Intelligent traffic control method and system, roadside edge computing device and storage medium

Abstract

The application relates to the technical field of intelligent traffic, and discloses an intelligent traffic control method and system, a roadside edge computing device and a storage medium, wherein the method comprises the following steps: collecting traffic state information acquired by video monitoring equipment, radar equipment and manual input equipment arranged at a traffic intersection and a surrounding area of the traffic intersection, wherein the traffic state information comprises vehicle queue length, average speed and waiting time; performing normalization processing on the traffic state information, calculating entropy values of various data sources, calculating and generating weights of the various data sources based on the entropy values; performing weighted fusion on the traffic state information based on the weights, obtaining a fused traffic state estimation vector, and constructing a Bayesian belief distribution based on the vector to represent a belief value of the traffic state. Through combination of the Bayesian belief distribution and a game model, real-time intelligent optimization scheduling of traffic signals is realized, traffic flow management is effectively improved, and signal waste is reduced.

Description

Technical Field

[0001] With the acceleration of urbanization, traffic management has become a key issue in urban development; especially during peak hours, the optimized scheduling of traffic signals directly affects road efficiency and traffic safety. However, traditional traffic signal control systems often fail to effectively cope with fluctuations in traffic flow, leading to road congestion or unnecessary waiting, wasting a significant amount of time and resources. Therefore, how to intelligently regulate traffic signals to maximize traffic flow has become a pressing technical challenge.

[0002] In existing technologies, traffic signal control largely relies on preset time periods and traffic flow statistics, switching signals according to a fixed cycle. This method is simple and easy to implement, ensuring basic traffic control, and is particularly suitable for areas with relatively stable traffic flow. For example, some intelligent traffic signal systems can adjust the signal cycle based on traffic flow data, avoiding the inefficiency caused by using the same signal cycle throughout the day. Furthermore, some existing technologies can adjust the signal duration for different directions within an intersection by setting priority strategies, thereby achieving a certain degree of traffic optimization.

[0003] However, existing technologies still have some shortcomings. First, traditional signal control schemes are often based on fixed models or flow counting, which cannot adapt to complex traffic flow fluctuations. Especially during peak hours or when unexpected events occur, the system's response speed is slow, failing to make real-time adjustments and exacerbating traffic congestion. Second, most existing systems do not consider the mutual constraints and coordinated control between different traffic directions, meaning that the priority of one direction may affect the capacity of other directions, causing unnecessary traffic conflicts and delays. Finally, existing signal control technologies largely rely on manual settings or historical data extrapolation, lacking the ability to make real-time intelligent decisions and respond promptly to sudden traffic situations. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an intelligent traffic control method, system, roadside edge computing device, and storage medium, which solves the problem that traffic signal control in existing technologies cannot adapt to dynamic traffic flow changes in real time.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an intelligent traffic control method, comprising the following steps:

[0006] Traffic status information is collected from video surveillance equipment, radar equipment, and manual input devices located at and around traffic intersections. The traffic status information includes vehicle queue length, average speed, and waiting time.

[0007] Traffic status information is normalized, and the entropy value of each data source is calculated. The weight of each data source is then calculated and generated based on the entropy value.

[0008] Based on the weights, traffic state information is weighted and fused to obtain a fused traffic state estimation vector, and a Bayesian belief distribution is constructed based on this vector to represent the belief value of traffic state.

[0009] Based on the Bayesian belief distribution, a game model is constructed that includes a set of strategies for left-turning and straight-going directions, and the optimal strategies of the game participants are inferred.

[0010] Based on the optimal strategy, a signal phase switching control command is generated and sent to the signal control device to execute the corresponding phase switching.

[0011] Preferably, the step of acquiring data from video surveillance equipment located at traffic intersections and their surrounding areas includes:

[0012] Acquire real-time video streams from traffic intersections and analyze vehicle queuing information, vehicle types, and flow status in the images;

[0013] Traffic flow speed is measured using radar equipment, and the speed measurement data is processed synchronously with video data;

[0014] Real-time traffic control instructions and emergency information can be input via manual input devices.

[0015] Preferably, the step of normalizing the traffic status information includes:

[0016] The collected vehicle queue length, average vehicle speed, and waiting time were normalized to a uniform numerical range.

[0017] The traffic state information is processed using a minimum-to-maximum standardization method, with the following formula:

[0018]

[0019] Where x represents the original data; x ′ The data is normalized; min(x) and max(x) are the minimum and maximum values ​​of this data dimension, respectively.

[0020] The normalized data is then normalized using standard deviation.

[0021] Preferably, the step of performing standard deviation normalization includes:

[0022] For the normalized traffic state information, the standard deviation of each data dimension is calculated to quantify the dispersion of its data distribution. Based on the calculated standard deviation, each normalized data value is adjusted to meet the requirements of standard deviation normalization. During the standard deviation normalization process, an appropriate scaling factor is used according to the actual fluctuation of the data to balance the fluctuation range of each traffic state data.

[0023] Preferably, the step of weighted fusion of traffic state information based on the weights includes:

[0024] Based on the entropy weights of each data source, a weighted average of the traffic state information is calculated to obtain the fused traffic state estimation vector, as shown in the formula:

[0025]

[0026] Where, x i Normalized traffic status data; w i The weights are calculated from the entropy value; f(x) is the weighted and fused traffic state estimation vector; n is the total number of traffic state indicators;

[0027] During the integration process, different weights are assigned based on the different sources of traffic status information.

[0028] Preferably, the step of constructing a game model that includes a set of strategies for left-turning and straight-going directions includes:

[0029] In the game theory model, the left-turning direction and the straight-going direction are the participants in the game, and the strategy set includes two strategies: "requesting a phase switch" and "maintaining the current phase".

[0030] The payoff function of the game model considers the queue length and efficiency of each direction of traffic; the waiting time of each direction of traffic; and the potential conflict cost when both parties request a phase switch at the same time.

[0031] Preferably, the step of generating signal phase switching control commands based on the optimal strategy includes:

[0032] The optimal strategy derived from the game theory model is mapped to traffic light control instructions, specifically by determining the time window for phase switching based on the optimal strategy.

[0033] The optimal strategy is to "request phase switching", which generates a control command to change the current traffic light phase.

[0034] The optimal strategy is to "maintain the current phase", which means that the current state of the traffic light will remain unchanged until the next decision cycle.

[0035] It also provides an intelligent traffic control system, including:

[0036] The information collection and normalization module is used to receive traffic status information from video surveillance equipment, radar equipment, and manual input devices in and around traffic intersections, and to normalize the information.

[0037] The weight determination and fusion module calculates the entropy value of each data source based on the normalized traffic state information, and performs a weighted fusion operation based on the weights to generate a fused traffic state estimation vector that represents the overall traffic state.

[0038] The traffic state inference module constructs a Bayesian belief distribution based on the fused traffic state estimation vector to represent the uncertainty and belief level of the current traffic state;

[0039] The game reasoning and instruction generation module is used to construct a game model containing a set of strategies for left turn and straight-through directions based on belief distribution, and to reason about the optimal strategies of the game participants, thereby generating corresponding signal phase switching control instructions.

[0040] The signal control execution module is used to receive the control command and drive the signal control device to perform a phase switching operation corresponding to the optimal strategy, thereby realizing dynamic adjustment of signal timing.

[0041] A roadside edge computing device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it is able to implement an intelligent traffic control method.

[0042] A storage medium is also provided, which stores a computer program that, when executed by a processor, enables an intelligent traffic control method.

[0043] This invention provides an intelligent traffic control method, system, roadside edge computing device, and storage medium. It possesses the following features:

[0044] Beneficial effects:

[0045] 1. This invention adopts a technical solution that combines Bayesian belief distribution and game theory model, which achieves the ability to accurately infer traffic flow and priority in complex traffic environments. Compared with the traditional timing control method in the prior art, this invention can adjust the signal timing in real time according to the traffic conditions, which solves the problem that static signal control cannot adapt to dynamic traffic flow changes.

[0046] 2. This invention improves the adaptability and real-time performance of traffic signal control by weighted fusion of multiple traffic data sources and generation of signal control commands. Compared to traditional signal control technology based on simple flow counting, this invention solves the problem of slow response in traditional systems when traffic flow fluctuates greatly, and can more flexibly cope with traffic peaks or emergencies.

[0047] 3. The technical solution of this invention, which generates signal phase switching control commands based on an optimal strategy, significantly improves the accuracy of signal control and scheduling efficiency. Unlike existing technologies that rely on manual intervention or fixed-cycle scheduling based on historical data, this invention optimizes phase switching decisions through intelligent reasoning, avoiding signal waste and traffic congestion.

[0048] 4. This invention introduces game theory to optimize traffic signal control strategies, achieving the effect of reducing traffic conflicts and improving traffic efficiency. Compared with the existing technology, which lacks consideration of the mutual influence between traffic strategies in different directions, this invention dynamically adjusts the signal cycle and priority through a game model, solving the problem of insufficient coordination of multi-directional traffic and effectively improving the traffic capacity of intersections. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the method flow of the present invention;

[0050] Figure 2 This is a system architecture diagram of the present invention;

[0051] Figure 3 This is a schematic diagram of the electronic device of the present invention. Detailed Implementation

[0052] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] Please see the appendix Figure 1 This invention provides an intelligent traffic control method, comprising the following steps:

[0054] S1. Collect traffic status information from video surveillance equipment, radar equipment, and manual input devices located at traffic intersections and their surrounding areas. The traffic status information includes vehicle queue length, average speed, and waiting time.

[0055] The acquisition and processing of traffic state information is a fundamental step supporting the generation of subsequent control strategies and the issuance of control commands. The accurate perception of traffic conditions directly impacts the efficiency of model inference and the targeted nature of control strategies. To achieve refined and real-time control of intersection traffic conditions, this invention introduces various types of sensing devices located at and around traffic intersections, including video surveillance equipment, radar equipment, and manual input terminals. These devices operate collaboratively to acquire key traffic state information, including vehicle queue length, average speed, and waiting time.

[0056] In this embodiment, the video surveillance equipment is mainly used to acquire real-time traffic flow images. With the help of image recognition algorithms, vehicle positions can be calibrated in real time, thereby identifying the number of vehicles in queue and calculating the queue length. Generally, by detecting the relative position of the vehicle's bounding box to the road area, vehicle density can be estimated, and thus the queue length can be obtained. Queue length L q It can be calculated using the following formula:

[0057] L q =N v ·l v +(N v -1)·d s ;

[0058] Among them, L q N represents the length of the vehicle queue. v Indicates the current number of vehicles in the queue; l v This represents the average vehicle length, which can be set to a fixed value or dynamically estimated based on the recognition results; d s This indicates the average distance between vehicles (unit: meters).

[0059] Specifically, the number of vehicles in the queue, N v Real-time statistics can be obtained through target detection models (such as the YOLO series), and combined with geographically labeled areas, it can be used to determine whether vehicles are in a queue, thus avoiding misjudging passing vehicles.

[0060] In one possible implementation, the radar equipment is installed above or beside the intersection. Its operation is independent of visible light, making it suitable for all-weather conditions, especially advantageous in low-light or inclement weather. The radar measures the instantaneous speed of vehicles using the Doppler effect and calculates the average speed of vehicles within a section. This average speed V avg It can be given by the following formula:

[0061]

[0062] Among them, V avg N represents the average velocity during the sampling period; r Indicates the number of vehicles detected; v iLet represent the instantaneous speed of the i-th vehicle.

[0063] Alternatively, radar data can be fused with video image recognition results to improve the accuracy of speed calculations. In areas where radar beams overlap, multi-source data verification can be performed by matching vehicle IDs with path trajectories to eliminate abnormal samples.

[0064] In this embodiment, the manual input device is mainly for facilitating the collection of signals from sudden events or localized interventions, such as on-site control information from traffic police, dispatch instructions, or temporary construction records. By configuring a mobile terminal or portable input panel, on-duty personnel can directly input abnormal status markers. For example, inputting "severe congestion in the left-turn lane" can trigger the model to adjust the left-turn phase duration. This input information enters the model processing flow in a structured format to correct or supplement the automatic sensing results.

[0065] In some embodiments, to avoid the impact of data errors on the strategy, this invention introduces a time-weighted filtering mechanism to correct the estimation of vehicle waiting time. Waiting time T w The calculation method is as follows:

[0066]

[0067] Among them, T w t represents the average waiting time (in seconds); j w represents the actual waiting time of the j-th vehicle; j This represents the time-weighted factor corresponding to the vehicle; M represents the number of vehicles in the statistical sample.

[0068] In one practical application, if the average waiting time on the main road is detected to exceed a threshold, the system can trigger priority passage logic and adjust the traffic light timing. The threshold setting can be adaptively learned based on the historical average fluctuation range.

[0069] Furthermore, to enhance system stability and redundancy, this invention supports a sensor data redundancy sampling mechanism. When any device fails or its accuracy is abnormal, the system will automatically activate a backup data source to compensate, ensuring overall data continuity.

[0070] Furthermore, before entering the scheduling model, multi-source traffic state data undergoes unified feature standardization processing, such as min-max normalization and Z-score transformation, to adapt to the data format requirements of the language model's input interface. This step ensures the consistency and comparability of data acquired by the subsequent semantic inference module, improving the accuracy and stability of the output strategy.

[0071] S2. Normalize the traffic status information and calculate the entropy value of each data source. Calculate and generate the weight of each data source based on the entropy value.

[0072] While acquiring traffic status information constitutes the foundation of data input, ensuring that this multi-source, heterogeneous data reliably supports subsequent reasoning and decision-making requires standardized processing and weighted optimization. Therefore, after completing the initial perception and extraction of traffic status information collected by video surveillance equipment, radar equipment, and manual input terminals, this invention further incorporates a normalization processing and information entropy weight calculation module. This module performs numerical normalization on data from different data sources and measures the data distribution characteristics using entropy theory, generating data source weights suitable for fusion calculations to ensure the reasonable distribution, significant features, and decision-making reference value of the input data.

[0073] In this embodiment, to ensure the comparability of traffic state data (e.g., vehicle queue length, average speed, waiting time) from different data sources, the original observation data are first normalized. Generally, the min-max normalization method is used. The original data x is standardized using the following formula:

[0074]

[0075] Where x represents the original data; x ′ The data is normalized; min(x) and max(x) are the minimum and maximum values ​​of this data dimension, respectively.

[0076] Alternatively, the above normalization process can be applied to the same traffic indicators from different data sources, such as average speed data collected by radar and video equipment, and input into the next entropy analysis module after being scaled uniformly.

[0077] In one possible implementation, the system calculates the information entropy of each normalized indicator data to measure the information distribution of that indicator in the overall sample, where p represents its proportion. ij The calculation is as follows:

[0078]

[0079] Where, p ij represents the normalized percentage of sample i on the j-th indicator; m represents the total number of samples; x is the original data.

[0080] In some embodiments, the system calculates the discriminative power of each data indicator based on the entropy value and generates a corresponding weight. A smaller entropy value indicates a more concentrated distribution of indicators, greater diversity, and higher information content, thus warranting a higher weight. The final weight calculation formula is as follows:

[0081]

[0082] Among them, w jThis indicates the time-weighted factor corresponding to the vehicle;

[0083] Generally, the system stores the calculated weights of each indicator and uses them as weighting factors in the feature aggregation process during the multi-source data fusion stage. Through the weight control mechanism, the reference status of different perception channels in complex traffic scenarios can be effectively balanced, the influence of noise redundancy can be suppressed, and the perception robustness of the strategy model can be improved.

[0084] As an extension of the processing, this invention supports a dynamic entropy update mechanism. Within a set period (e.g., every 5 minutes), the system automatically recalculates the entropy values ​​and weights of each indicator for the current time period, thereby adapting to the time-varying characteristics of traffic conditions. Furthermore, for abnormal samples (such as data mutations caused by localized congestion due to sudden events), the system can introduce a sliding window averaging process to smooth the data at individual points, preventing mutations from affecting overall entropy fluctuations.

[0085] S3. Based on the weights, the traffic state information is weighted and fused to obtain the fused traffic state estimation vector, and a Bayesian belief distribution is constructed based on the vector to represent the belief value of the traffic state.

[0086] After normalizing the multi-source traffic state data and determining the weights of each indicator based on information entropy theory, a unified and structured state estimation vector needs to be constructed to support the inference and execution of the subsequent intelligent traffic control model. The core of this stage lies in fusing the normalized and weighted data into a unified vector form representing the current traffic state, and establishing a Bayesian belief distribution on this basis to quantitatively express the system's level of trust in the current traffic state. This invention integrates state data obtained from multiple sensing sources using a weighted fusion method, and constructs a belief expression model with the help of Bayesian theory to achieve robust modeling of traffic states under uncertain conditions.

[0087] In this embodiment, after completing the normalization processing and entropy weight calculation of each traffic state index, the weight w corresponding to each index is used as the basis for the calculation. i The normalized data is weighted and fused to obtain a fused traffic state estimation vector f(x). This estimation vector represents the integration of information from multiple data sources regarding key states (such as queue length, average speed, and waiting time) in the current traffic scenario. The calculation formula is as follows:

[0088]

[0089] Where, x i Normalized traffic status data; w i The weights are calculated from the entropy value; f(x) is the weighted and fused traffic state estimation vector; n is the total number of traffic state indicators.

[0090] In one possible implementation, the fusion vector f(x) is further used to construct a Bayesian belief distribution. This belief distribution expresses the current system's level of confidence in different traffic state classifications (such as "smooth traffic," "mild congestion," and "severe congestion"). The belief distribution is built based on Bayesian inference, using the fusion vector as observational evidence and combining it with prior knowledge to obtain posterior belief values.

[0091] Generally, the fused state vector is taken as f(x), which is given a state θ. k The likelihood function under the following expression is:

[0092]

[0093] Wherein, P(f(x)|θ k ) is in state θ k The observation probability of the fused state vector; The mean is μ k The covariance is Σ k It follows a multivariate normal distribution.

[0094] As an alternative, this invention supports a multi-round fusion mechanism. By accumulating the f(x) vectors from multiple time points, a time-series belief propagation model can be constructed, further improving the robustness of traffic state prediction. Introducing a Hidden Markov Model (HMM) for Bayesian updates in dynamic scenarios helps model state transition probabilities and compensates for inference biases caused by short-term anomalies or data interruptions.

[0095] S4. Based on the Bayesian belief distribution, construct a game model that includes the strategy set of left-turning and straight-going directions, and deduce the optimal strategy of the game participants.

[0096] After obtaining the fused traffic state estimation vector and constructing a Bayesian belief distribution based on this vector, the system still needs to rationally select possible traffic control strategies under the current traffic conditions. To this end, this invention further proposes a strategy game model to deduce the optimal control behavior of each player in the game based on belief-driven reasoning. This model uses the traffic state characterized by the belief distribution as the decision-making environment, sets a set of strategies including left-turn and straight-ahead directions, constructs a game framework for traffic participation behavior, and realizes strategy evolution reasoning under competitive or cooperative game mechanisms.

[0097] In this embodiment, the game theory model focuses on the key traffic directions at the intersection, mainly including two typical traffic modes: left turn and straight ahead. Each direction is abstracted as a game participant, whose strategy set includes control options such as "allow passage," "delay," and "restricted passage." The goal of the game is to achieve a balance between traffic efficiency and conflict minimization under different traffic conditions.

[0098] Generally, the system first determines the category and confidence level of the current traffic state based on the aforementioned Bayesian belief distribution, and then infers the traffic pressure in each direction of travel. A higher belief value indicates greater traffic density in the corresponding direction and a higher urgency for travel strategies. Alternatively, this pressure information can be used to construct the payoff function for each participant in each direction.

[0099] Specifically, there is a potential conflict between the strategies for left turns and going straight. For example, when both directions simultaneously receive a "go" control signal, there will be a high probability of traffic conflict at the intersection, thus affecting traffic safety. Therefore, in the game theory model, each strategy combination corresponds to a payoff matrix, the elements of which are constructed from the following factors: current traffic state belief value;

[0100] The estimated number of vehicles passing through in each direction;

[0101] Potential delay costs and conflict penalties under various strategy combinations;

[0102] Cost of switching control strategies.

[0103] It is affected by both traffic delays and the clearance status of adjacent straight-ahead directions. This dependency is described by constructing a conditional payoff model, which is dynamically updated based on Bayesian belief values.

[0104] In some embodiments, the system employs Nash equilibrium analysis to solve the constructed strategy game model, deduce the optimal strategy combination for each participant under the current traffic conditions. If a strategy combination remains stable without any unilateral deviation motivation among the participants, it is considered a Nash equilibrium solution. This optimal solution will serve as the input strategy for the next cycle of signal control, used to update the traffic signal timing state.

[0105] As a specific operational approach, the system can incorporate the framework of incomplete information game theory during the strategy reasoning process, combining it with historical strategy execution records to construct a hybrid strategy model. In this model, each direction can choose to execute a certain strategy probabilistically based on the uncertainty of the opponent's strategy. For example, when there is heavy traffic in the left-turn direction and a low belief value in the straight-ahead direction, the system will execute the "left-turn priority" strategy with a higher probability, improving overall traffic efficiency.

[0106] Furthermore, to enhance the model's adaptability and real-time performance, this invention introduces a dynamic game evolution mechanism. During continuous traffic cycles, the system adjusts the weighting coefficients of the payoff function based on feedback from the effects of each round's strategy implementation (such as changes in queue length and the number of vehicles passing through), thereby achieving dynamic iteration and self-optimization of the strategy game framework.

[0107] S5. Generate a signal phase switching control command based on the optimal strategy, and send the control command to the signal control device to execute the corresponding phase switching;

[0108] After selecting the optimal strategy based on Bayesian belief distribution reasoning, the next key task is to translate the optimal strategy into signal control instructions and transmit these instructions to the signal control equipment for real-time execution of corresponding signal phase switching. This process needs to be closely integrated with the aforementioned game theory reasoning module to ensure that each decision effectively reflects the traffic flow status, optimizes traffic signal timing in a timely manner, and avoids traffic congestion or signal waste.

[0109] In this embodiment, the system first converts the output information of the optimal strategy into signal control instructions. These instructions specifically include specifying the switching order and duration of each traffic signal phase within the current cycle, as well as priority settings. For example, if the belief value for the straight-ahead direction is high during a certain period, the system will generate a "straight-ahead priority" strategy, instructing the signal control equipment to activate the straight-ahead phase and extend its duration. Simultaneously, the signal duration for other phases (such as left turns or right turns) will be shortened accordingly to avoid unnecessary traffic congestion.

[0110] Specifically, the system extracts the corresponding signal control parameters from the optimal strategy output obtained based on the game theory model, including the signal duration in each direction (e.g., T). left T represents the time of the left turn phase. straight This indicates the time of the straight-ahead phase and the timing of phase switching. Based on these parameters, the system generates control commands through a signal control algorithm, which involves the traffic light cycle P. cycle The settings are as follows:

[0111] P cycle =T left +T straight +T turnright ;

[0112] Among them, P cycle T is the total duration of the signal period; left T is the duration of the left turn phase; straight T is the duration of the straight-line phase; turnright This represents the duration of the right-turn phase.

[0113] Alternatively, the system can dynamically adjust the duration of each phase based on traffic flow, belief value, and conflict weights to ensure that traffic flow is optimized to the maximum extent.

[0114] Next, the generated signal control commands are sent to the signal control equipment via the communication interface. This process requires consideration of the data transmission protocol between the system and the signal control equipment. Generally, the system uses standardized traffic signal control protocols (such as SCOOT, ITS, etc.) to ensure that signal commands are transmitted to the signal control equipment in a timely and accurate manner. In actual operation, the signal control equipment parses the received commands and performs corresponding signal phase switching and duration adjustments based on the command content.

[0115] Specifically, after parsing the instructions, the signal control equipment will control the status of the traffic lights in each traffic direction according to the time allocation specified in the instructions. The control process of the equipment typically includes the following steps:

[0116] Initiate the corresponding signal phase and maintain it for the specified time;

[0117] When a phase ends, the signal switches to the next phase.

[0118] At the end of the cycle, the next signal cycle begins.

[0119] In some embodiments, to enhance the system's responsiveness and stability, the signal control equipment can also adjust the command execution process based on a feedback mechanism. For example, when traffic flow changes significantly, the equipment can dynamically adjust the signal cycle or phase switching sequence by detecting traffic conditions in real time, thereby further optimizing traffic efficiency.

[0120] As an extension, the system can also combine real-time traffic data and historical traffic patterns to further optimize the generated signal control commands. By introducing machine learning algorithms or predictive models, the system can infer the most likely changes in traffic flow based on traffic patterns in different time periods, and adjust the signal phases in advance accordingly to improve traffic flow smoothness and the ability to respond to emergencies.

[0121] The intelligent traffic control system described below can be referenced in correspondence with the intelligent traffic control method described above.

[0122] Please see the appendix Figure 2 The present invention also provides an intelligent traffic control system, comprising:

[0123] The information collection and normalization module is used to receive traffic status information from video surveillance equipment, radar equipment, and manual input devices in and around traffic intersections, and to normalize the information.

[0124] The weight determination and fusion module calculates the entropy value of each data source based on the normalized traffic state information, and performs a weighted fusion operation based on the weights to generate a fused traffic state estimation vector that represents the overall traffic state.

[0125] The traffic state inference module constructs a Bayesian belief distribution based on the fused traffic state estimation vector to represent the uncertainty and belief level of the current traffic state;

[0126] The game reasoning and instruction generation module is used to construct a game model containing a set of strategies for left turn and straight-through directions based on belief distribution, and to reason about the optimal strategies of the game participants, thereby generating corresponding signal phase switching control instructions.

[0127] The signal control execution module is used to receive control commands and drive the signal control equipment to perform phase switching operations corresponding to the optimal strategy, thereby realizing dynamic adjustment of signal timing.

[0128] The system in this embodiment can be used to execute the above method embodiments, and its principle and technical effect are similar, so they will not be described again here.

[0129] The roadside edge computing device described below can be referenced in correspondence with the intelligent traffic control method described above.

[0130] Please see the appendix Figure 3 The present invention also provides a roadside edge computing 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 computer program, it is able to perform the above method.

[0131] The present invention also provides a storage medium storing a computer program, which is executed by a processor to perform the method described above.

[0132] The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0133] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method of intelligent traffic control, characterized by The method comprises the following steps: Collecting traffic state information obtained by video monitoring devices, radar devices and manual input devices arranged at traffic intersections and their surrounding areas, wherein the traffic state information comprises vehicle queue length, average speed and waiting time; Normalizing the traffic state information and calculating the entropy value of each data source, and calculating and generating the weight of each data source based on the entropy value; Weighted fusion of the traffic state information based on the weight, obtaining a fused traffic state estimation vector, and constructing a Bayesian belief distribution based on the vector to represent the belief value of the traffic state; According to the Bayesian belief distribution, a game model including a strategy set of left-turn and straight-through directions is constructed, and the optimal strategy of the game participants is inferred; Generating a signal phase switching control instruction based on the optimal strategy, and sending the control instruction to the signal control device to execute the corresponding phase switching.

2. The intelligent traffic control method according to claim 1, characterized by, The step of collecting video monitoring devices arranged at traffic intersections and their surrounding areas comprises: Obtaining real-time video streams of traffic intersections, analyzing vehicle queue information, vehicle types and flow states in the images; Speed measurement of traffic flow by radar devices, and synchronous processing of speed measurement data and video data; Input real-time traffic control instructions and emergency information through manual input devices.

3. The intelligent traffic control method according to claim 1, wherein, The step of normalizing the traffic state information comprises: Normalizing the collected vehicle queue length, average speed and waiting time to a unified numerical range; Using the minimum-to-maximum standardization method to process the traffic state information, the formula is: where x is the original data; x ′ is the normalized data; min(x) and max(x) are the minimum and maximum values of the data dimension, respectively; Standard deviation normalization processing of the normalized data.

4. The intelligent traffic control method according to claim 3, wherein, The step of standard deviation normalization processing comprises: Calculating the standard deviation of each data dimension of the normalized traffic state information to quantify the dispersion degree of its data distribution; adjusting each normalized data value according to the calculated standard deviation to meet the requirements of standard deviation normalization; In the standard deviation normalization processing process, a proper scaling factor is used according to the actual fluctuation of the data.

5. The intelligent traffic control method of claim 1, wherein, The step of weighted fusion of the traffic state information based on the weight comprises: According to the entropy weight of each data source, the traffic state information is weighted and averaged to obtain a fused traffic state estimation vector, and the formula is: where x i is the normalized traffic state data; w i is the weight calculated by the entropy value; f(x) is the traffic state estimation vector after weighted fusion; and n is the total number of traffic state indicators. In the fusion process, different weights are assigned according to different sources of traffic state information.

6. The intelligent traffic control method of claim 1, wherein, The step of constructing a game model including a strategy set of left-turn and straight-through directions comprises: In the game model, the left-turn and straight-through directions are the participants of the game, and the strategy set includes two strategies: "request phase switching" and "maintain current phase"; The payoff function of the game model considers the queue length and traffic efficiency of each direction, the waiting time of each direction, and the potential conflict cost when both sides request phase switching at the same time.

7. The intelligent traffic control method of claim 1, wherein, The step of generating a signal phase switching control instruction based on the optimal strategy comprises: Mapping the optimal strategy inferred by the game model into a signal light control instruction, specifically determining the time window of phase switching according to the optimal strategy; If the optimal strategy is "request phase switch", a control instruction is generated to change the current signal phase; If the optimal strategy is "maintain current phase", the current signal state is maintained until the next decision period.

8. The intelligent traffic control system for use in the intelligent traffic control method according to any one of claims 1 to 7, characterized in that, The method comprises: An information collection and normalization module configured to receive traffic state information from video monitoring devices, radar devices and manual input devices at the intersection and its surrounding area, and normalize the information; A weight determination and fusion module configured to calculate the entropy value of each data source based on the normalized traffic state information, and perform a weighted fusion operation based on the weights to generate a fusion traffic state estimation vector representing the overall traffic state; A traffic state inference module configured to construct a Bayesian belief distribution based on the fusion traffic state estimation vector to represent the uncertainty and belief level of the current traffic state; A game reasoning and instruction generation module configured to construct a game model including a strategy set of left turn and straight traffic directions based on the belief distribution, reason the optimal strategy of the game participants, and generate a corresponding signal phase switch control instruction; A signal control execution module configured to receive the control instruction and drive the signal control device to perform a phase switch operation corresponding to the optimal strategy to realize dynamic adjustment of signal timing.

9. A roadside edge computing device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the intelligent traffic control method of any one of claims 1-7.

10. A storage medium, characterized by The computer program is stored in the memory and executed by the processor to implement the intelligent traffic control method of any one of claims 1-7.

Images