An intelligent traffic control method and system based on real-time vehicle flow detection

Abstract

The present invention proposes an intelligent traffic control method and system based on real-time vehicle flow detection, which belongs to the field of intelligent traffic control technology. The problems it addresses are: the existing technology has single vehicle flow detection, inaccurate data collection, lack of overall network analysis, inaccurate vehicle flow prediction, and unreasonable signal light timing. It includes: obtaining vehicle transmission information data, supplementing and correcting the vehicle transmission information data through mobile terminals and road network monitoring data, and performing edge computing analysis, pre-processing the parsed vehicle information, and constructing a prediction model through machine learning algorithms to predict traffic flow; using a graph theory model to describe intersection relationships and create a self-organizing traffic network; defining three indicators based on the predicted traffic flow and constructing a traffic light timing calculation model; using a chaotic particle swarm algorithm to calculate the optimal timing strategy to achieve collaborative control. The present invention improves the accuracy of vehicle flow prediction and the rationality of signal light timing.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent traffic control, and in particular relates to an intelligent traffic control method and system based on real-time vehicle flow detection. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Currently, traffic lights are installed at numerous intersections in cities to facilitate the passage of motor vehicles and pedestrians according to traffic light regulations. However, if the traffic light release times at multiple intersections are not set properly, it will lead to traffic congestion, making the traffic conditions in the entire city very complicated.

[0004] Traditional traffic control systems usually use fixed signal timing, and each intersection often operates independently, lacking coordination and optimization. It cannot be flexibly adjusted according to real-time traffic conditions, which can easily lead to reduced traffic efficiency.

[0005] Therefore, with the development of the automobile industry, modern urban transportation has gradually developed towards intelligent transportation, and has gradually begun to adopt artificial intelligence technology to design signal control systems, and has begun to apply various intelligent algorithms to traffic control systems.

[0006] However, existing intelligent traffic control systems rely on a single method for detecting traffic flow, resulting in low data accuracy and a failure to meet the requirements of intelligent transportation systems. Traditional systems also fail to manage multiple adjacent intersections as a complex network, leading to inaccurate overall predictions and inefficient traffic management during peak traffic periods. Existing traffic control technologies cannot guarantee accurate traffic flow predictions or the rationality of signal timing, nor can they improve traffic efficiency. Summary of the Invention

[0007] To overcome the shortcomings of the aforementioned prior art, the present invention provides an intelligent traffic control method and system based on real-time traffic flow detection. This method breaks away from the traditional fixed traffic light timing scheme and links the traffic light timing cycle with traffic flow data within the network. This allows for dynamic traffic light timing based on real-time traffic flow detection, enabling real-time adjustment of intersection traffic light timing and improving traffic efficiency. Traffic flow is calculated using a variety of methods, including road network image recognition, mobile terminal quantity estimation, and vehicle network data reporting. Traffic flow data is calculated by treating multiple adjacent intersections as a complex network, improving the accuracy of traffic flow predictions and the rationality of traffic light timing.

[0008] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0009] A first aspect of the present invention provides an intelligent traffic control method based on real-time vehicle flow detection.

[0010] An intelligent traffic control method based on real-time traffic flow detection, comprising:

[0011] Obtain vehicle transmission information data through the intelligent connected vehicle OBU and perform corrections;

[0012] Perform edge computing analysis and preprocessing on the corrected vehicle transmission information data;

[0013] Based on the pre-processed vehicle information, a prediction model is constructed through machine learning algorithms to predict traffic flow;

[0014] Use graph theory models to describe intersection relationships, analyze network characteristics, and create self-organizing transportation networks;

[0015] Based on the predicted traffic flow, three indicators are defined: average queue length, average waiting time, and area entry and exit ratio;

[0016] Based on the self-organizing traffic network and three indicators, a traffic light timing calculation model is constructed;

[0017] Based on the traffic light timing calculation model, the chaotic particle swarm algorithm is used to calculate the optimal timing strategy and coordinate the control of multiple intersections in the area.

[0018] Among them, based on the self-organizing traffic network and three indicators, a traffic light timing calculation model is constructed, and the formula is:

[0019]

[0020] Among them, α, β, γ and δ are weight coefficients. represents the average delay time at the intersection; l q (X) is the average queue length of vehicles at the intersection; n is the number of intersections; ρ(X) is the vehicle entry and exit ratio of the area; q 预测 represents the traffic flow predicted by the LSTM model; q 实际 Represents the actual traffic flow; ∈·∑ n (λ n -λ avg ) 2 is the regularization term, λ n Indicates the green-to-signal ratio, the ratio of green light time to the total signal cycle time; λ avg Represents the average green-to-signal ratio.

[0021] As an implementation method, the vehicle transmission information data is corrected to obtain the corrected vehicle transmission information data. The specific process is as follows:

[0022] Real-time collection of terminal connection data of roadside 5G base stations;

[0023] Estimate regional vehicle density based on terminal connection data from roadside 5G base stations, and supplement vehicle transmission data;

[0024] Based on road network monitoring data, AI algorithms are used to calculate the number of vehicles on the road;

[0025] Based on the regional vehicle density and the number of vehicles on the road, the vehicle transmission information data is corrected through the spatiotemporal alignment method to obtain the corrected vehicle transmission information data.

[0026] As an implementation method, based on the pre-processed vehicle transmission information, a prediction model is constructed by a machine learning algorithm to predict traffic flow, wherein the prediction model is an LSTM model, specifically:

[0027] Get historical traffic data;

[0028] Input the historical traffic data into the LSTM model, perform model fitting training, and obtain a trained LSTM model;

[0029] The preprocessed vehicle transmission information is input into the trained LSTM model to predict the traffic flow at the next moment.

[0030] As an implementation method, a graph theory model is used to describe intersection relationships, analyze network characteristics, and create a self-organizing traffic network. Specifically, based on the graph theory model, the intersection relationships are described, and vehicles and roadside facilities are used as network nodes, equipped with communication and computing equipment, and C-V2X and Wi-Fi Direct technologies are used to dynamically connect them and self-organize the network topology.

[0031] As an implementation method, the formula for calculating the average vehicle queue length is:

[0032]

[0033] Among them, l q is the average queue length of vehicles; n is the number of intersections; represents the remaining queue length at intersection i; represents the current queue length at intersection i; a and b are weight coefficients, representing the contribution ratio of the remaining queue length and the newly added queue length, respectively.

[0034] As an implementation method, a chaotic particle swarm algorithm is used to calculate the optimal timing strategy, specifically:

[0035] Based on a self-organizing traffic network, with the optimization objectives of minimizing the average queue length, the average waiting time, and the area entry-exit ratio, a chaotic particle swarm optimization algorithm is used to obtain the optimal solution of the traffic light timing calculation model.

[0036] The optimal solution of the traffic light timing calculation model is converted into signal light control instructions to obtain the optimal timing strategy.

[0037] A second aspect of the present invention provides an intelligent traffic control system based on real-time vehicle flow detection, comprising:

[0038] The data acquisition module is configured to: obtain vehicle transmission information data through the intelligent connected vehicle OBU and perform correction;

[0039] The edge computing module is configured to: perform edge computing analysis on the corrected vehicle transmission information data and perform preprocessing;

[0040] The core computing module is configured to: build a prediction model based on pre-processed vehicle information through machine learning algorithms to predict traffic flow;

[0041] Use graph theory models to describe intersection relationships, analyze network characteristics, and create self-organizing transportation networks;

[0042] Based on the predicted traffic flow, three indicators are defined: average queue length, average waiting time, and area entry and exit ratio;

[0043] Based on the self-organizing traffic network and three indicators, a traffic light timing calculation model is constructed;

[0044] Based on the traffic light timing calculation model, the chaotic particle swarm algorithm is used to calculate the optimal timing strategy and coordinate the control of multiple intersections in the area.

[0045] Among them, based on the self-organizing traffic network and three indicators, a traffic light timing calculation model is constructed, and the formula is:

[0046]

[0047] Among them, α, β, γ and δ are weight coefficients. represents the average delay time at the intersection, i.e. the average waiting time; l q (X) is the average queue length of vehicles at the intersection; n is the number of intersections; ρ(X) is the vehicle entry and exit ratio of the area; q 预测 represents the traffic flow predicted by the LSTM model; q 实际 Represents the actual traffic flow; ∈·∑ n (λ n -λ avg ) 2 is the regularization term, λn Indicates the green-to-signal ratio, the ratio of green light time to the total signal cycle time; λ avg Represents the average green-to-signal ratio.

[0048] The third aspect of the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in a method as described in the first aspect of the present invention are implemented.

[0049] A fourth aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps of a method as described in the first aspect of the present invention.

[0050] A fifth aspect of the present invention provides a computer program product comprising instructions, which, when run on a computer, enables the computer program to be executed by a processor to implement the steps in a method as described in the first aspect of the present invention.

[0051] One or more of the above technical solutions have the following beneficial effects:

[0052] The present invention adopts multiple data source collection methods, uses 5G smart devices as a supplement to traffic flow data, introduces more mature road network monitoring to correct traffic flow data, and improves the accuracy of traffic flow measurement.

[0053] This invention enhances vehicle-road collaboration by utilizing next-generation 5G communication technology and integrating it with the industry-standard C-V2X vehicle-to-everything (C-V2X) technology. This enables real-time communication between vehicles (V2V), between vehicles and roadside infrastructure (V2I), and between vehicles and pedestrians (V2P). By leveraging V2P and V2I to predict traffic control signals, forward traffic flow, speed limit signs, road slope, and other information, neighboring vehicles can plan their routes more accurately, improving the accuracy of vehicle-road collaborative control.

[0054] The present invention adopts a graph theory model to describe the relationship between intersections, manages multiple adjacent intersections as a complex network, improves the accuracy of traffic flow prediction and the rationality of timing, enhances the management and optimization capabilities of the overall traffic network, and realizes the optimization of the traffic network.

[0055] The present invention collects traffic vehicle information, defines three timing indicators for signal light timing, and uses a chaotic particle swarm algorithm to dynamically adjust signal light timing, thereby reducing urban traffic congestion, improving road capacity, and enhancing traffic efficiency.

[0056] The overall effect of this invention is that, under ideal environmental conditions, long-distance congestion at a single intersection is reduced by 20%, and the average waiting time for a single vehicle is reduced by 15%. It establishes communication between basic roadside traffic facilities, vehicles, and pedestrians, enabling information sharing and laying the foundation for vehicle-road collaborative control.

[0057] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0059] Figure 1 This is a flow chart of an intelligent traffic control method based on real-time vehicle flow detection according to the first embodiment of the present invention;

[0060] Figure 2 This is a schematic diagram of a framework of an intelligent traffic control system based on real-time vehicle flow detection according to a second embodiment of the present invention;

[0061] Figure 3 This is a schematic diagram of an obvious congestion situation on the road under the fixed timing according to the first embodiment of the present invention;

[0062] Figure 4 This is a schematic diagram of a simulation experiment of an intelligent traffic control method based on real-time vehicle flow detection according to the first embodiment of the present invention. DETAILED DESCRIPTION

[0063] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0064] It should be noted that the terms used herein are for describing particular embodiments only and are not intended to limit the exemplary embodiments according to the present invention.

[0065] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0066] Example 1

[0067] This embodiment discloses an intelligent traffic control method based on real-time traffic flow detection. It collects traffic vehicle information, defines three timing indicators for signal light timing, and uses a chaotic particle swarm optimization algorithm to achieve dynamic signal light timing adjustment. This method solves the traffic congestion problem caused by traditional fixed signal light timing, improves the accuracy of traffic flow measurement, and enhances the management and optimization capabilities of the overall traffic network. The method includes:

[0068] S1. Obtain vehicle transmission information data through the intelligent connected vehicle OBU and perform correction;

[0069] S2. Perform edge computing analysis and preprocessing on the corrected vehicle transmission information data;

[0070] S3. Based on the pre-processed vehicle information, a prediction model is constructed through machine learning algorithms to predict traffic flow;

[0071] S4. Use graph theory models to describe intersection relationships, analyze network characteristics, and create a self-organizing transportation network;

[0072] S5. Based on the predicted traffic flow, define three indicators: average queue length, average waiting time, and area entry and exit ratio;

[0073] Based on the self-organizing traffic network and three indicators, a traffic light timing calculation model is constructed;

[0074] The chaotic particle swarm algorithm is used to calculate the optimal timing strategy and coordinate the control of multiple intersections in the area.

[0075] In order to explain this embodiment more clearly, Figure 1 As shown in FIG, the implementation process of an intelligent traffic control method based on real-time traffic flow detection can be specifically described as follows:

[0076] like Figure 1 As shown, in step S1, vehicle transmission information data is obtained through the intelligent connected vehicle OBU and corrected to obtain corrected vehicle transmission information data.

[0077] In this embodiment, data is collected through intelligent connected vehicle OBU and real-time road network monitoring (implemented by industrial cameras + visual video), and the collected data is transmitted to the edge computing device.

[0078] S1-1. Obtain vehicle transmission information data through the intelligent connected vehicle OBU.

[0079] In this embodiment, the vehicle-mounted OBU obtains vehicle transmission information data, including real-time vehicle position, speed, and vehicle density.

[0080] S1-2. Supplement and correct the vehicle transmission information data to obtain corrected vehicle transmission information data.

[0081] Since the driving data uploaded by the OBU is the most accurate and reliable data, it is used as the direct data source in this embodiment. However, considering that the number of connected cars on the market that support OBU data transmission is insufficient, it will cause large errors in data collection. Therefore, two other data sources are designed for supplementation and correction.

[0082] Real-time collection of terminal connection data of roadside 5G base stations and real-time road network monitoring data are obtained, and vehicle transmission information data is supplemented and corrected through 5GC positioning enhancement service technology and image recognition target detection technology respectively.

[0083] Two methods for supplementing and correcting data sources include:

[0084] (1) And supplement the vehicle transmission information data.

[0085] Obtain the number of mobile terminals, estimate the regional vehicle density based on the number of mobile phones connected to the 5G base station (assuming the average passenger capacity of a single vehicle), and supplement the vehicle transmission information data.

[0086] In this embodiment, the number of smart mobile terminals is collected: Taking into account the problem of insufficient number of connected cars at this stage, this embodiment uses smart mobile terminals as a supplementary data source in actual applications. By utilizing the precise positioning technology in the 5GC-R17 standard, information is broadcast to smart mobile terminals within the coverage of the 5G base station. The average number of passengers per vehicle and the location of the smart terminal during this period are used to calculate the current traffic flow in all directions at the intersection, thereby enhancing the data accuracy during periods when the number of connected cars is insufficient.

[0087] Real-time terminal connection data from roadside 5G base stations is collected. Using 5GC positioning enhancement services, high-precision positioning of mobile terminals is achieved. The number of 5G smart mobile terminals currently on the road is calculated. Based on the average number of mobile terminals per vehicle, the number of vehicles on the road at that time can be estimated, supplemented, and corrected for vehicle density. This can then be combined with dynamic traffic light timing solutions to provide more appropriate route planning for vehicles on the road.

[0088] (2) Visual image data correction.

[0089] Obtain road network monitoring data and process it. Specifically:

[0090] First, the captured video stream is preprocessed using the Canny algorithm for edge detection and median filtering for denoising. After this initial processing, the image is passed through the YOLOv5 object detection algorithm for real-time object detection and recognition, enabling accurate vehicle counting and location. Finally, the non-maximum suppression (NMS) algorithm is used to remove overlapping detection frames, reducing false and missed detections.

[0091] Then, use cameras and AI algorithms (such as YOLOv5 target detection) to directly count the number of vehicles on the road to correct traffic flow.

[0092] In this embodiment, the existing traffic management system is connected to the road network, using real-time monitoring images. Image recognition and target detection technologies are used for analysis to avoid the risk of unreliable user-side data. During the simulation phase, industrial cameras can be used to simulate and capture simulated cars on the road to correct traffic flow.

[0093] Specifically, through spatiotemporal alignment, OBU data, data collected by smart terminals, and road network monitoring data are mapped into a unified integrated time series feature vector to make up for the coverage blind spots of a single data source.

[0094] After the above steps, the corrected vehicle transmission information data is finally obtained, which fills the coverage blind spot of a single data source and avoids errors in data collection.

[0095] Step 2: Perform edge computing analysis on the corrected vehicle transmission information data to obtain the analyzed vehicle information; and perform preprocessing to obtain the preprocessed vehicle information.

[0096] In this embodiment, edge computing devices are deployed on roadside facilities (traffic light host computers), which can process and aggregate information sent by vehicles locally, thereby achieving rapid response and real-time control, providing more appropriate route planning for road vehicles, and optimizing traffic flow.

[0097] The specific steps include:

[0098] Step 2-1: Perform edge computing analysis on the corrected vehicle transmission information data to obtain the analyzed vehicle information.

[0099] Receive vehicle transmission information and parse it:

[0100] The edge computing device receives information sent by the vehicle through the Uu interface and PC5 interface in the C-V2X standard; then, it parses and filters the received information to obtain the parsed vehicle information data, which includes the vehicle's location, speed and density.

[0101] To ensure the security and reliability of data transmission, in this embodiment, RabbitMQ message queue is used to cache and load balance received data packets to reduce packet loss caused by excessive load on a single node.

[0102] After the above steps, the accuracy and reliability of the information are ensured.

[0103] Step 2-2: Preprocess the parsed vehicle information to obtain preprocessed vehicle information.

[0104] Vehicle information processing: Edge computing devices use their own processing capabilities to further process and calculate the parsed vehicle information, primarily including statistics and analysis of vehicle density and speed.

[0105] Step 2-3: Data reporting:

[0106] To improve service compatibility with edge computing facilities, this project can use HTTP and WebSocket protocols to achieve real-time data transmission. At the same time, considering the high amount of duplicate data and high data compressibility, this project compresses the data before transmission.

[0107] In this embodiment, the cloud computing as a whole adopts SpringBoot as the server framework, which mainly accepts long connection data sent by the WebSocket protocol and short-term data sent by HTTP.

[0108] The pre-processed vehicle speed, density and vehicle position are uploaded.

[0109] Step 3: Based on the preprocessed vehicle information, a prediction model is constructed through machine learning algorithms to predict traffic flow.

[0110] In this embodiment, prediction is performed using an LSTM model. The specific prediction process is as follows:

[0111] (1) The traffic data of historical periods are saved through memory units and used as training data for model fitting.

[0112] In the traffic flow prediction scenario, the fitting model in this embodiment uses the long short-term memory network (LSTM) model. The historical traffic data is integrated with the vehicle OBU data, the number of mobile terminal network accesses, and road network monitoring data to build a dynamic prediction framework oriented to time dependency. The specific process is as follows:

[0113] 1) Build a prediction model.

[0114] LSTM (Long Short-Term Memory) is a special recurrent neural network (RNN) that solves the problem of traditional RNN's insufficient ability to model long-term dependencies through internal "memory units" and "gating mechanisms".

[0115] 2) Input the historical traffic data into the LSTM model, perform model fitting training, and obtain a trained LSTM model.

[0116] The preprocessed vehicle transmission information is input into the trained LSTM model to predict the traffic flow at the next moment.

[0117] The memory unit stores historical key information (such as periodic traffic during morning and evening rush hours), and the forget gate / input gate dynamically determines which information to retain or discard (for example, a sudden traffic accident will weaken the focus on regular traffic).

[0118] The output gate generates the current prediction result, and the trained prediction model (LSTM model) is obtained.

[0119] (2) Dynamic prediction and adaptive optimization.

[0120] To achieve real-time response and long-term adaptability, the system adopts a two-stage mechanism:

[0121] Edge computing inference: Deploy the trained LSTM model to the roadside edge server, directly process the local data stream, and complete the prediction within 5 minutes (reducing cloud transmission latency);

[0122] Online incremental learning: Regularly fine-tune model parameters with newly added data to continuously improve predictive capabilities as traffic patterns evolve (such as new road openings and changes in holiday traffic).

[0123] (3) The pre-processed vehicle information is input into the trained prediction model to predict traffic flow.

[0124] Based on historical traffic data (OBU, 5G base stations, and road network monitoring), a trained LSTM model is obtained. Based on real-time input, key indicators such as traffic volume, queue length, and delay time at each intersection in the next 5-15 minutes are predicted. In other words, the traffic volume at the intersection at the next moment is obtained, which is expressed as:

[0125] q 预测 =LSTM(vehicle position, vehicle password, vehicle speed).

[0126] Step 4: Use graph theory models to describe intersection relationships, analyze network characteristics, and create a self-organizing transportation network.

[0127] This example introduces network science concepts, treating each intersection as a network node and connecting roads as edges to build a graph model. This model describes intersection relationships. By combining multiple intersections into a single network for research, a self-organizing traffic network is created, improving adaptability to complex road conditions, system reliability, and traffic flow forecast accuracy and scheduling rationality.

[0128] Vehicles and roadside infrastructure are used as network nodes, equipped with communication and computing devices, and leveraging technologies such as C-V2X and Wi-Fi Direct to achieve dynamic connectivity and self-organizing network topologies. Nodes can instantly exchange information such as traffic conditions and accident alerts.

[0129] By analyzing network characteristics, traffic management decisions are supported. Using network science methods, key intersections are identified to optimize traffic planning, dynamically adjust traffic light timing, and reduce congestion. Compared to traditional centralized network architectures, self-organizing networks can better adapt to complex road conditions and provide higher reliability.

[0130] Step 5: Define the three indicators of average queue length, average waiting time, and area entry and exit ratio based on the predicted traffic flow;

[0131] Based on the self-organizing traffic network and three indicators, a traffic light timing calculation model is constructed;

[0132] The chaotic particle swarm algorithm is used to calculate the optimal timing strategy and coordinate the control of multiple intersections in the area.

[0133] (1) Based on the predicted traffic flow, three indicators are defined: average queue length, average waiting time, and area entry and exit ratio.

[0134] An algorithm is established with the optimization objectives of minimizing the average queue length of vehicles, minimizing the average vehicle delay time, and minimizing the vehicle entry and exit ratio in the area; a timing model is constructed based on the chaotic particle swarm algorithm, and the optimal state is achieved by optimizing the above indicators.

[0135] In this embodiment, the average queue length, average vehicle delay time, and vehicle entry and exit ratio of each of the five intersections (including the current intersection and the four adjacent intersections in the east, west, south, and north) are used as input parameters to establish a traffic light timing calculation model. The optimal timing strategy is finally calculated. The specific model is as follows:

[0136] Calculation of average vehicle queue length:

[0137]

[0138] Among them, l q is the average queue length of vehicles; n is the number of intersections; represents the remaining queue length at intersection i; represents the current queue length at intersection i; a and b are weight coefficients, representing the contribution ratio of the remaining queue length and the newly added queue length, respectively, for example, a = 0.7, b = 0.3. By adjusting the weights, queue assessment can be flexibly optimized for different congestion scenarios (for example, paying more attention to newly added queues during peak hours);

[0139]

[0140] l q is the average queue length of vehicles; l s is the remaining vehicle queue length; l p is the length of the queued vehicles; n is the number of intersections; g i1 represents the effective green light time of the upstream intersection of the i-th intersection; g i2 represents the effective green light time of the intersection downstream of the i-th intersection; v represents the average vehicle speed; v t is the parking wave speed; T s=1 means the last vehicle arrives at the road section within the first red light; T s =2 means the last vehicle arrives at the road section within the second red light; q Indicates the starting wave speed; T t Indicates the stop signal interval where the first vehicle arrives at the road section; l max Indicates the maximum vehicle parking queue length.

[0141] The vehicle delay time is calculated using the following formula:

[0142]

[0143] in,

[0144]

[0145] T delay is the delay time of all intersections, n is one of all intersections; k represents an entrance road in intersection n; r n represents all entrances of intersection n; η k Indicates saturation, s k is the lane saturation flow; q k represents the road traffic flow at intersection entrance k; k represents the green signal ratio of intersection entrance k; d k (w,λ n , o) represents the delay function, which is used to describe the impact of intersection turning ratio on delay (such as the proportion of left-turning vehicles), w represents the turning ratio, λ n represents the green-to-signal ratio, which is the ratio of green light time to the total signal cycle time. o represents the relative importance of balancing the green-to-signal ratio (the ratio of green light time to the total cycle) and the turn ratio (w) in calculating the delay of a single entrance lane. represents the signal period; q total It represents the sum of traffic flows at all entrances of intersection n and is used to balance delay calculations under different traffic scales.

[0146] The formula for calculating the vehicle entry and exit ratio of a region is:

[0147]

[0148] Where ρ is the area entry and exit ratio, that is, the ratio of the maximum number of output vehicles to the total number of vehicles entering the area, which is set as the objective function, Q is the total number of vehicles entering the area; t = 1 represents the first signal cycle, t = 1, ..., n, f = 1 represents the first phase, F is the maximum number of phases; l out is the number of entrance channels in the boundary area; is the number of vehicles coming out of entrance lane k within t signal cycles,

[0149]

[0150] Is the entrance k light green when the phase is q? If yes, it is 1, otherwise it is 0; k (t) is the road traffic flow at intersection k in the tth signal cycle; is the effective green light time of the fth phase of intersection n in the tth signal cycle; Delayed vehicles at the entrance lane k in the fth phase of the tth signal cycle; s k is the lane saturation flow.

[0151] (2) Based on the self-organizing traffic network and three indicators, a traffic light timing calculation model is constructed and optimized.

[0152] In this example, a regional traffic signal optimization model was established with the optimization objectives of minimizing average vehicle delay time, minimizing average queue length, and minimizing regional entry and exit ratio. Assuming that the number of intersections in the region is n, and each intersection has f phases, the optimization vector is:

[0153] X=[λ1,t1,λ2,t2,…,λ m ,t m ](4)

[0154] Among them, λ m represents the green signal ratio at the mth intersection, t m Indicates the effective green light time at intersection m.

[0155] The final optimization model formula is:

[0156]

[0157] Among them, α, β, γ and δ are weight coefficients. represents the average delay time at the intersection; l q (X) is the average queue length of vehicles at the intersection; n is the number of intersections; ρ(X) is the vehicle entry and exit ratio of the area; q 预测 represents the traffic flow predicted by the LSTM model; q 实际 Represents the actual traffic flow; ∈·∑ n (λ n -λ avg ) 2 is the regularization term, λ n Indicates the green-to-signal ratio, the ratio of green light time to the total signal cycle time; λ avg represents the average green-to-signal ratio, By regularizing the fluctuation range of the green-to-light ratio, we can prevent the green light time at a single intersection in the timing plan from being too long or too short, thereby improving coordination stability.

[0158] The constraints are

[0159] Among them, T represents the traffic light cycle, T min Indicates the minimum cycle of the traffic light, that is, the shortest time to complete a traffic light, T max Indicates the maximum cycle of the traffic light, that is, the maximum time to complete a traffic light, which limits the cycle range of the traffic light; g indicates the time the green light is on, g min Indicates the shortest time the green light is on, g max Indicates the maximum time the green light is on.

[0160] (3) Using the chaotic particle swarm algorithm, the optimal timing strategy is calculated.

[0161] The prediction capability of the LSTM model is deeply integrated with the dynamic decision-making capability of the traffic light timing model, and the real-time response and global optimization capabilities of the transportation system are improved through multi-dimensional collaboration.

[0162] The LSTM model uses historical traffic data (from OBUs, 5G base stations, and road network monitoring) and real-time input to predict key indicators such as traffic volume, queue length, and delay time at each intersection within the next 5-15 minutes. The traffic light timing model uses the LSTM predictions as input and combines them with current traffic conditions (such as queue length lq and delay time Tdelay) to solve the optimal timing solution using a chaotic particle swarm algorithm.

[0163] The optimal timing strategy refers to the length of the next cycle timing for multiple directions at the current intersection.

[0164] Traffic light timing converts traffic flow data into specific green light durations and phase switching strategies. Each indicator in formulas (1), (2), and (3) (such as delay time and queue length) is directly related to the actual control parameters of the traffic light, and the chaotic particle swarm algorithm finds the optimal balance point through iterative optimization. Ultimately, the system achieves real-time and dynamic traffic control through edge computing and V2I communication, significantly improving road network efficiency.

[0165] Based on the constraints, with the optimization objectives of minimizing the average vehicle delay time, minimizing the average queue length, and minimizing the area entry and exit ratio, the traffic light timing model formula (5) is solved to obtain the optimal solution, that is, the optimal timing strategy, and the optimal timing strategy is converted into signal control instructions.

[0166] (4) After obtaining the optimal timing strategy, instructions are issued to coordinate control of multiple intersections in the area.

[0167] Instruction issuance means that the data center will send the optimal timing strategy (i.e. the optimal timing for the next cycle) obtained through the traffic light timing calculation model to the edge device, and the edge device will issue specific timing adjustment instructions.

[0168] The instructions include periodic time adjustment, forced yellow flash, forced red light, return to normal and other operations, which are used to control the operation of the entire system.

[0169] Coordinated control of multiple intersections within the area. The specific process is as follows:

[0170] ① Command Receiving: The command source can be a command sent through the communication network or a command from the local control center. The command issuing module needs to establish a stable communication connection and have a command receiving and caching mechanism to ensure reliable command transmission.

[0171] ② Instruction Parsing: After receiving an instruction, the instruction issuing module needs to parse it to extract the instruction's operational requirements and parameters. This parsing process should use standard parsing algorithms and rules to correctly parse the instruction and extract key information. The parsed instruction information will be used for subsequent instruction execution.

[0172] ③ Command Execution: Based on the parsed command information, the command issuing module calls the corresponding control interface to operate the traffic light. Specific operations may include changing the traffic light's timing, adjusting the phase of the traffic light, or triggering an emergency signal. When executing commands, the correctness and safety of the operation must be ensured to avoid adverse effects on traffic operations.

[0173] ④ Execution Result Feedback: After executing a command, the command issuing module needs to provide feedback to the upper-level management system. This feedback should include the command's execution status, results, and related parameter information. The upper-level management system can make decisions and adjustments based on this feedback.

[0174] In order to verify the effectiveness of this embodiment, the existing high-performance computing cluster in the laboratory is used to simulate the cloud computing center, and the data collected by the edge computing server is processed and summarized, and simulation is performed through software, such as Figure 3 and Figure 4 As shown in the figure, when the traffic volume is large, if the signal duration cannot be effectively adjusted, it will cause congestion on the entire road, leading to traffic paralysis. Figure 3 The figure shows the obvious congestion on the road under fixed timing. Figure 4 As shown, after the control method in this embodiment is introduced, the vehicle congestion situation is significantly alleviated.

[0175] According to the signal light timing algorithm designed in this embodiment, the green phase duration in different directions is adjusted. It is found that the above-mentioned large traffic flow can be reasonably regulated to optimize road traffic congestion, better adapt to changes in traffic flow, avoid congestion and delays, and improve road utilization.

[0176] Example 2

[0177] The purpose of this embodiment is to provide an intelligent traffic control system based on real-time traffic flow detection, including:

[0178] The data acquisition module is configured to: obtain vehicle transmission information data through the intelligent connected vehicle OBU and perform correction;

[0179] The edge computing module is configured to: perform edge computing analysis on the corrected vehicle transmission information data and perform preprocessing;

[0180] The core computing module is configured to: build a prediction model based on pre-processed vehicle information through machine learning algorithms to predict traffic flow;

[0181] Use graph theory models to describe intersection relationships, analyze network characteristics, and create self-organizing transportation networks;

[0182] Based on the predicted traffic flow, three indicators are defined: average queue length, average waiting time, and area entry and exit ratio;

[0183] Based on the self-organizing traffic network and three indicators, a traffic light timing calculation model is constructed;

[0184] Based on the traffic light timing calculation model, the chaotic particle swarm algorithm is used to calculate the optimal timing strategy and coordinate the control of multiple intersections in the area.

[0185] Among them, based on the self-organizing traffic network and three indicators, a traffic light timing calculation model is constructed, and the formula is:

[0186]

[0187] Among them, α, β, γ and δ are weight coefficients. represents the average delay time at the intersection; l q (X) is the average queue length of vehicles at the intersection; n is the number of intersections; ρ(X) is the vehicle entry and exit ratio of the area; q 预测 represents the traffic flow predicted by the LSTM model; q 实际 Represents the actual traffic flow; ∈·∑ n (λ n -λ avg ) 2 is the regularization term, λ nIndicates the green-to-signal ratio, the ratio of green light time to the total signal cycle time; λ avg Represents the average green-to-signal ratio.

[0188] The data acquisition module includes vehicle-mounted OBU, 5G smart mobile terminal and image data interface.

[0189] The edge computing module mainly includes roadside edge computing facilities (ECN).

[0190] The main functions of the cloud computing part include data receiving module, core computing module, data storage module, instruction issuing module, and cloud-edge collaborative unloading module.

[0191] The data receiving module features a universal and scalable data access layer compatible with various communication protocols, such as RESTful and WebSocket, ensuring seamless data transfer to the cloud. To address data spikes, Redis is used as an in-memory cache to store some of the received data, ensuring data loss and significantly reducing backend processing pressure. Kafka, a message queue, is deployed for asynchronous data processing, achieving data decoupling and peak-load shifting.

[0192] In the core computing module, a chaotic particle swarm algorithm is employed to define three metrics based on predicted traffic flow: average queue length, average waiting time, and regional entry / exit ratio. These metrics are used as input parameters, combined with the self-organizing traffic network to establish a traffic light timing calculation model. This model calculates the optimal timing strategy, enabling coordinated control of multiple intersections within the region and optimizing traffic flow across the entire road network through vehicle-to-infrastructure (V2I) communication. This ensures the module can respond efficiently and accurately to complex road conditions, effectively improving commuting efficiency at intersections.

[0193] This embodiment also includes a data storage module, which primarily stores and manages various data generated during system operation, including traffic flow data collected by the data collection component, traffic light cycle timing strategies, and details of cloud-edge collaborative task dispatches. This module is designed to utilize the government cloud object storage and existing cloud services for implementation. Data security is handled by the government cloud maintainer.

[0194] Instruction issuing module: mainly issues instructions or cloud-edge collaborative computing instructions based on the calculation results of the core computing module.

[0195] Cloud-edge collaborative offloading module: The main function of this module is to detect the load details of each edge device, and according to the cloud load and the load of each edge node, send collaborative computing tasks to the edge node computing module on demand to enhance data response speed and processing efficiency.

[0196] In this embodiment, the entire system adopts SpringBoot as the server framework, which mainly accepts long connection data sent by the WebSocket protocol and short-term data sent by HTTP.

[0197] Example 3

[0198] The purpose of this embodiment is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the program.

[0199] Example 4

[0200] The purpose of this embodiment is to provide a computer-readable storage medium.

[0201] A computer-readable storage medium stores a computer program, which, when executed by a processor, performs the steps of the above method.

[0202] Example 5

[0203] The purpose of this embodiment is to provide a computer program product containing instructions, which, when running on a computer, enables the computer to execute the methods and functions involved in any of the above embodiments.

[0204] The steps involved in the apparatus of the above embodiment correspond to those of the method embodiment 1. For detailed implementation, please refer to the relevant description of embodiment 1. The term "computer-readable storage medium" should be understood to mean a single medium or multiple media containing one or more instruction sets; it should also be understood to include any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and causing the processor to perform any method of the present invention.

[0205] Those skilled in the art will appreciate that the modules or steps of the present invention described above can be implemented using a general-purpose computer device. Alternatively, they can be implemented using program code executable by a computing device, which can then be stored in a storage device and executed by the computing device. Alternatively, they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0206] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. An intelligent traffic control method based on real-time traffic flow detection, characterized in that: include: Obtain vehicle transmission information data through the intelligent connected vehicle OBU and perform corrections; Perform edge computing analysis and preprocessing on the corrected vehicle transmission information data; Based on the pre-processed vehicle information, a prediction model is constructed through machine learning algorithms to predict traffic flow; Use graph theory models to describe intersection relationships, analyze network characteristics, and create self-organizing transportation networks; Based on the predicted traffic flow, three indicators are defined: average queue length, average waiting time, and area entry and exit ratio; Based on the self-organizing traffic network and three indicators, a traffic light timing calculation model is constructed; Based on the traffic light timing calculation model, the chaotic particle swarm algorithm is used to calculate the optimal timing strategy and coordinate the control of multiple intersections in the area. Among them, based on the self-organizing traffic network and three indicators, a traffic light timing calculation model is constructed, and the formula is: Among them, α, β, γ and δ are weight coefficients. represents the average delay time at the intersection; l q (X) is the average queue length of vehicles at the intersection; n is the number of intersections; ρ(X) is the vehicle entry and exit ratio of the area; q 预测 represents the traffic flow predicted by the LSTM model; q 实际 Represents the actual traffic flow; ∈·∑ n (λ n -λ avg ) 2 is the regularization term, λ n Indicates the green-to-signal ratio, the ratio of green light time to the total signal cycle time; λ avg Represents the average green-to-signal ratio.

2. The intelligent traffic control method based on real-time traffic flow detection according to claim 1, characterized in that: Correct the vehicle transmission information data. The specific process is as follows: Real-time collection of terminal connection data of roadside 5G base stations; Estimate regional vehicle density based on terminal connection data from roadside 5G base stations, and supplement vehicle transmission data; Based on road network monitoring data, AI algorithms are used to calculate the number of vehicles on the road; Based on the regional vehicle density and the number of vehicles on the road, the vehicle transmission information data is corrected through the spatiotemporal alignment method to obtain the corrected vehicle transmission information data.

3. The intelligent traffic control method based on real-time traffic flow detection according to claim 1, characterized in that: Based on the pre-processed vehicle transmission information, a prediction model is constructed through a machine learning algorithm to predict traffic flow. The prediction model is an LSTM model, specifically: Get historical traffic data; Input the historical traffic data into the LSTM model, perform model fitting training, and obtain a trained LSTM model; The preprocessed vehicle transmission information is input into the trained LSTM model to predict the traffic flow at the next moment.

4. The intelligent traffic control method based on real-time traffic flow detection according to claim 1, characterized in that: A graph theory model is used to describe intersection relationships, analyze network characteristics, and create a self-organizing traffic network. Specifically, based on the graph theory model, the intersection relationships are described, and vehicles and roadside facilities are used as network nodes, equipped with communication and computing equipment, and C-V2X and Wi-Fi Direct technologies are used to dynamically connect them and establish a self-organizing network topology.

5. The intelligent traffic control method based on real-time traffic flow detection according to claim 1, characterized in that: The calculation formula for the average vehicle queue length is: Among them, l q is the average queue length of vehicles; n is the number of intersections; represents the remaining queue length at intersection i; represents the current queue length at intersection i; a and b are weight coefficients, representing the contribution ratio of the remaining queue length and the newly added queue length, respectively.

6. The intelligent traffic control method based on real-time traffic flow detection according to claim 1, characterized in that: The optimal timing strategy is obtained by using the chaotic particle swarm algorithm, which is as follows: Based on a self-organizing traffic network, with the optimization objectives of minimizing the average queue length, the average waiting time, and the area entry-exit ratio, a chaotic particle swarm optimization algorithm is used to obtain the optimal solution of the traffic light timing calculation model. The optimal solution of the traffic light timing calculation model is converted into signal light control instructions to obtain the optimal timing strategy.

7. An intelligent traffic control system based on real-time traffic flow detection, characterized in that: include: The data acquisition module is configured to: obtain vehicle transmission information data through the intelligent connected vehicle OBU and perform correction; The edge computing module is configured to: perform edge computing analysis on the corrected vehicle transmission information data and perform preprocessing; The core computing module is configured to: build a prediction model based on pre-processed vehicle information through machine learning algorithms to predict traffic flow; Use graph theory models to describe intersection relationships, analyze network characteristics, and create self-organizing transportation networks; Based on the predicted traffic flow, three indicators are defined: average queue length, average waiting time, and area entry and exit ratio; Based on the self-organizing traffic network and three indicators, a traffic light timing calculation model is constructed; Based on the traffic light timing calculation model, the chaotic particle swarm algorithm is used to calculate the optimal timing strategy and coordinate the control of multiple intersections in the area. Among them, based on the self-organizing traffic network and three indicators, a traffic light timing calculation model is constructed, and the formula is: Among them, α, β, γ and δ are weight coefficients. represents the average delay time at the intersection; l q (X) is the average queue length of vehicles at the intersection; n is the number of intersections; ρ(X) is the vehicle entry and exit ratio of the area; q 预测 represents the traffic flow predicted by the LSTM model; q 实际 Represents the actual traffic flow; ∈·∑ n (λ n -λ avg ) 2 is the regularization term, λ n Indicates the green-to-signal ratio, the ratio of green light time to the total signal cycle time; λ avg Represents the average green-to-signal ratio.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are performed.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images