A V2X testing method, system, device and computer program product based on cloud platform

Abstract

The present invention discloses a V2X test method, system, device and computer program product based on a cloud platform, the method comprising: selecting a preset test scenario; the cloud platform generates a configuration information package based on the key parameters of the test scenario and the node configuration rules; the edge computing node performs initialization settings based on the configuration information package, starts data receiving and transmission services, and returns configuration confirmation information to the cloud platform; the test vehicle terminal collects vehicle data in real time through the on-board sensor and the V2X communication module according to the instructions in the configuration information package; the edge computing node pre-processes the vehicle data and sends the pre-processed vehicle data to the cloud platform; the cloud platform constructs a multi-dimensional data matrix based on the vehicle data and the test scenario data; the multi-dimensional data matrix is ​​input into the pre-trained CNN-RNN hybrid network model for analysis to obtain the factors affecting the communication quality of the V2X communication module. Through the processing scheme disclosed in the present invention, V2X can be optimized and improved in a targeted manner.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a V2X testing method, system, device and computer program product based on a cloud platform. Background Art

[0002] With the development of Internet of Vehicles technology, V2X communication has become an important means to achieve information exchange between vehicles and the surrounding environment. In order to ensure the reliability, safety and efficiency of the V2X system, it is necessary to conduct comprehensive tests on the communication performance in various scenarios.

[0003] Although the existing V2X test technology meets the basic test requirements to a certain extent, the following problems still exist:

[0004] Limited testing scenarios: Laboratory testing is difficult to simulate complex and changing traffic environments, while outdoor field testing is limited by actual conditions and cannot fully cover all possible scenarios.

[0005] Low testing efficiency: Outdoor field testing requires a lot of manpower and material resources, with a long testing cycle and high costs.

[0006] Insufficient data processing capabilities: Existing test systems have limited capabilities in data collection, transmission and analysis, making it difficult to process large-scale test data in real time.

[0007] The test results are not intuitive: due to the lack of effective visualization tools, it is difficult for users to intuitively understand the test results and to quickly identify problems and optimization solutions.

[0008] Poor flexibility: The existing test system is not flexible enough in adjusting and expanding test scenarios, and it is difficult to adapt to changing test requirements.

[0009] It can be seen that the above existing V2X test methods still have inconveniences and defects in use, and are in urgent need of further improvement. How to create a new V2X test method has become a goal that the industry urgently needs to improve. Summary of the invention

[0010] In view of this, an embodiment of the present disclosure provides a V2X testing method based on a cloud platform, which at least partially solves the problems existing in the prior art.

[0011] In a first aspect, an embodiment of the present disclosure provides a V2X testing method based on a cloud platform, the method comprising the following steps:

[0012] Select a preset test scenario;

[0013] The cloud platform generates a configuration information package based on the key parameters of the test scenario and the node configuration rules;

[0014] The edge computing node performs initialization settings based on the configuration information package, starts data reception and transmission services, and returns configuration confirmation information to the cloud platform;

[0015] The test vehicle terminal collects vehicle data in real time through vehicle sensors and a V2X communication module according to the instructions in the configuration information package;

[0016] The edge computing node pre-processes the vehicle data and sends the pre-processed vehicle data to the cloud platform;

[0017] The cloud platform constructs a multidimensional data matrix based on the vehicle data and the test scenario data;

[0018] The multidimensional data matrix is ​​input into a pre-trained CNN-RNN hybrid network model for analysis to obtain factors affecting the communication quality of the V2X communication module.

[0019] According to a specific implementation of an embodiment of the present disclosure, the vehicle data includes: vehicle driving data, the vehicle's own position, speed, driving direction, information about surrounding vehicles, and road environment information.

[0020] According to a specific implementation of an embodiment of the present disclosure, the vehicle data includes: a driving trajectory, acceleration data and braking data of the vehicle.

[0021] According to a specific implementation of the embodiment of the present disclosure, the edge computing node pre-processes the vehicle data, including the following steps:

[0022] Performing integrity checks on the vehicle data to remove erroneous or missing data points;

[0023] Locally caching the preset timeliness data in the vehicle data;

[0024] Classification and labeling are performed according to the type and weight of the vehicle data.

[0025] According to a specific implementation of the embodiment of the present disclosure, the pre-training CNN-RNN hybrid network model includes:

[0026] Integrate the test data collected by the cloud platform according to time series and data types to build a multidimensional data matrix;

[0027] The multidimensional data matrix is ​​normalized;

[0028] Divide the normalized multidimensional data matrix into a training set, a validation set, and a test set;

[0029] Initialize the parameters of the CNN-RNN hybrid network model;

[0030] The training set data is input into the CNN-RNN hybrid network model and passed through CNN according to the forward propagation algorithm to obtain a feature vector sequence;

[0031] Inputting the feature vector sequence into the RNN for time series modeling to obtain network output;

[0032] Calculate a loss function based on the network output and the true labels of the training data;

[0033] The back propagation algorithm is used to calculate the gradient of the loss function to the network parameters, and then the optimization algorithm is used to update the network parameters according to the gradient;

[0034] After each training cycle, the performance of the CNN-RNN hybrid network model is evaluated using the validation set data. When the validation set performance no longer improves or begins to decline, the training is stopped early.

[0035] The CNN-RNN hybrid network model is evaluated based on the test set data; when the evaluation results do not meet the preset standards, the network architecture, hyperparameters, the amount of training data, data enhancement, and re-training and evaluation are adjusted until the evaluation results meet the preset standards.

[0036] According to a specific implementation of the embodiment of the present disclosure, the method further includes:

[0037] Preset solution mapping rule base based on factors affecting the communication quality of V2X communication module;

[0038] Prioritize solutions based on the impact and scope of the impact.

[0039] According to a specific implementation of the embodiment of the present disclosure, before the cloud platform constructs a multidimensional data matrix based on the vehicle data and the test scenario data, the method further includes:

[0040] Based on the preset conditions of the test scenario, the cloud platform verifies the vehicle data in real time to determine whether the vehicle data meets the preset threshold. When the preset threshold is not met, the data is marked and a data review request is sent to the edge computing node, while adjusting the data collection parameters of the vehicle terminal.

[0041] In a second aspect, an embodiment of the present disclosure provides a V2X test system based on a cloud platform, the system comprising:

[0042] A configuration information module is configured to select a preset test scenario;

[0043] And, the cloud platform generates a configuration information package based on the key parameters and node configuration rules of the test scenario;

[0044] An initialization module is configured to initialize the edge computing node based on the configuration information package, start data reception and transmission services, and return configuration confirmation information to the cloud platform;

[0045] A data acquisition module is configured for the test vehicle terminal to collect vehicle data in real time through vehicle sensors and a V2X communication module according to instructions in the configuration information package;

[0046] A preprocessing module, configured to preprocess the vehicle data at the edge computing node and send the preprocessed vehicle data to the cloud platform;

[0047] An evaluation module, configured for the cloud platform to construct a multi-dimensional data matrix based on the vehicle data and the test scenario data;

[0048] And, the multidimensional data matrix is ​​input into a pre-trained CNN-RNN hybrid network model for analysis to obtain factors affecting the communication quality of the V2X communication module.

[0049] In a third aspect, an embodiment of the present disclosure further provides an electronic device, the electronic device comprising:

[0050] at least one processor; and,

[0051] a memory communicatively connected to the at least one processor; wherein,

[0052] The memory stores instructions that can be executed by the at least one processor. When the instructions are executed by the at least one processor, the at least one processor implements the cloud platform-based V2X testing method as described in any one of the first aspect or any one of the implementations of the first aspect.

[0053] In a fourth aspect, an embodiment of the present disclosure further provides a computer program product, comprising a computer program stored on a non-transitory computer-readable storage medium, wherein the computer program comprises program instructions, and when the program instructions are executed by a computer, the computer executes the cloud platform-based V2X testing method in the aforementioned first aspect or any implementation of the first aspect.

[0054] The cloud-based V2X test method in the disclosed embodiment inputs actual test data into a trained deep learning model, and the model analyzes the data based on the learned characteristic patterns and rules. When the model output indicates a potential problem, such as the output failure probability exceeds the set threshold or the predicted communication indicator is abnormal, further in-depth analysis of the hidden layer state and neuron activation inside the model is performed, and the key data factors and characteristic patterns that cause the problem are located in combination with data features and network structure knowledge. Then, based on the pre-defined problem-solution mapping rule base, specific improvement suggestions for the problem are generated, including adjusting relevant equipment parameters, optimizing algorithm strategies, changing data transmission methods, etc., and the suggestions are prioritized according to the severity of the problem and the scope of impact, so that R&D personnel can optimize and improve the V2X system in a targeted manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The above is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0056] Figure 1 A schematic diagram of a V2X testing method flow based on a cloud platform provided in an embodiment of the present disclosure;

[0057] Figure 2 A flowchart of a V2X testing method based on a cloud platform provided in an embodiment of the present disclosure;

[0058] Figure 3 is a schematic diagram of a test vehicle terminal structure provided by an embodiment of the present disclosure;

[0059] Figure 4 It is a schematic diagram of a V2X test process based on a cloud platform provided in an embodiment of the present disclosure;

[0060] Figure 5 A schematic diagram of a V2X test system structure based on a cloud platform provided in an embodiment of the present disclosure; and

[0061] Figure 6 A schematic diagram of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0062] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0063] The following describes the embodiments of the present disclosure through specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present disclosure.

[0064] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement a device and / or practice a method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0065] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0066] The present invention discloses a V2X test system and method based on a cloud platform, aiming to provide an efficient and flexible test platform for evaluating and optimizing V2X communication performance. The system realizes rapid deployment and flexible adjustment of multiple test scenarios through the collaborative work of cloud servers, edge computing nodes, test vehicle terminals and virtual simulation platforms.

[0067] Solve the problem of limited test scenarios: Through the virtual simulation platform, various test scenarios can be easily created and adjusted to simulate different traffic environments and communication conditions, thereby improving the comprehensiveness and accuracy of the test.

[0068] Improve test efficiency: Use cloud computing and edge computing technologies to achieve efficient data processing and transmission, shorten test cycles, and reduce test costs. The system supports automated test processes, from task allocation to data collection, analysis, and report generation, with full automation to reduce manual intervention.

[0069] Improve test accuracy: The test vehicle terminal not only collects V2X communication data, but also records vehicle status information to provide comprehensive data support. Using machine learning and data mining technology, the data is deeply analyzed to identify communication bottlenecks and abnormal situations, providing accurate performance evaluation and problem diagnosis.

[0070] Support for diverse test scenarios: The system can simulate a variety of scenarios such as urban traffic management, autonomous driving, rural area testing and special environment testing to ensure the diversity and practicality of the test.

[0071] Provide intuitive visualization tools: Through the graphical user interface, users can view the test progress and results in real time, generate detailed test reports including charts, statistics and improvement suggestions, and help users quickly understand the test results and develop optimization plans.

[0072] Improve system flexibility and scalability: Support multiple users to operate online simultaneously, and each user can independently create and manage their own test scenarios. The system architecture is flexible and can easily add new test vehicles, edge computing nodes, and virtual simulation scenarios to support large-scale testing.

[0073] System architecture: The system consists of cloud servers, edge computing nodes, test vehicle terminals, and virtual simulation platforms. The cloud servers are responsible for task scheduling, data processing, and analysis; edge computing nodes provide low-latency data transmission services; test vehicle terminals are used to perform actual V2X communication tasks; and the virtual simulation platform can simulate different traffic environments and communication conditions.

[0074] Test process: After the user selects or customizes the test scenario through the cloud interface, the system automatically sends the corresponding configuration information to each node. During the test, the edge computing node collects data from the test vehicle in real time and uploads it to the cloud, which is analyzed and processed by the cloud server. At the same time, the virtual simulation platform runs synchronously to provide users with intuitive visual feedback.

[0075] Data analysis and optimization: The cloud server uses advanced algorithms to conduct in-depth learning on the collected data, identify potential problems, and make suggestions for improvement. In addition, the system also supports comparative analysis of historical data to help R&D personnel continuously optimize the performance of the V2X system.

[0076] Figure 1 A schematic diagram of a cloud-based V2X testing method process provided in an embodiment of the present disclosure.

[0077] Figure 2 For Figure 1 Corresponding flow chart of V2X test method based on cloud platform.

[0078] like Figure 1As shown, in step S110, a preset test scene is selected.

[0079] More specifically, users log in to the cloud server interface, where a variety of preset test scenario templates are presented, covering different traffic flows, road layouts, weather conditions, and V2X communication interference factors. Users can directly select one of the templates based on specific test needs, or use the custom tools provided by the interface to flexibly adjust the parameters of each scenario to create a unique test scenario. For example, users can accurately set the vehicle density, speed distribution, signal obstruction, etc. of a specific road section.

[0080] More specifically, the process proceeds to step S120.

[0081] In step S120, the cloud platform generates a configuration information package based on the key parameters of the test scenario and the node configuration rules.

[0082] More specifically, when the user completes the selection or customization of the test scenario, click the "Start Test" button. After receiving the instruction, the cloud server (or cloud platform) first parses the selected or customized test scenario information, extracts key parameters, and generates a detailed configuration information package for the edge computing node and the test vehicle terminal according to the node configuration rules pre-set by the system. The configuration information package not only contains basic communication protocol settings, data collection frequency and other general information, but also customizes data screening conditions, transmission priority, etc. for different test scenarios.

[0083] Next, go to step S130.

[0084] In step S130, the edge computing node performs initialization settings based on the configuration information package, starts data reception and transmission services, and returns configuration confirmation information to the cloud platform.

[0085] More specifically, the cloud server sends the generated configuration information package to the edge computing node and the test vehicle terminal (such as Figure 3 After receiving the configuration information, the edge computing node immediately performs initialization settings, starts data reception and transmission services, and returns configuration confirmation information to the cloud server.

[0086] After receiving the configuration information, the test vehicle terminal also performs corresponding initialization operations, including activating the on-board V2X communication module, calibrating sensors, etc., and then feeds back a ready signal to the cloud server.

[0087] Next, go to step S140.

[0088] In step S140, the test vehicle terminal collects vehicle data in real time through vehicle-mounted sensors and a V2X communication module according to the instructions in the configuration information package.

[0089] In the embodiment of the present invention, the vehicle data includes: vehicle driving data, the vehicle's own position, speed, driving direction, information about surrounding vehicles and road environment information.

[0090] In the embodiment of the present invention, the driving data of the vehicle includes: the driving trajectory, acceleration data and braking data of the vehicle.

[0091] More specifically, during the actual driving process, the test vehicle terminal collects various data in real time through the on-board sensors and V2X communication module according to the instructions in the configuration information, such as the vehicle's own position, speed, driving direction, information about surrounding vehicles (obtained through V2X communication), road environment information (such as traffic sign recognition results), etc. These data are uploaded in real time through the low-latency connection established with the edge computing node in accordance with the preset data format and transmission protocol.

[0092] Next, go to step S150.

[0093] In step S150, the edge computing node preprocesses the vehicle data and sends the preprocessed vehicle data to the cloud platform.

[0094] In an embodiment of the present invention, the edge computing node preprocesses the vehicle data, including the following steps: performing an integrity check on the vehicle data to remove erroneous or missing data points; locally caching preset timeliness data in the vehicle data; and classifying and marking the vehicle data according to its type and weight.

[0095] More specifically, after receiving the data uploaded by the test vehicle terminal, the edge computing node does not simply forward it to the cloud server, but first performs preliminary data preprocessing. The data is checked for integrity to remove data points that are obviously erroneous or missing; some data with high timeliness requirements are cached locally to ensure data continuity when the network fluctuates; at the same time, the data is classified and labeled according to its type and importance to provide auxiliary information for subsequent cloud analysis. After preprocessing, the edge computing node uploads the data to the cloud server according to the optimized transmission strategy. Unlike the prior art, this preprocessing operation of the edge computing node greatly reduces the burden on the cloud server and improves the data processing efficiency and reliability of the entire system.

[0096] Next, go to step S160.

[0097] In step S160, the cloud platform constructs a multi-dimensional data matrix based on the vehicle data and the test scenario data.

[0098] In an embodiment of the present invention, the method also includes: the cloud platform verifies the vehicle data in real time based on the preset conditions of the test scenario to determine whether the vehicle data meets the preset threshold; wherein, when the preset threshold is not met, the data is marked, and a data review request is sent to the edge computing node, and the data collection parameters of the vehicle terminal are adjusted at the same time.

[0099] More specifically, after receiving the data uploaded by the edge computing node, the cloud server first classifies the data according to the source of the data (i.e., the corresponding test vehicle terminal identification) and the test scenario information, and stores the data in a dedicated test data storage area.

[0100] Then, the multi-threaded data processing mechanism is started, and the corresponding processing modules are called for different types of data. For vehicle position and speed data, the trajectory analysis algorithm is used for processing to calculate the vehicle's driving trajectory, acceleration, sudden braking and other events; for V2X communication data, signal strength analysis, communication delay measurement, data packet loss rate statistics and other operations are performed. During the data processing process, the cloud server will also verify the data in real time based on the preset conditions of the test scenario to determine whether the data meets the expected test requirements. If data anomalies are found, such as the speed data of a vehicle exceeds the unreasonable range of the road speed limit in the scenario, the cloud server will immediately mark the data and send a data review request to the edge computing node, while adjusting the data collection parameters of the vehicle terminal to ensure the accuracy of the data.

[0101] Finally, different types of data (such as vehicle motion data, V2X communication data, road environment data, etc.) are integrated according to time series and spatial relationships to construct a multidimensional data matrix as the input of the algorithm model. For example, for vehicle motion data, the vehicle position, speed, acceleration and other information at consecutive time points are combined into a data sequence, and similar data sequences of surrounding vehicles and corresponding road environment information are associated to form a complete input data unit.

[0102] Next, go to step S170.

[0103] At step S170, the multidimensional data matrix is ​​input into a pre-trained CNN-RNN hybrid network model for analysis to obtain factors affecting the communication quality of the V2X communication module.

[0104] In an embodiment of the present invention, the pre-training CNN-RNN hybrid network model includes: integrating the test data collected by the cloud platform according to the time series and data type to construct a multidimensional data matrix; normalizing the multidimensional data matrix; dividing the normalized multidimensional data matrix into a training set, a verification set and a test set; initializing the parameters of the CNN-RNN hybrid network model; inputting the training set data into the CNN-RNN hybrid network model, passing through CNN according to the forward propagation algorithm, to obtain a feature vector sequence; inputting the feature vector sequence into RNN for time series modeling to obtain network output; and The loss function is calculated based on the real labels of the training data; the back-propagation algorithm is used to calculate the gradient of the loss function to the network parameters, and then the optimization algorithm is used to update the network parameters according to the gradient; after each training cycle, the performance of the CNN-RNN hybrid network model is evaluated using the validation set data; when the validation set performance no longer improves or begins to decline, the training is stopped in advance; the CNN-RNN hybrid network model is evaluated based on the test set data; when the evaluation result does not meet the preset standard, the network architecture, hyperparameters, increase the amount of training data, data enhancement, and re-training and evaluation are performed until the evaluation result meets the preset standard.

[0105] More specifically, after completing the collection and preliminary processing of test data, the cloud server enters the deep data analysis phase, using advanced algorithm models based on deep learning, such as the architecture combining convolutional neural network (CNN) and recurrent neural network (RNN), to learn and analyze the large amount of stored test data.

[0106] During the training process, the algorithm model automatically extracts characteristic patterns and potential laws in the data by learning a large number of data units. Unlike existing technologies, this deep learning algorithm can discover complex nonlinear relationships between data, rather than being limited to simple linear correlation analysis. For example, it can identify the subtle correlation between a vehicle's driving behavior and the quality of V2X communication in a specific traffic scenario, as well as the impact of different road environment factors on this correlation. Through multiple iterations of training, the algorithm model continuously optimizes its own parameters, improves its ability to understand and analyze data, and can accurately identify potential problem points. For example, when the model finds that vehicles in a certain area frequently experience a sudden increase in V2X communication delays accompanied by abnormal vehicle driving trajectories, the area will be marked as a potential problem area.

[0107] The present invention adopts a hybrid architecture that combines the advantages of convolutional neural network (CNN) and recurrent neural network (RNN).

[0108] Convolutional Neural Network (CNN) part:

[0109] Input layer: Receives the integrated multidimensional data matrix, one dimension of which represents the time series, and the other dimensions correspond to vehicle motion data (such as position, speed, acceleration, etc.), V2X communication data (signal strength, delay, data packet information, etc.), and road environment data (road type, weather condition coding, etc.). For example, the continuous vehicle position information is arranged into a two-dimensional image-like data structure at a certain time interval (similar to mapping time series data to spatial dimensions) so that CNN can effectively extract local features.

[0110] Convolutional layer: Set up multiple convolutional layers and use convolution kernels of different sizes. Small convolution kernels are used to capture subtle changes and local correlations in the data. For example, a 3x3 convolution kernel can focus on the relationship between adjacent time points and similar vehicles or data dimensions; large convolution kernels are used to extract more macroscopic feature patterns. For example, a 5x5 convolution kernel can cover a slightly longer time range or a wider data range to discover the correlation between vehicle group behavior and communication conditions in a larger time scale and spatial range. Each convolution layer is followed by a batch normalization layer to accelerate network convergence and improve training stability. At the same time, ReLU is used as the activation function to introduce nonlinear factors and enhance the network's expressiveness.

[0111] Pooling layer: Set the pooling layer after some convolutional layers, use the maximum pooling operation to reduce the data dimension and the amount of calculation while retaining the main feature information. For example, use a 2x2 pooling window to downsample the data output by the convolutional layer, so that the network can focus on more abstract and representative features.

[0112] Fully connected layer: After several convolution and pooling layers, the data is flattened and connected to the fully connected layer. The function of the fully connected layer is to integrate the various local and global features extracted previously and map them to a more abstract feature space so as to effectively connect with the subsequent RNN part. The number of neurons in the fully connected layer is adjusted according to the complexity of the data and the learning ability of the network, and the Dropout technology can be introduced between the fully connected layers to prevent overfitting, randomly discard a certain proportion of neuron connections, and improve the generalization ability of the network.

[0113] Recurrent Neural Network (RNN) part:

[0114] Input layer: Receives the feature vector sequence output by the CNN fully connected layer. This sequence retains the information of the data in the time dimension, allowing RNN to model the temporal nature of the data.

[0115] Circulatory layer: Long short-term memory network (LSTM) is used as the basic unit of the circulatory layer. LSTM can effectively handle long-term dependencies in long sequence data and avoid the gradient vanishing problem in traditional RNN. In the circulatory layer, multiple LSTM units are set, and each unit calculates based on the hidden state of the previous moment and the feature vector of the current input to update the hidden state of the current moment. In this way, the network can learn the dynamic change rules of data in time series, such as the evolution trend of vehicle behavior and communication status over time. The dimension of the hidden state is set according to the characteristics of the data and the learning needs of the network, and the optimal value is generally determined through experimental adjustment.

[0116] Output layer: Connected after the loop layer, the structure of the output layer is determined according to the specific analysis task. If it is a classification task, such as determining whether there is a V2X system failure or a specific type of problem, the output layer can use a softmax activation function to output the probability distribution of each category; if it is a regression task, such as predicting the specific value of communication delay or the changing trend of vehicle driving trajectory, the output layer uses a linear activation function to directly output the predicted value.

[0117] The specific process of training a neural network is as follows:

[0118] Data preprocessing and preparation:

[0119] The various test data collected from the cloud server are integrated according to time series and data types, and a multidimensional data matrix is ​​constructed as described above. The data is normalized and the eigenvalues ​​of different data ranges are mapped to a unified interval (such as [0, 1] or [-1, 1]) to facilitate network learning and training. At the same time, the data is divided into training set, validation set and test set according to a certain ratio (such as 70% - 80%). The training set is used for network parameter learning and optimization, the validation set is used to monitor the performance of the network during training to prevent overfitting, and the test set is used to finally evaluate the generalization ability and analysis effect of the network.

[0120] Network initialization and training:

[0121] Initialize the parameters of the CNN-RNN hybrid network, including convolution kernel weights, biases, LSTM unit weight matrices and bias vectors, etc. Use the random initialization method and set the appropriate initialization range based on experience to ensure that the network has good learning ability and convergence in the early stages of training.

[0122] The training set data is input into the network, and the convolution, pooling, and full connection operations of the CNN part are performed in sequence according to the forward propagation algorithm to obtain the feature vector sequence, which is then input into the RNN part for time series modeling, and finally the output of the network is obtained. The loss function is calculated based on the output and the true label of the training data (such as known problem types, actual communication delay values, etc.). The loss function is selected in an appropriate form according to the specific task. For example, the cross entropy loss function is commonly used for classification tasks, and the mean square error loss function is commonly used for regression tasks.

[0123] The back-propagation algorithm is used to calculate the gradient of the loss function to the network parameters, and then an optimization algorithm (such as stochastic gradient descent (SGD), Adagrad, Adadelta or Adam, etc.) is used to update the network parameters according to the gradient. During the training process, set an appropriate learning rate. The learning rate can be a fixed value or a strategy that gradually decays as the training progresses to balance the learning speed and convergence accuracy of the network. At the same time, after each training cycle (epoch), the performance of the network is evaluated using the validation set data, and the loss value and accuracy (if it is a classification task) or other evaluation indicators (if it is a regression task) are recorded. If it is found that the performance of the validation set is no longer improving or begins to decline, stop training in advance to prevent overfitting.

[0124] Model evaluation and optimization:

[0125] After training is completed, use the test set data to perform a final evaluation of the network model. Calculate indicators such as loss value, accuracy (classification task) or prediction error (regression task) on the test set to measure the generalization ability and analysis effect of the network model on unknown data. If the evaluation result is not ideal, you can further adjust the network architecture, hyperparameters (such as convolution kernel size, number of layers, number of LSTM units, learning rate, etc.) or increase the amount of training data, data enhancement and other operations, and then re-train and evaluate until satisfactory model performance is obtained.

[0126] Problem Identification and Suggestion Generation:

[0127] The actual test data is input into the trained deep learning model, and the model analyzes the data based on the learned characteristic patterns and rules. When the model output indicates a potential problem, such as the output failure probability exceeds the set threshold or the predicted communication indicator is abnormal, further in-depth analysis of the hidden layer state and neuron activation inside the model is performed, and the key data factors and characteristic patterns that cause the problem are located by combining data features and network structure knowledge. Then, based on the pre-defined problem-solution mapping rule base, specific improvement suggestions for the problem are generated, including adjusting relevant equipment parameters, optimizing algorithm strategies, changing data transmission methods, etc., and the suggestions are prioritized according to the severity of the problem and the scope of impact, so that R&D personnel can optimize and improve the V2X system in a targeted manner.

[0128] In an embodiment of the present invention, the method further includes: presetting a solution mapping rule base based on factors affecting the communication quality of the V2X communication module; and prioritizing the solutions according to the degree and scope of influence of the influencing factors.

[0129] More specifically, after identifying potential problem points, the cloud server generates corresponding improvement suggestions based on the internal logic of the algorithm model and predefined optimization rules. These suggestions are not just simple problem descriptions, but also include specific solutions and optimization strategies. For example, if it is found that the V2X communication problem is caused by signal interference at a certain intersection, improvement suggestions may include adjusting the communication base station parameters near the intersection, optimizing the signal reception algorithm of the vehicle terminal, or suggesting adding signal relay equipment in the area. At the same time, the cloud server will also prioritize the improvement suggestions and determine the optimization measures that need to be implemented first based on the severity of the problem and the scope of impact.

[0130] In addition to analyzing single test data, the cloud server also supports comparative analysis of historical data. The system will regularly organize and archive different batches of test data to establish a historical data warehouse. When performing comparative analysis, the cloud server first selects relevant data subsets from the historical data warehouse based on the similarity and time sequence of the test scenarios. Then, the difference analysis algorithm is used to compare and calculate the key indicators in different data subsets (such as V2X communication success rate, vehicle driving safety indicators, etc.) to find out the changing trends and differences of the indicators. For example, by comparing the test data of the same section of road in different seasons and different time periods, it is found that in winter, due to the cold weather and snow on the road, the braking distance of the vehicle increases, resulting in the need to advance the safety warning time in V2X communication accordingly. Based on the results of this historical data comparison and analysis, R&D personnel can have a more comprehensive understanding of the performance of the V2X system under different conditions, carry out targeted system optimization and upgrades, and continuously improve the stability and adaptability of the V2X system. Compared with the existing technology of only analyzing a single test data in isolation, this can provide a more in-depth and comprehensive basis for system performance evaluation and optimization.

[0131] The V2X testing method based on the cloud platform of the present invention has the following improvements:

[0132] Support for diverse test scenarios: Through the virtual simulation platform, various test scenarios can be easily created and adjusted to simulate different traffic environments and communication conditions, thereby improving the comprehensiveness and accuracy of the test.

[0133] Improve test efficiency: Use cloud computing and edge computing technologies to achieve efficient data processing and transmission, shorten test cycles, and reduce test costs.

[0134] Enhanced data processing capabilities: The use of advanced data processing and analysis technologies, such as distributed computing and machine learning, can process large-scale test data in real time and provide accurate analysis results.

[0135] Provide intuitive visualization tools: Through the graphical user interface, users can intuitively view test results and system status, quickly identify problems and optimization solutions.

[0136] Improve flexibility: The system supports multiple users to operate online simultaneously, and can flexibly adjust test scenarios and parameters to adapt to changing test needs.

[0137] In the V2X test method based on a cloud platform proposed in the present invention, during the entire test process, the virtual simulation platform and the actual test run synchronously. The cloud server transmits some key data (such as vehicle location, traffic condition changes, etc.) after real-time processing to the virtual simulation platform. Based on these data, the virtual simulation platform performs visual display in a pre-built virtual traffic environment model to provide users with intuitive visual feedback. Users can view the test progress in the virtual simulation platform in real time through the cloud interface, including the driving dynamics of the vehicle, the simulation effect of V2X communication, etc. Compared with the prior art, this close integration of virtual simulation and actual testing enables users to have a more comprehensive understanding of the test situation, discover potential problems in a timely manner, and make some hypothetical test adjustments in a virtual environment to provide a reference for actual testing.

[0138] In the embodiment of the present invention, Figure 4 As shown, the system of the present invention may include the following structure:

[0139] 1. System Architecture

[0140] 1.1 Cloud Server

[0141] Function: Responsible for the management and control of the entire test system, including task scheduling, data storage, analysis and report generation.

[0142] Technical details:

[0143] Distributed computing: Use distributed computing technologies, such as Hadoop or Spark, to improve data processing capabilities and response speed.

[0144] Database: Use relational databases (such as MySQL) and NoSQL databases (such as MongoDB) to store different types of data.

[0145] API interface: Provides a RESTful API interface to facilitate calls from other systems and services.

[0146] Security: Use SSL / TLS protocol to encrypt data transmission to ensure data security.

[0147] 1.2 Edge Computing Nodes

[0148] Function: Deployed near the test area to reduce data transmission delay and speed up data processing.

[0149] Technical details:

[0150] Hardware configuration: high-performance processor (such as Intel Xeon), large-capacity memory (such as 64GB RAM), high-speed storage (such as NVMe SSD).

[0151] Software configuration: Run a lightweight operating system (such as Ubuntu Server) and be equipped with data processing and transmission software (such as Node.js).

[0152] Network connection: Supports wired and wireless connections (such as 5G, Wi-Fi) to ensure low latency and high bandwidth.

[0153] 1.3 Test vehicle terminal

[0154] Function: Equipment installed on the test vehicle, including V2X communication module, GPS positioning module and sensor interface, used to perform V2X communication tasks.

[0155] Technical details:

[0156] V2X communication module: supports DSRC (Dedicated Short-Range Communications) and C-V2X (Cellular V2X) standards.

[0157] GPS positioning module: High-precision GPS module, providing real-time location information.

[0158] Sensor interface: supports the access of multiple sensors (such as radar, camera, accelerometer) to collect vehicle status information.

[0159] Communication protocols: Support standard V2X communication protocols such as IEEE 802.11p and 3GPP R14.

[0160] 1.4 Virtual Simulation Platform

[0161] Function: A cloud-based virtualized environment capable of simulating various traffic scenarios and communication conditions.

[0162] Technical details:

[0163] Simulation engine: Use high-performance simulation engines (such as SUMO and CARLA) to model traffic scenarios.

[0164] Virtualization technology: Containerization technology based on Kubernetes and Docker supports simultaneous online operations by multiple users.

[0165] Interactive interface: Provides a graphical user interface (GUI) that allows users to intuitively create and edit test scenarios.

[0166] Real-time feedback: Supports real-time data streaming to provide users with instant feedback on test results.

[0167] In the embodiment of the present invention, the following method is used to Figure 4 The system shown was tested:

[0168] 2. Testing process

[0169] 2.1 Scene Creation

[0170] User Interface: Through the web interface, users can select preset test scenarios or customize new scenarios.

[0171] Scenario configuration: Users can specify test objectives (such as communication stability, latency), test parameters (such as vehicle speed, traffic density), and test conditions (such as weather conditions, road type).

[0172] Scene saving: Scenes created by users can be saved in the cloud server for easy subsequent call and modification.

[0173] 2.2 Task Allocation

[0174] Task generation: The cloud server generates a detailed test plan based on the scenario selected by the user, including test time, participating vehicles, test path, etc.

[0175] Task distribution: The test plan is distributed to each edge computing node and test vehicle terminal through the API interface.

[0176] Task confirmation: After receiving the task, each node and terminal sends a confirmation message to the cloud server to ensure that the task is successfully assigned.

[0177] 2.3 Data Collection

[0178] Mission execution: The test vehicle performs V2X communication tasks as planned, including sending and receiving V2X messages and recording vehicle status information.

[0179] Data collection: Edge computing nodes collect data from test vehicles in real time, including communication data, location information, sensor data, etc.

[0180] Data upload: The collected data is uploaded to the cloud server via wired or wireless network to ensure data integrity and real-time.

[0181] 2.4 Data Analysis

[0182] Data processing: The cloud server cleans and preprocesses the uploaded data to remove invalid and abnormal data.

[0183] Performance evaluation: Use statistical analysis methods (such as mean, standard deviation) to evaluate the stability, reliability and efficiency of V2X communications.

[0184] Problem diagnosis: Discover potential problems such as signal interference and network congestion through data mining techniques (such as cluster analysis and decision trees).

[0185] Optimization suggestions: Based on the analysis results, provide specific optimization suggestions, such as adjusting the communication frequency, optimizing the routing algorithm, etc.

[0186] 2.5 Result Feedback

[0187] Report generation: The cloud server generates detailed test reports including charts, statistics, and improvement suggestions.

[0188] Result display: The analysis results and reports are presented to users through the web interface, and can be downloaded and shared.

[0189] Real-time monitoring: Provides real-time monitoring function, users can check the test progress and results at any time.

[0190] 3. Data analysis and optimization

[0191] 3.1 Performance Evaluation

[0192] Communication stability: Evaluate the packet loss rate, number of retransmissions, and number of connection interruptions in V2X communications.

[0193] Communication reliability: Evaluate the success rate, bit error rate, and transmission error rate of V2X communications.

[0194] Communication efficiency: Evaluate latency, throughput, and bandwidth utilization of V2X communications.

[0195] 3.2 Problem Diagnosis

[0196] Signal interference: Detect signal interference sources such as other wireless devices, electromagnetic wave interference, etc. through spectrum analysis technology.

[0197] Network congestion: Detect network congestion points such as insufficient bandwidth and packet queuing through traffic analysis technology.

[0198] Hardware failure: Identify hardware failures through fault detection algorithms, such as communication module damage, sensor failure, etc.

[0199] Figure 5 The cloud platform-based V2X test system 500 provided by the present invention is shown, including a configuration information module 510, an initialization module 520, a data acquisition module 530, a pre-processing module 540 and an evaluation module 550.

[0200] The configuration information module 510 is used to select a preset test scenario; and the cloud platform generates a configuration information package based on the key parameters and node configuration rules of the test scenario;

[0201] The initialization module 520 is used for the edge computing node to perform initialization settings based on the configuration information package, start data reception and transmission services, and return configuration confirmation information to the cloud platform;

[0202] The data acquisition module 530 is used by the test vehicle terminal to collect vehicle data in real time through the vehicle-mounted sensors and the V2X communication module according to the instructions in the configuration information package;

[0203] The preprocessing module 540 is used for the edge computing node to preprocess the vehicle data and send the preprocessed vehicle data to the cloud platform;

[0204] The evaluation module 550 is used for the cloud platform to construct a multidimensional data matrix based on the vehicle data and the test scenario data; and to input the multidimensional data matrix into a pre-trained CNN-RNN hybrid network model for analysis to obtain factors affecting the communication quality of the V2X communication module.

[0205] See also Figure 6 The embodiment of the present disclosure further provides an electronic device 60, the electronic device comprising:

[0206] at least one processor; and,

[0207] a memory communicatively connected to the at least one processor; wherein,

[0208] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the cloud platform-based V2X testing method in the aforementioned method embodiment.

[0209] The embodiments of the present disclosure also provide a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the cloud platform-based V2X testing method in the aforementioned method embodiment.

[0210] The embodiments of the present disclosure also provide a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the cloud platform-based V2X testing method in the aforementioned method embodiment.

[0211] Reference below Figure 6 , which shows a schematic diagram of the structure of an electronic device 60 suitable for implementing the embodiment of the present disclosure. The electronic device in the embodiment of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0212] like Figure 6 As shown, the electronic device 60 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 60 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0213] Typically, the following devices may be connected to the I / O interface 605: input devices 606 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 608 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 609. The communication device 609 may allow the electronic device 60 to communicate with other devices wirelessly or by wire to exchange data. Although the electronic device 60 with various devices is shown in the figure, it should be understood that it is not required to implement or have all the devices shown. More or fewer devices may be implemented or have alternatively.

[0214] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 609, or installed from a storage device 608, or installed from a ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0215] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0216] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0217] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains at least two Internet Protocol addresses; sends a node evaluation request including the at least two Internet Protocol addresses to a node evaluation device, wherein the node evaluation device selects an Internet Protocol address from the at least two Internet Protocol addresses and returns it; receives the Internet Protocol address returned by the node evaluation device; wherein the obtained Internet Protocol address indicates an edge node in a content distribution network.

[0218] Alternatively, the computer-readable medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device: receives a node evaluation request including at least two Internet Protocol addresses; selects an Internet Protocol address from the at least two Internet Protocol addresses; and returns the selected Internet Protocol address; wherein the received Internet Protocol address indicates an edge node in a content distribution network.

[0219] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0220] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0221] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a unit does not limit the unit itself in some cases. For example, the first acquisition unit may also be described as a "unit for acquiring at least two Internet Protocol addresses".

[0222] It should be understood that various parts of the present disclosure may be implemented in hardware, software, firmware, or a combination thereof.

[0223] The above is only a specific implementation of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present disclosure should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A V2X testing method based on a cloud platform, characterized in that: The method comprises the following steps: Select a preset test scenario; The cloud platform generates a configuration information package based on the key parameters of the test scenario and the node configuration rules; The edge computing node performs initialization settings based on the configuration information package, starts data reception and transmission services, and returns configuration confirmation information to the cloud platform; The test vehicle terminal collects vehicle data in real time through the vehicle-mounted sensors and the V2X communication module according to the instructions in the configuration information package; The edge computing node pre-processes the vehicle data and sends the pre-processed vehicle data to the cloud platform; The cloud platform constructs a multidimensional data matrix based on the vehicle data and the test scenario data; Inputting the multidimensional data matrix into a pre-trained CNN-RNN hybrid network model for analysis to obtain factors affecting the communication quality of the V2X communication module; Pre-trained CNN-RNN hybrid network model, including: Integrate the test data collected by the cloud platform according to time series and data types to build a multidimensional data matrix; The multidimensional data matrix is ​​normalized; Divide the normalized multidimensional data matrix into a training set, a validation set, and a test set; Initialize the parameters of the CNN-RNN hybrid network model; The training set data is input into the CNN-RNN hybrid network model and passed through CNN according to the forward propagation algorithm to obtain a feature vector sequence; Inputting the feature vector sequence into the RNN for time series modeling to obtain network output; Calculate a loss function based on the network output and the true labels of the training data; The back propagation algorithm is used to calculate the gradient of the loss function to the network parameters, and then the optimization algorithm is used to update the network parameters according to the gradient; After each training cycle, the performance of the CNN-RNN hybrid network model is evaluated using the validation set data. When the validation set performance no longer improves or begins to decline, the training is stopped early. Evaluate the CNN-RNN hybrid network model based on the test set data; when the evaluation results do not meet the preset standards, adjust the network architecture, hyperparameters, increase the amount of training data, and enhance the data, and retrain and evaluate until the evaluation results meet the preset standards; The method further comprises: Preset solution mapping rule base based on factors affecting the communication quality of V2X communication module; Prioritize solutions based on the impact and scope of the impact.

2. The V2X testing method based on a cloud platform according to claim 1, characterized in that: The vehicle data includes: vehicle driving data, the vehicle's own position, speed, driving direction, information about surrounding vehicles and road environment information.

3. The cloud platform-based V2X testing method according to claim 2, characterized in that: The vehicle driving data includes: the vehicle's driving trajectory, acceleration data and braking data.

4. The cloud platform-based V2X testing method according to claim 1, characterized in that: The edge computing node preprocesses the vehicle data, comprising the following steps: Performing integrity checks on the vehicle data to remove erroneous or missing data points; Locally caching the preset timeliness data in the vehicle data; Classification and labeling are performed according to the type and weight of the vehicle data.

5. The V2X testing method based on a cloud platform according to claim 1, characterized in that: Before the cloud platform constructs a multidimensional data matrix based on the vehicle data and the test scenario data, the method further includes: Based on the preset conditions of the test scenario, the cloud platform verifies the vehicle data in real time to determine whether the vehicle data meets the preset threshold. When the preset threshold is not met, the data is marked and a data review request is sent to the edge computing node, while adjusting the data collection parameters of the vehicle terminal.

6. A V2X test system based on a cloud platform, characterized in that: The system comprises: A configuration information module is configured to select a preset test scenario; And, the cloud platform generates a configuration information package based on the key parameters and node configuration rules of the test scenario; An initialization module is configured to initialize the edge computing node based on the configuration information package, start data reception and transmission services, and return configuration confirmation information to the cloud platform; A data acquisition module is configured to test that the vehicle terminal collects vehicle data in real time through the vehicle-mounted sensors and the V2X communication module according to the instructions in the configuration information package; A preprocessing module, configured to preprocess the vehicle data at the edge computing node and send the preprocessed vehicle data to the cloud platform; An evaluation module is configured to construct a multidimensional data matrix based on the vehicle data and the test scenario data on the cloud platform; and input the multidimensional data matrix into a pre-trained CNN-RNN hybrid network model for analysis to obtain factors affecting the communication quality of the V2X communication module; Pre-trained CNN-RNN hybrid network model, including: Integrate the test data collected by the cloud platform according to time series and data types to build a multidimensional data matrix; The multidimensional data matrix is ​​normalized; Divide the normalized multidimensional data matrix into a training set, a validation set, and a test set; Initialize the parameters of the CNN-RNN hybrid network model; The training set data is input into the CNN-RNN hybrid network model and passed through CNN according to the forward propagation algorithm to obtain a feature vector sequence; Inputting the feature vector sequence into the RNN for time series modeling to obtain network output; Calculate a loss function based on the network output and the true labels of the training data; The back propagation algorithm is used to calculate the gradient of the loss function to the network parameters, and then the optimization algorithm is used to update the network parameters according to the gradient; After each training cycle, the performance of the CNN-RNN hybrid network model is evaluated using the validation set data. When the validation set performance no longer improves or begins to decline, the training is stopped early. Evaluate the CNN-RNN hybrid network model based on the test set data; when the evaluation results do not meet the preset standards, adjust the network architecture, hyperparameters, increase the amount of training data, and enhance the data, and retrain and evaluate until the evaluation results meet the preset standards; The system is also configured to: Preset solution mapping rule base based on factors affecting the communication quality of V2X communication module; Prioritize solutions based on the impact and scope of the impact.

7. An electronic device, characterized in that: The electronic device includes: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor executes the cloud platform-based V2X testing method as described in any one of claims 1 to 5.

8. A computer program product, characterized in that The computer program product includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer is caused to execute the cloud platform-based V2X testing method as described in any one of claims 1 to 5.

Images