Traffic Flow Induction Monitoring Method and System Based on Convolutional Denoising Autoencoder

Abstract

The present invention discloses a traffic flow induction monitoring method and system based on convolutional denoising autoencoder, which relates to the technical field of traffic control and includes: generating a fused spatio-temporal traffic data set; extracting in-vehicle and navigation data from the spatio-temporal traffic data set, and initially classifying the traveling vehicles into vehicles with known navigation information and vehicles with unknown navigation information; using a convolutional denoising autoencoder, a spatio-temporal convolutional neural network, and a long short-term memory network to process the spatio-temporal traffic data of the vehicles with unknown navigation information to generate a traffic flow prediction model; training and optimizing the traffic flow prediction model using a joint loss function; calculating the traffic index of the next section according to the spatio-temporal traffic data of the vehicles with known navigation information and the vehicles with unknown navigation information, and if the traffic index of the next section is greater than a preset traffic threshold, guiding the vehicle to switch to the recommended alternative path to provide the optimal driving path suggestion, so as to realize the induction and optimization of the traffic flow.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic control, and specifically to a traffic flow induction monitoring method and system based on convolutional denoising autoencoder. Background Art

[0002] With the acceleration of the urbanization process and the continuous growth of traffic demand, the problem of traffic congestion has become increasingly serious, which has had a serious impact on people's travel efficiency and urban operation efficiency. In order to effectively alleviate traffic congestion and improve road traffic capacity, the traffic flow induction system has emerged. Existing traffic flow induction systems usually adopt traditional data processing and prediction methods, such as time series analysis, machine learning algorithms, etc. However, when dealing with large-scale and high-dimensional traffic data, these methods often have problems such as low computational efficiency and insufficient prediction accuracy. At the same time, it is difficult for these methods to effectively capture the spatio-temporal characteristics of traffic data, that is, the mutual correlation and dependence relationships of traffic data in space and time.

[0003] In recent years, the development of deep learning technology has provided new solutions for traffic flow induction systems. In particular, deep learning models such as convolutional neural network (CNN), recurrent neural network (RNN) and its variants (such as long short-term memory network LSTM) have achieved remarkable results in the fields of image recognition, speech recognition, natural language processing, etc. These models have powerful feature extraction and pattern recognition capabilities, can process large-scale and high-dimensional data, and capture complex patterns and correlation relationships in the data.

[0004] However, there are still some deficiencies in existing deep learning-based traffic flow induction systems. For example, in the data preprocessing stage, existing systems often adopt simple filtering and normalization methods, which are difficult to effectively remove noise and outliers in the data. In the prediction stage, existing systems usually only consider the traffic data of all vehicles, while ignoring the mutual influence between vehicles with known navigation information and vehicles with unknown navigation information, resulting in inaccurate and incomplete prediction results. Summary of the Invention

[0005] (I) Technical Problems to be Solved

[0006] Aiming at the deficiencies of the prior art, the present invention provides a traffic flow induction monitoring method and system based on convolutional denoising autoencoders. By collecting and fusing multi-source heterogeneous traffic data, the convolutional denoising autoencoder is used to denoise the spatio-temporal traffic data of vehicles with unknown navigation information, and then the spatio-temporal convolutional neural network and the long short-term memory network are combined for feature extraction and sequence modeling, so as to obtain a high-precision traffic flow prediction model. On this basis, by introducing a joint loss function and comprehensively considering the losses of vehicles with known navigation information and vehicles with unknown navigation information, the model is trained and optimized to improve the prediction performance. Finally, according to the prediction results and the traffic capacity of the road section, the optimal driving route suggestion is provided for the driver to realize the induction and optimization of the traffic flow.

[0007] (II)Technical solutions

[0008] To achieve the above objectives, the present invention is realized through the following technical solutions: A traffic flow induction monitoring method based on convolutional denoising autoencoders, comprising the following steps:

[0009] Collect video data, radar data, in-vehicle data, and navigation data of driving vehicles, match them according to timestamps and geographical locations, and generate a fused spatio-temporal traffic data set;

[0010] Extract in-vehicle and navigation data from the spatio-temporal traffic data set, initially divide the driving vehicles into vehicles with known navigation information and vehicles with unknown navigation information. For vehicles with known navigation information, judge whether their current positions deviate from the navigation path. If they deviate from the navigation path, reclassify them as vehicles with unknown navigation information, otherwise do not process;

[0011] Screen out the spatio-temporal traffic data of vehicles with unknown navigation information from the spatio-temporal traffic data set, and use the convolutional denoising autoencoder, spatio-temporal convolutional neural network, and long short-term memory network to process the spatio-temporal traffic data of vehicles with unknown navigation information to generate a traffic flow prediction model;

[0012] Extract the spatio-temporal traffic data of vehicles with known navigation information from the spatio-temporal traffic data set, generate a joint loss function through the loss functions of vehicles with known navigation information and vehicles with unknown navigation information, and use the joint loss function and the spatio-temporal traffic data of vehicles with known navigation information to train and optimize the traffic flow prediction model;

[0013] According to the spatio-temporal traffic data of vehicles with known navigation information and vehicles with unknown navigation information, obtain the traffic flow of the next road section, combine the saturation and traffic capacity of the next road section, calculate the traffic passing index of the next road section. If the traffic passing index of the next road section is greater than the preset passing threshold, guide the vehicle to switch to the recommended alternative path.

[0014] Further, initially dividing the driving vehicles into vehicles with known navigation information and vehicles with unknown navigation information includes:

[0015] Extract the in-vehicle data and navigation data of all vehicles from the fused spatio-temporal traffic dataset. The in-vehicle data includes the unique identifier of the vehicle and real-time location information. The navigation data includes the departure location, destination, and planned navigation path information of the vehicle.

[0016] All vehicles are initially divided into vehicles with known navigation information and vehicles with unknown navigation information according to whether the vehicle is associated with valid navigation data. Among them, vehicles with known navigation information are vehicles whose navigation data contains navigation path information, and vehicles with unknown navigation information are vehicles that are not associated with navigation data or whose navigation data does not contain navigation path information.

[0017] Furthermore, determine whether its current position deviates from the navigation path, including:

[0018] Preset a path deviation threshold. If the shortest distance from the current position of the vehicle to the navigation path exceeds the path deviation threshold, the vehicle is determined to have deviated from the navigation path, and the vehicle is removed from the list of vehicles with known navigation information and reclassified as a vehicle with unknown navigation information. If the shortest distance from the current position of the vehicle to the navigation path does not exceed the path deviation threshold, the vehicle is determined not to have deviated from the navigation path.

[0019] Furthermore, generate a traffic flow prediction model, including:

[0020] Screen out the spatio-temporal traffic data of vehicles with unknown navigation information from the spatio-temporal traffic dataset, including the video data, radar data, and in-vehicle data of the vehicle, and use a convolutional denoising autoencoder to denoise the spatio-temporal traffic data of vehicles with unknown navigation information. Use the denoised spatio-temporal traffic data of vehicles with unknown navigation information as input, and use a spatio-temporal convolutional neural network to extract high-order spatio-temporal features.

[0021] Connect a long short-term memory network to perform sequence modeling on the extracted high-order spatio-temporal features. Perform a concatenation operation on the output features of the spatio-temporal convolutional neural network and the long short-term memory network to fuse the feature information from different network layers, and send the fused features to a fully connected layer for prediction to obtain a preliminary traffic flow prediction model.

[0022] Furthermore, use a joint loss function to train and optimize the traffic flow prediction model, including:

[0023] Calculate the proportion of vehicles with known navigation information among all driving vehicles: p = the number of vehicles with known navigation information / the number of all driving vehicles. Set a loss function for vehicles with unknown navigation information and vehicles with known navigation information respectively, and assign a weight coefficient to the loss function of vehicles with known navigation information: , where ω represents the weight coefficient and p represents the proportion of vehicles with known navigation information among all moving vehicles.

[0024] Furthermore, combining the loss functions of vehicles with unknown navigation information and vehicles with known navigation information, as well as the weight coefficient, a joint loss function is generated: , where L_co represents the joint loss function, L_un represents the loss function of vehicles with unknown navigation information, and L_kn represents the loss function of vehicles with known navigation information;

[0025] The traffic flow prediction model is trained using the joint loss function and the spatio-temporal traffic data of vehicles with known navigation information. The parameters of the traffic flow prediction model are updated through the backpropagation algorithm until the preset number of training rounds is reached or the loss function converges.

[0026] Furthermore, obtaining the traffic flow of the next section includes:

[0027] When a vehicle starts its in-vehicle navigation or navigation software, the traffic flow of all vehicles with known navigation information when the current vehicle proceeds to the next section is counted through the spatio-temporal traffic data of all vehicles with known navigation information. The spatio-temporal traffic data of vehicles with unknown navigation information is used as input, and the traffic flow prediction model is used to predict the traffic flow of vehicles with unknown navigation information in the next section.

[0028] Furthermore, calculating the traffic index of the next section includes:

[0029] Based on the traffic flow of vehicles with known navigation information and vehicles with unknown navigation information, combined with the saturation and traffic capacity of the next section, the traffic index of the next section is calculated. The calculation formula is as follows:

[0030] ;

[0031] where TI represents the traffic index, Tv_u represents the traffic flow of vehicles with known navigation information, Tv_k represents the traffic flow of vehicles with unknown navigation information, Se represents the saturation, Tc represents the traffic capacity, and Ct represents the congestion critical value.

[0032] Furthermore, when the traffic index of the next section is less than or equal to the preset traffic threshold, no processing is performed. When the traffic index of the next section is greater than the preset traffic threshold, alternative sections are searched, and the traffic indices of all alternative sections are calculated; if the traffic indices of all alternative sections are greater than the preset traffic threshold, the section with the lowest traffic index is selected as the recommended section, and a navigation instruction is sent to the vehicle to guide the vehicle to switch to the recommended alternative path.

[0033] A traffic flow induction monitoring system based on convolutional denoising autoencoder includes:

[0034] A data acquisition module collects video data, radar data, vehicle-mounted data, and navigation data of a moving vehicle, matches them according to timestamps and geographical locations, and generates a fused spatio-temporal traffic dataset;

[0035] A vehicle classification module extracts vehicle-mounted and navigation data from the spatio-temporal traffic dataset, initially classifies moving vehicles into vehicles with known navigation information and vehicles with unknown navigation information. For vehicles with known navigation information, it determines whether their current position deviates from the navigation path. If it deviates from the navigation path, it re-classifies them as vehicles with unknown navigation information; otherwise, no processing is performed.

[0036] A model construction module filters out the spatio-temporal traffic data of vehicles with unknown navigation information from the spatio-temporal traffic dataset, and uses a convolutional denoising autoencoder, a spatio-temporal convolutional neural network, and a long short-term memory network to process the spatio-temporal traffic data of vehicles with unknown navigation information to generate a traffic flow prediction model;

[0037] A model training and optimization module extracts the spatio-temporal traffic data of vehicles with known navigation information from the spatio-temporal traffic dataset, generates a joint loss function through the loss functions of vehicles with known navigation information and vehicles with unknown navigation information, and uses the joint loss function and the spatio-temporal traffic data of vehicles with known navigation information to train and optimize the traffic flow prediction model;

[0038] A traffic flow guidance decision module obtains the traffic flow of the next road segment based on the spatio-temporal traffic data of vehicles with known navigation information and vehicles with unknown navigation information, combines the saturation and traffic capacity of the next road segment, calculates the traffic index of the next road segment. If the traffic index of the next road segment is greater than a preset traffic threshold, it guides the vehicle to switch to the recommended alternative path.

[0039] (III) Beneficial effects

[0040] The present invention provides a traffic flow guidance and monitoring method and system based on convolutional denoising autoencoding, having the following beneficial effects:

[0041] (1) By extracting and classifying vehicle information from the fused spatio-temporal traffic dataset to distinguish vehicles with known navigation information and vehicles with unknown navigation information, it is possible to more accurately grasp the driving status and path planning of vehicles. For vehicles with known navigation information, through path matching and deviation judgment, it is possible to timely discover and handle the situation where the vehicle deviates from the navigation path, thereby ensuring the accuracy and integrity of traffic data.

[0042] (2) By using CDAE to denoise the spatio-temporal traffic data of vehicles with unknown navigation information, the noise and outliers in the data can be effectively removed, and the data becomes more accurate and stable. The influence of vehicles with known navigation information is excluded, and a large amount of computational effort is reduced. The output features of STCN and LSTM are concatenated to fuse the feature information from different network layers, which helps the model to comprehensively utilize multiple information sources for prediction.

[0043] (3) By introducing the spatio-temporal traffic data of vehicles with known navigation information and jointly training it with the data of vehicles with unknown navigation information, the model can learn more diverse traffic flow characteristics, thereby enhancing its generalization ability in new scenarios or unknown conditions. Since the data of two types of vehicles are combined and reasonable weight coefficients are assigned according to their proportions among all moving vehicles, the model can consider various factors more comprehensively during prediction, thus improving the prediction accuracy.

[0044] (4) By real-time monitoring and predicting the traffic flow of each road section and combining the navigation information of vehicles, real-time traffic guidance can be provided during the vehicle's driving process to help the vehicle select the optimal driving route, avoid congestion, and improve travel efficiency. By using the traffic flow prediction model and the calculation of the traffic index, the traffic conditions of each road section can be intelligently judged, and based on the prediction results and real-time traffic data, intelligent navigation suggestions can be provided for the vehicle to reduce the time and cost losses caused by traffic congestion. Description of the Drawings

[0045] Figure 1 It is a schematic flowchart of the traffic flow induction monitoring method based on convolutional denoising autoencoder of the present invention;

[0046] Figure 2 It is a schematic diagram of the steps of the traffic flow induction monitoring method based on convolutional denoising autoencoder of the present invention;

[0047] Figure 3 It is a schematic diagram of the structure of the traffic flow induction monitoring system based on convolutional denoising autoencoder of the present invention. Detailed Embodiments

[0048] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0049] Please refer to Figure 1 - Figure 2 , the present invention provides a traffic flow induction monitoring method based on convolutional denoising autoencoder, including the following steps:

[0050] Step 1: Collect video data, radar data, in-vehicle data, and navigation data of moving vehicles, match them according to timestamps and geographical locations, and generate a fused spatio-temporal traffic dataset;

[0051] The first step includes the following:

[0052] Step 101: Deploy high-definition video cameras and microwave radar sensors at key traffic locations (such as intersections, highway entrances, urban arterial roads, etc.) to collect video data and radar data, use intelligent in-vehicle terminals to collect in-vehicle data, and use navigation software to collect navigation data;

[0053] Among them, the camera is used to capture vehicle images, identify vehicle models, colors, and license plate numbers. Considering the all-weather working requirements, the camera is equipped with an infrared night vision function. The microwave radar sensor is used to measure dynamic traffic data such as vehicle speed, traffic flow, and vehicle spacing. The radar uses frequency modulation continuous wave technology to improve detection accuracy. The sampling frequency of the sensor can reach the second level to meet the real-time data collection requirements. The intelligent in-vehicle terminal integrates a GPS positioning and wireless communication module to report the GPS position, driving speed, etc. of the vehicle in real time. The navigation software is used to collect road condition information (such as congestion, road construction, etc.) and the departure and destination locations, etc.;

[0054] Step 102: Preprocess the collected video data, radar data, in-vehicle data, and navigation data, including: using median filtering and weighted moving average methods to remove outliers and random noise, and using the Z-score normalization method to unify the data of different sensors to the same scale;

[0055] Step 103: Use the 5G network to transmit the preprocessed data from each collection point to the data center, match the camera data, radar data, in-vehicle data, and navigation data according to timestamps and geographical locations, and generate a fused spatio-temporal traffic dataset;

[0056] When in use, combine the content of steps 101 to 103:

[0057] By integrating video data, radar data, in-vehicle data, and navigation data, comprehensive information about traffic conditions can be obtained, providing a rich data basis for traffic flow induction monitoring. The fusion of multiple data sources can complement each other's deficiencies and reduce the errors and uncertainties that may be brought by a single data source.

[0058] Step 2: Extract in-vehicle and navigation data from the spatio-temporal traffic dataset, initially divide the moving vehicles into vehicles with known navigation information and vehicles with unknown navigation information. For vehicles with known navigation information, judge whether their current positions deviate from the navigation path. If they deviate from the navigation path, reclassify them as vehicles with unknown navigation information, otherwise do not process them;

[0059] Step 2 includes the following contents:

[0060] Step 201: Extract the in-vehicle data and navigation data of all vehicles from the fused spatio-temporal traffic dataset. The in-vehicle data includes the unique identifier of the vehicle (such as license plate number or device ID), and the real-time location information reported through the intelligent in-vehicle terminal (such as GPS coordinates). The navigation data includes the departure place, destination, and planned navigation path information of the vehicle;

[0061] Step 202: Initially divide all vehicles into vehicles with known navigation information and vehicles with unknown navigation information according to whether the vehicle is associated with valid navigation data. Among them, vehicles with known navigation information are vehicles whose navigation data contains navigation path information, and vehicles with unknown navigation information are vehicles that are not associated with navigation data or whose navigation data does not contain navigation path information;

[0062] Step 203: For vehicles with known navigation information, perform path matching through a path matching algorithm based on their reported real-time location information (GPS coordinates) and navigation path information. The path matching algorithm includes Euclidean distance, Manhattan distance based on geographical coordinates, or other matching algorithms;

[0063] Step 204: Preset a path deviation threshold to determine whether its current position deviates from the navigation path. For each vehicle with known navigation information, if the shortest distance from the vehicle's current position to the navigation path exceeds the path deviation threshold, the vehicle is determined to have deviated from the navigation path, removed from the list of vehicles with known navigation information, and reclassified as a vehicle with unknown navigation information; if the shortest distance from the vehicle's current position to the navigation path does not exceed the path deviation threshold, no processing is performed;

[0064] It should be noted that the setting of the path deviation threshold can be determined by analyzing historical traffic data to determine the average deviation and maximum deviation between the vehicle and the navigation path during normal driving, so as to set a reasonable threshold range. The path deviation threshold should accurately reflect whether the vehicle has truly deviated from the navigation path;

[0065] When in use, combine the contents of Step 201 to Step 204:

[0066] By extracting and classifying vehicle information from the fused spatio-temporal traffic dataset, distinguishing vehicles with known navigation information and vehicles with unknown navigation information, the driving state and path planning of vehicles can be more accurately grasped. For vehicles with known navigation information, through path matching and deviation judgment, the situation where the vehicle deviates from the navigation path can be discovered and processed in a timely manner, thereby ensuring the accuracy and integrity of traffic data.

[0067] Step 3: Screen out the spatio-temporal traffic data of vehicles with unknown navigation information from the spatio-temporal traffic dataset, and use a convolutional denoising autoencoder, a spatio-temporal convolutional neural network, and a long short-term memory network to process the spatio-temporal traffic data of vehicles with unknown navigation information to generate a traffic flow prediction model;

[0068] The third step includes the following contents:

[0069] Step 301: Screen out the spatio-temporal traffic data of vehicles with unknown navigation information from the spatio-temporal traffic dataset, including the video data, radar data, and in-vehicle data of the vehicles, and use a convolutional denoising autoencoder (CDAE) to denoise the spatio-temporal traffic data of vehicles with unknown navigation information;

[0070] Among them, constructing a CDAE model includes a convolutional encoder and a convolutional decoder. The encoder maps the input data to a low-dimensional latent space, and the decoder then restores the latent representation to clean output data. Random noise (such as Gaussian noise, salt noise, etc.) is added to the input data, and the CDAE network is trained to minimize the reconstruction error between the noisy data and the original data. After the training is completed, the final CDAE model is obtained;

[0071] Step 302: Use the spatio-temporal traffic data of vehicles with unknown navigation information after denoising as input, and use a spatio-temporal convolutional neural network (Spatial-Temporal Convolutional Network, STCN) to extract high-order spatio-temporal features;

[0072] STCN uses a graph convolutional network in the spatial dimension to capture the road network topology, and uses causal convolution in the time dimension to achieve one-way time information transmission to avoid using future data. The Inception module is used at the bottom layer of the network to increase the receptive field and capture more context information. Residual connections are used at the top layer to accelerate the model convergence speed and prevent overfitting;

[0073] Step 303: After STCN, connect a long short-term memory network (LSTM) to perform sequence modeling on the extracted high-order spatio-temporal features;

[0074] It should be noted that LSTM can adaptively select the information to be remembered and forgotten through a gating mechanism and can capture long-term dependencies. On the basis of LSTM, an attention mechanism can also be introduced to dynamically adjust the weights of different time steps and highlight the influence of critical moments;

[0075] Step 304: Concatenate the output features of STCN and LSTM to fuse the feature information from different network layers, and send the fused features into a fully connected layer for prediction to obtain a preliminary traffic flow prediction model, which predicts the traffic flow of each road section within a future time period (such as within 15 - 60 minutes), including traffic volume, average speed, and congestion index;

[0076] It should be noted that the fully connected layer can adopt a recursive prediction structure, that is, first predict the traffic conditions (such as traffic volume, average speed, congestion index, etc.) at the first future time step, and then use it as input to predict the traffic conditions at the next time step, and so on, until predicting the traffic conditions within the next 15 - 60 minutes;

[0077] When in use, combine the content of Steps 301 to 304:

[0078] By using CDAE to denoise the spatio - temporal traffic data of vehicles with unknown navigation information, the noise and outliers in the data can be effectively removed, and it is more accurate and stable. The influence of vehicles with known navigation information is excluded, reducing a large amount of computational complexity. Concatenating the output features of STCN and LSTM to fuse the feature information from different network layers helps the model comprehensively utilize multiple information sources for prediction. When predicting through the fully connected layer, key indicators such as traffic volume, average speed, and congestion index of each road section within the future time period can be output, providing strong decision - making support for traffic management and guidance.

[0079] Step Four: Extract the spatio - temporal traffic data of vehicles with known navigation information from the spatio - temporal traffic dataset, generate a joint loss function through the loss functions of vehicles with known navigation information and vehicles with unknown navigation information, and use the joint loss function and the spatio - temporal traffic data of vehicles with known navigation information to train and optimize the traffic flow prediction model;

[0080] The above Step Four includes the following content:

[0081] Step 401: Extract the spatio - temporal traffic data of vehicles with known navigation information from the spatio - temporal traffic dataset, including video data, radar data, and in - vehicle data of the vehicles, and divide the spatio - temporal traffic data of vehicles with known navigation information into a training set, a validation set, and a test set to ensure the randomness and representativeness of data division;

[0082] Step 402: Use the traffic flow prediction model (based on the data of vehicles with unknown navigation information) trained in Step Three to predict the test set of vehicles with known navigation information, and evaluate the performance indicators of the traffic flow prediction model according to the prediction results, including accuracy, recall rate, F1 - score, and mean squared error MSE, etc.;

[0083] Step 403: Calculate the proportion of vehicles with known navigation information among all running vehicles: p = number of vehicles with known navigation information / number of all running vehicles. Set a loss function (such as mean squared error MSE, cross-entropy loss, etc.) for both vehicles with unknown navigation information and vehicles with known navigation information, and assign a weight coefficient to the loss function of vehicles with known navigation information: , where ω represents the weight coefficient, and p represents the proportion of vehicles with known navigation information among all running vehicles;

[0084] Step 404: Combine the loss functions of vehicles with unknown navigation information and vehicles with known navigation information, as well as the weight coefficient, to generate a joint loss function: , where L_co represents the joint loss function, L_un represents the loss function of vehicles with unknown navigation information, and L_kn represents the loss function of vehicles with known navigation information;

[0085] Step 405: Use the joint loss function to train the traffic flow prediction model, update the parameters of the traffic flow prediction model through the backpropagation algorithm until the preset number of training rounds is reached or the loss function converges. Evaluate the performance of the optimized model on the validation set, and adjust the model parameters (such as learning rate, batch size, number of network layers, etc.) according to the evaluation results to obtain better model performance;

[0086] When in use, combine the content of Steps 401 to 405:

[0087] By introducing the spatio-temporal traffic data of vehicles with known navigation information and combining it with the data of vehicles with unknown navigation information for joint training, the model can learn more diverse traffic flow characteristics, thereby enhancing its generalization ability in new scenarios or unknown conditions. Since the data of two types of vehicles are combined and a reasonable weight coefficient is assigned according to their proportion among all running vehicles, this enables the model to consider various factors more comprehensively during prediction, thus improving the prediction accuracy.

[0088] Step Five: Obtain the traffic flow of the next road segment based on the spatio-temporal traffic data of vehicles with known navigation information and vehicles with unknown navigation information. Combine the saturation and traffic capacity of the next road segment to calculate the traffic index of the next road segment. If the traffic index of the next road segment is greater than the preset traffic threshold, guide the vehicle to switch to the recommended alternative route.

[0089] The above Step Five includes the following content:

[0090] Step 501: when the vehicle starts the in-vehicle navigation or navigation software, the spatiotemporal traffic data of all vehicles with known navigation information are used to count the traffic volume of all vehicles with known navigation information when the current vehicle is on the next road section, and the data of vehicles with unknown navigation information is used as input to predict the traffic volume of vehicles with unknown navigation information on the next road section using a traffic flow prediction model;

[0091] It should be noted that when the navigation information of the vehicle is known, the time it takes for the vehicle to arrive at the next road section can be calculated through the vehicle's current position, driving speed and the distance of each road section. Therefore, the time it takes for all vehicles with known navigation information to enter the next road section can be calculated, and the number of vehicles with known navigation information that enter the next road section at the same time as the current vehicle can be counted;

[0092] Step 502: Calculate the traffic index of the next road segment by combining the traffic volume of vehicles with known navigation information and vehicles with unknown navigation information with the saturation and traffic capacity of the next road segment. The calculation formula is as follows:

[0093] ;

[0094] Among them, TI represents the traffic index, Tv_u represents the traffic flow of vehicles with known navigation information, Tv_k represents the traffic flow of vehicles with unknown navigation information, Se represents saturation, Tc represents traffic capacity, and Ct represents the congestion threshold;

[0095] It should be noted that the traffic capacity refers to the maximum traffic volume that can pass through the next road section in a unit time, which is a fixed road parameter. The saturation refers to the maximum traffic volume that the next road section can bear under normal operation, but it usually does not exceed the traffic capacity. It reflects the degree of congestion on the road and is a value between the current traffic volume and the traffic capacity. The congestion threshold refers to the traffic volume value when the next road section begins to be obviously congested, which is an empirical value or a threshold determined based on historical data.

[0096] Step 503: pre-set a traffic threshold. When the traffic index of the next road segment is less than or equal to the preset traffic threshold, no processing is performed. When the traffic index of the next road segment is greater than the preset traffic threshold, it is considered that the road segment may be congested or have poor traffic, and alternative road segments are searched, and the traffic indexes of all the alternative road segments are calculated.

[0097] It should be noted that the setting of the traffic threshold needs to take into account the normal traffic capacity of the road. Under normal circumstances, the road should be able to accommodate a certain number of vehicles to pass smoothly. This amount is the traffic capacity of the road. The traffic threshold should be set within a certain proportion of the road capacity to ensure that the road can remain unobstructed in most cases.

[0098] Step 504: If the traffic indices of all alternative road segments are greater than a preset traffic threshold, select the road segment with the lowest traffic index as the recommended road segment, send a navigation instruction to the vehicle, and guide the vehicle to switch to the recommended alternative route;

[0099] When in use, combine the content of Steps 501 to 504:

[0100] By real-time monitoring and predicting the traffic flow of each road segment and combining the navigation information of the vehicle, it is possible to provide real-time traffic guidance during the vehicle's driving process, help the vehicle select the optimal driving route, avoid congestion, improve travel efficiency. By using the traffic flow prediction model and traffic index calculation, it is possible to intelligently judge the traffic conditions of each road segment, and based on the prediction results and real-time traffic data, provide intelligent navigation suggestions for the vehicle, reducing the time and cost losses caused by traffic congestion.

[0101] Please refer to Figure 3 , the present invention also provides a traffic flow induction monitoring system based on convolutional denoising autoencoder, including: a data acquisition module, a vehicle classification module, a model construction module, a model training and optimization module, and a traffic flow induction decision module; wherein,

[0102] The data acquisition module collects video data, radar data, in-vehicle data, and navigation data of the driving vehicles, matches them according to the time stamp and geographical location, and generates a fused spatio-temporal traffic data set;

[0103] The vehicle classification module extracts in-vehicle and navigation data from the spatio-temporal traffic data set, initially classifies the driving vehicles into vehicles with known navigation information and vehicles with unknown navigation information. For vehicles with known navigation information, it judges whether their current position deviates from the navigation path. If it deviates from the navigation path, it is re-classified as a vehicle with unknown navigation information, otherwise no processing is performed;

[0104] The model construction module screens out the spatio-temporal traffic data of vehicles with unknown navigation information from the spatio-temporal traffic data set, and uses a convolutional denoising autoencoder, a spatio-temporal convolutional neural network, and a long short-term memory network to process the spatio-temporal traffic data of vehicles with unknown navigation information to generate a traffic flow prediction model;

[0105] The model training and optimization module extracts the spatio-temporal traffic data of vehicles with known navigation information from the spatio-temporal traffic data set, generates a joint loss function through the loss functions of vehicles with known navigation information and vehicles with unknown navigation information, and uses the joint loss function and the spatio-temporal traffic data of vehicles with known navigation information to train and optimize the traffic flow prediction model;

[0106] The traffic flow induction decision-making module obtains the traffic volume of the next road section based on the spatio-temporal traffic data of vehicles with known navigation information and vehicles with unknown navigation information, combines the saturation and traffic capacity of the next road section, calculates the traffic index of the next road section. If the traffic index of the next road section is greater than the preset traffic threshold, it guides the vehicle to switch to the recommended alternative route.

[0107] In the application, several formulas involved are calculated by taking their numerical values after dimensionless processing. The formula is a formula obtained by software simulation of a large amount of collected data to approximate the real situation as closely as possible. The coefficients in the formula are set by those skilled in the art according to the actual situation.

[0108] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by a combination of electronic hardware, computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution.

[0109] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units. They can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0110] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed in this application can easily think of changes or substitutions, which should all be covered within the protection scope of this application.

Claims

1. A traffic flow induction monitoring method based on convolutional denoising autoencoder, characterized by: The following steps are involved: Collect video data, radar data, vehicle data, and navigation data of moving vehicles, match them according to timestamps and geographic locations, and generate a fused spatiotemporal traffic data set; Extract vehicle-mounted and navigation data from the spatiotemporal traffic dataset, and preliminarily divide the vehicles into vehicles with known navigation information and vehicles with unknown navigation information. For vehicles with known navigation information, determine whether their current position deviates from the navigation path. If they deviate from the navigation path, they are reclassified as vehicles with unknown navigation information, otherwise no processing is done; The spatiotemporal traffic data of vehicles with unknown navigation information are screened out from the spatiotemporal traffic data set, and the spatiotemporal traffic data of vehicles with unknown navigation information are processed using convolutional denoising autoencoders, spatiotemporal convolutional neural networks, and long short-term memory networks to generate a traffic flow prediction model; Extract the spatiotemporal traffic data of vehicles with known navigation information from the spatiotemporal traffic data set, calculate the proportion of vehicles with known navigation information in all running vehicles: p = number of vehicles with known navigation information / number of all running vehicles, set a loss function for vehicles with unknown navigation information and vehicles with known navigation information respectively, and assign a weight coefficient to the loss function of vehicles with known navigation information: ω = 1 / p, where ω represents the weight coefficient and p represents the proportion of vehicles with known navigation information in all running vehicles; The joint loss function is generated by the loss functions of the unknown navigation information vehicle and the known navigation information vehicle, as well as the weight coefficient: L_co=L_un+ω·L_kn, where L_co represents the joint loss function, L_un represents the loss function of the unknown navigation information vehicle, and L_kn represents the loss function of the known navigation information vehicle; The traffic flow prediction model is trained using the joint loss function and the spatiotemporal traffic data of vehicles with known navigation information, and the parameters of the traffic flow prediction model are updated through the back propagation algorithm until a preset number of training rounds is reached or the loss function converges; Based on the spatiotemporal traffic data of vehicles with known navigation information and vehicles with unknown navigation information, the traffic flow of the next road section is obtained. Combined with the saturation and traffic capacity of the next road section, the traffic index of the next road section is calculated. If the traffic index of the next road section is greater than the preset traffic threshold, the vehicle is guided to switch to the recommended alternative route.

2. The traffic flow induction monitoring method based on convolutional denoising autoencoder according to claim 1 is characterized in that: The vehicles are initially divided into vehicles with known navigation information and vehicles with unknown navigation information, including: Extract the vehicle-mounted data and navigation data of all vehicles from the fused spatiotemporal traffic dataset. The vehicle-mounted data includes the vehicle's unique identifier and real-time location information, and the navigation data includes the vehicle's departure point, destination, and planned navigation path information; Based on whether the vehicle is associated with valid navigation data, all vehicles are initially divided into vehicles with known navigation information and vehicles with unknown navigation information. Vehicles with known navigation information are vehicles whose navigation data contains navigation path information, and vehicles with unknown navigation information are vehicles that are not associated with navigation data or whose navigation data does not contain navigation path information.

3. The traffic flow induction monitoring method based on convolutional denoising autoencoder according to claim 2 is characterized in that: For a vehicle with known navigation information, determine whether its current position deviates from the navigation path, including: A path deviation threshold is set in advance. If the shortest distance between the current position of a vehicle with known navigation information and the navigation path exceeds the path deviation threshold, the vehicle is judged to have deviated from the navigation path, and the vehicle is removed from the list of vehicles with known navigation information and reclassified as a vehicle with unknown navigation information; if the shortest distance between the current position of the vehicle and the navigation path does not exceed the path deviation threshold, the vehicle is judged to have not deviated from the navigation path.

4. The traffic flow induction monitoring method based on convolutional denoising autoencoder according to claim 1 is characterized in that: Generate traffic flow prediction models, including: The spatiotemporal traffic data of vehicles with unknown navigation information are screened out from the spatiotemporal traffic data set, including the vehicle's video data, radar data, and vehicle-mounted data. The spatiotemporal traffic data of vehicles with unknown navigation information are denoised using a convolutional denoising autoencoder. The denoised spatiotemporal traffic data of vehicles with unknown navigation information are used as input, and high-order spatiotemporal features are extracted using a spatiotemporal convolutional neural network. The long short-term memory network is connected to perform sequence modeling on the extracted high-order spatiotemporal features; the output features of the spatiotemporal convolutional neural network and the long short-term memory network are spliced ​​to fuse the feature information from different network layers, and the fused features are sent to the fully connected layer for prediction to obtain a preliminary traffic flow prediction model.

5. The traffic flow induction monitoring method based on convolutional denoising autoencoder according to claim 1 is characterized in that: Get the traffic volume of the next road segment, including: When the vehicle starts the on-board navigation or navigation software, the spatiotemporal traffic data of all vehicles with known navigation information are used to count the traffic flow of all vehicles with known navigation information when the current vehicle is proceeding to the next section. The spatiotemporal traffic data of vehicles with unknown navigation information is used as input, and the traffic flow prediction model is used to predict the traffic flow of vehicles with unknown navigation information on the next section.

6. The traffic flow induction monitoring method based on convolutional denoising autoencoder according to claim 5 is characterized in that: Calculate the traffic index of the next road segment, including: The traffic index of the next road section is calculated by combining the traffic flow of vehicles with known navigation information and vehicles with unknown navigation information with the saturation and traffic capacity of the next road section. The calculation formula is as follows: Among them, TI represents the traffic index, Tv_u represents the traffic flow of vehicles with known navigation information, Tv_k represents the traffic flow of vehicles with unknown navigation information, Se represents saturation, Tc represents traffic capacity, and Ct represents the congestion threshold.

7. The traffic flow induction monitoring method based on convolutional denoising autoencoder according to claim 6 is characterized in that: When the traffic index of the next road segment is less than or equal to the preset traffic threshold, no processing is performed. When the traffic index of the next road segment is greater than the preset traffic threshold, alternative road segments are searched and the traffic indexes of all alternative road segments are calculated. If the traffic indexes of all candidate sections are greater than the preset traffic threshold, the section with the lowest traffic index is selected as the recommended section, and a navigation instruction is sent to the vehicle to guide the vehicle to switch to the recommended alternative route.

8. A traffic flow induction monitoring system based on convolutional denoising autoencoder, used to implement the method according to any one of claims 1 to 7, characterized in that: include: The data acquisition module collects video data, radar data, vehicle data, and navigation data of moving vehicles, matches them according to timestamps and geographic locations, and generates a fused spatiotemporal traffic data set; The vehicle classification module extracts vehicle-mounted and navigation data from the spatiotemporal traffic data set, and preliminarily divides the moving vehicles into vehicles with known navigation information and vehicles with unknown navigation information. For vehicles with known navigation information, it determines whether their current position deviates from the navigation path. If it deviates from the navigation path, it is reclassified as a vehicle with unknown navigation information, otherwise no processing is done; The model building module selects the spatiotemporal traffic data of vehicles with unknown navigation information from the spatiotemporal traffic data set, processes the spatiotemporal traffic data of vehicles with unknown navigation information using convolutional denoising autoencoders, spatiotemporal convolutional neural networks, and long short-term memory networks to generate a traffic flow prediction model; The model training and optimization module extracts the spatiotemporal traffic data of vehicles with known navigation information from the spatiotemporal traffic data set, calculates the proportion of vehicles with known navigation information in all running vehicles: p = number of vehicles with known navigation information / number of all running vehicles, sets a loss function for vehicles with unknown navigation information and vehicles with known navigation information respectively, and assigns a weight coefficient to the loss function of vehicles with known navigation information: ω = 1 / p, where ω represents the weight coefficient and p represents the proportion of vehicles with known navigation information in all running vehicles; The joint loss function is generated by the loss functions of the unknown navigation information vehicle and the known navigation information vehicle, as well as the weight coefficient: L_co=L_un+ω·L_kn, where L_co represents the joint loss function, L_un represents the loss function of the unknown navigation information vehicle, and L_kn represents the loss function of the known navigation information vehicle; The traffic flow prediction model is trained using the joint loss function and the spatiotemporal traffic data of vehicles with known navigation information, and the parameters of the traffic flow prediction model are updated through the back propagation algorithm until a preset number of training rounds is reached or the loss function converges; The traffic flow induction decision module obtains the traffic volume of the next road section based on the spatiotemporal traffic data of vehicles with known navigation information and vehicles with unknown navigation information, and calculates the traffic index of the next road section based on the saturation and traffic capacity of the next road section. If the traffic index of the next road section is greater than the preset traffic threshold, the vehicle is guided to switch to the recommended alternative route.

Images