the vehicle status data and traffic environment perception data to the edge computing node in the corresponding area;

[0182] The edge computing node is configured to aggregate vehicle status data and traffic environment perception data to obtain a local digital twin model, upload the local digital twin model to the cloud digital twin center, and make network switching decisions for the target vehicle based on the network switching strategy issued by the cloud digital twin center. Based on the decision results, the target vehicle is controlled to switch to an on-board network across regions.

[0183] The cloud-based digital twin center is configured to integrate meteorological data, operator base station data, and local digital twin models to obtain a global digital twin model. Based on the global digital twin model, it predicts the traffic congestion and network congestion conditions of each road network area in the future period. The network switching strategy is determined based on the traffic congestion and network congestion conditions, and the network switching strategy is sent to the edge computing node.

[0184] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0185] Reference Figure 3 , an embodiment of the present invention provides a cross-regional vehicle network switching device based on digital twins, including:

[0186] at least one processor;

[0187] at least one memory for storing at least one program;

[0188] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned cross-regional vehicle network switching method based on digital twins.

[0189] The contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0190] An embodiment of the present invention also provides a computer-readable storage medium, which stores a program executable by a processor. When the program is executed by the processor, it is used to execute the above-mentioned cross-regional vehicle network switching method based on digital twins.

[0191] A computer-readable storage medium according to an embodiment of the present invention can execute a cross-regional vehicle network switching method based on digital twins provided by an embodiment of the method of the present invention, can execute any combination of implementation steps of the method embodiment, and has the corresponding functions and beneficial effects of the method.

[0192] The embodiment of the present invention also discloses a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs Figure 1 The method shown.

[0193] In some optional embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the above-mentioned boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided in an exemplary manner for the purpose of providing a more comprehensive understanding of the technology. The disclosed method is not limited to the operation and logic flow presented herein. Optional embodiments are contemplated in which the order of the various operations is changed and the sub-operations described as a part of a larger operation are performed independently.

[0194] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the above-mentioned functions and / or features can be integrated into a single physical device and / or software module, or one or more functions and / or features can be implemented in separate physical devices or software modules. It is also understood that a detailed discussion of the actual implementation of each module is not necessary for understanding the present invention. More specifically, given the properties, functions, and internal relationships of the various functional modules in the devices disclosed herein, the actual implementation of the module will be understood within the routine skills of an engineer. Therefore, a person skilled in the art can implement the present invention set forth in the claims using ordinary skills without undue experimentation. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0195] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the above methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0196] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0197] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable media on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0198] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0199] In the above description of this specification, reference to the terms "one embodiment / example," "another embodiment / example," or "certain embodiments / examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0200] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

[0201] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the above embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A cross-regional vehicle network switching method based on digital twins, characterized in that: The following steps are involved: Build a cloud-edge-device digital twin network architecture based on the terminal perception layer, edge computing nodes, and cloud-based digital twin center; Acquire vehicle status data and traffic environment perception data of the target vehicle through the terminal perception layer, and transmit the vehicle status data and the traffic environment perception data to the edge computing node in the corresponding area; Aggregating the vehicle status data and the traffic environment perception data through the edge computing node to obtain a local digital twin model, and uploading the local digital twin model to the cloud digital twin center; The cloud-based digital twin center integrates meteorological data, operator base station data, and the local digital twin model to obtain a global digital twin model, predicts traffic congestion and network congestion conditions in each road network area in the future period based on the global digital twin model, determines a network switching strategy based on the traffic congestion and network congestion conditions, and sends the network switching strategy to the edge computing node; The edge computing node makes a network switching decision for the target vehicle according to the network switching strategy, and controls the target vehicle to switch the cross-region vehicle network according to the decision result.

2. A cross-regional vehicle network switching method based on digital twins according to claim 1, characterized in that: The terminal perception layer includes an on-board terminal installed on the target vehicle and roadside units installed on both sides of the road. The edge computing nodes are set in each road network area, and the cloud-based digital twin center is connected to the meteorological monitoring system and the operator service system.

3. The cross-regional vehicle network switching method based on digital twin according to claim 2 is characterized in that: The traffic environment perception data includes vehicle-side environment perception data and roadside environment perception data. The terminal perception layer obtains the vehicle status data and traffic environment perception data of the target vehicle, and transmits the vehicle status data and traffic environment perception data to the edge computing node in the corresponding area, which specifically includes: Collecting the vehicle status data, the vehicle-side environment perception data and the vehicle positioning information of the corresponding target vehicle through the vehicle-mounted terminal; Determine the edge computing node in the corresponding area according to the vehicle positioning information, and transmit the vehicle status data and the vehicle-side environment perception data to the edge computing node in the corresponding area; collecting the roadside environment perception data through the roadside unit; The edge computing node in the corresponding area is determined according to the geographical location of the roadside unit, and the roadside environment perception data is transmitted to the edge computing node in the corresponding area.

4. The method for switching cross-regional vehicle networks based on digital twins according to claim 1 is characterized in that: The data aggregation of the vehicle status data and the traffic environment perception data by the edge computing node to obtain a local digital twin model specifically includes: Acquire the vehicle status data and the traffic environment perception data in the corresponding area through the edge computing node; Performing spatiotemporal alignment and coordinate system conversion on the vehicle state data and the traffic environment perception data, and extracting vehicle entities, pedestrian entities, traffic signal entities, and roadblock entities; The vehicle entity, the pedestrian entity, the traffic signal entity, and the roadblock entity are abstracted into nodes, and the interaction relationship between each node is modeled through a graph neural network to obtain a spatiotemporal relationship graph; Obtain the static road network topology of the corresponding area, fuse the spatiotemporal relationship map and the static road network topology, and obtain the local digital twin model of the corresponding area.

5. The method for switching cross-regional vehicle networks based on digital twins according to claim 1 is characterized in that: The cloud-based digital twin center integrates meteorological data, operator base station data, and the local digital twin model to obtain a global digital twin model, which specifically includes: Obtaining the local digital twin models of each road network area through the cloud digital twin center, and aggregating the local digital twin models according to the global road network topology to obtain a road network digital twin model; The meteorological data and the operator base station data are obtained through the cloud-based digital twin center, and the meteorological data, the operator base station data and the road network digital twin model are integrated to obtain the global digital twin model.

6. The method for switching cross-regional vehicle networks based on digital twins according to claim 1 is characterized in that: The prediction of traffic congestion and network congestion conditions in each road network area in the future period based on the global digital twin model specifically includes: Extracting traffic flow data, road condition data, network load data, and device connection status for each road network area based on the global digital twin model; Inputting the traffic flow data and the road condition data into a pre-trained traffic congestion prediction model to obtain the traffic congestion situation of each road network area in the future period; The network congestion situation of each road network area in the future period is predicted based on the traffic congestion situation, the network load data and the device connection status.

7. A cross-regional vehicle network switching method based on digital twin according to any one of claims 1 to 6, characterized in that: The determining of the network switching strategy according to the traffic congestion situation and the network congestion situation specifically includes: Injecting the traffic congestion situation and the network congestion situation into the global digital twin model to obtain a digital twin simulation environment; Based on the digital twin simulation environment, multiple regional agents and multiple base station agents are determined, and the network switching strategy for multi-network regional collaboration is obtained through multi-agent reinforcement learning based on the regional agents and the base station agents.

8. The method for switching cross-regional vehicle networks based on digital twins according to claim 7 is characterized in that: The making a network switching decision for the target vehicle according to the network switching strategy, and controlling the target vehicle to perform cross-regional vehicle network switching according to the decision result, specifically includes: Determine the target coverage range of each communication base station and the target road network area of each edge computing node according to the network switching strategy; Determine the current communication base station and current edge computing node to which the target vehicle is currently connected, and determine, based on the vehicle status data, whether the target vehicle has entered the target coverage range of other communication base stations and whether it has entered the target road network area of other edge computing nodes; When the target vehicle enters the target coverage range of another communication base station, a target communication link is established between the target vehicle and the other communication base station, and dual-link parallel transmission is performed based on the target communication link and the original communication link of the target vehicle until it is confirmed that the network performance of the target communication link meets expectations, and the target vehicle is switched to the target communication link; When the target vehicle enters the target road network area of another edge computing node, the computing task of the target vehicle being executed on the current edge computing node is migrated to the other edge computing node for continued execution.

9. A cross-regional vehicle network switching system based on digital twins, characterized in that: It includes a terminal perception layer, edge computing nodes and a cloud-based digital twin center. The terminal perception layer, the edge computing nodes and the cloud-based digital twin center constitute a cloud-edge-end digital twin network architecture, wherein: The terminal perception layer is configured to obtain vehicle status data and traffic environment perception data of the target vehicle, and transmit the vehicle status data and the traffic environment perception data to the edge computing node of the corresponding area; The edge computing node is configured to perform data aggregation on the vehicle status data and the traffic environment perception data to obtain a local digital twin model, and upload the local digital twin model to the cloud digital twin center, and make a network switching decision for the target vehicle according to the network switching strategy issued by the cloud digital twin center, and control the target vehicle to perform cross-regional vehicle network switching according to the decision result; The cloud-based digital twin center is configured to integrate meteorological data, operator base station data, and the local digital twin model to obtain a global digital twin model, and predict the traffic congestion and network congestion conditions of each road network area in the future period based on the global digital twin model. The network switching strategy is determined based on the traffic congestion and the network congestion conditions, and the network switching strategy is sent to the edge computing node.

10. A computer-readable storage medium storing a program executable by a processor, characterized in that: The program executable by the processor is used to execute a cross-regional vehicle network switching method based on digital twins as described in any one of claims 1 to 8 when executed by the processor.

Images