Intelligent road traffic signal lamp dynamic regulation and control system and method based on artificial intelligence

Abstract

The invention discloses an intelligent road traffic signal lamp dynamic regulation and control system and method based on artificial intelligence, and relates to the related field of signal lamp regulation and control, and the system comprises the steps: constructing a traffic state feature vector based on traffic flow data; the traffic state feature vector is analyzed in real time through a deep learning model, the traffic flow change trend in a future time window is predicted, a prediction result is generated, a dynamic signal lamp regulation and control strategy is generated and issued to a signal lamp control unit, the signal lamp state is adjusted in real time through the signal lamp control unit, and a regulation and control result is generated; and storing the regulation and control result and the traffic flow data to a traffic management database to perform data mining optimization, and generating a signal lamp control optimization suggestion to perform dynamic regulation and control on the intelligent traffic signal lamp. The technical problem that a traditional signal lamp is difficult to flexibly adjust according to the real-time traffic flow is solved, and the technical effects of dynamically adjusting the signal lamp time sequence, achieving intelligent timing and reducing the congestion phenomenon and the traffic accident rate are achieved.

Description

Technical Field

[0001] This application relates to the field of signal lamp regulation, and particularly to an intelligent road traffic signal lamp dynamic regulation system and method based on artificial intelligence. Background Art

[0002] The control of traffic signal lamps mainly relies on preset fixed timing schemes, which are often formulated based on historical traffic data or empirical judgments and are difficult to accurately reflect the changes in real-time traffic conditions. With the acceleration of urbanization and the continuous increase in traffic flow, the limitations of this static control method have become increasingly prominent, resulting in more serious problems such as traffic congestion, low traffic efficiency, and increased environmental pollution, leading to the technical problem that traditional signal lamps are difficult to flexibly adjust according to real-time traffic flow. Summary of the Invention

[0003] This application provides an intelligent road traffic signal lamp dynamic regulation system and method based on artificial intelligence, solves the technical problem that traditional signal lamps are difficult to flexibly adjust according to real-time traffic flow, and achieves the technical effect of dynamically adjusting the signal lamp timing to realize intelligent timing, reducing congestion and the accident rate. This application provides an intelligent road traffic signal lamp dynamic regulation system based on artificial intelligence. The system is applied to an intelligent road traffic signal lamp dynamic regulation method based on artificial intelligence, and includes: a vector construction module for constructing a traffic state feature vector based on traffic flow data; a real-time analysis module for performing real-time analysis on the traffic state feature vector through a deep learning model, predicting the traffic flow change trend within a future time window, and generating a prediction result; a regulation analysis module for generating a dynamic signal lamp regulation strategy according to the prediction result; a real-time adjustment module for sending the dynamic signal lamp regulation strategy to a signal lamp control unit, and adjusting the signal lamp state in real time through the signal lamp control unit to generate a regulation result; a mining and optimization module for storing the regulation result and the traffic flow data in a traffic management database for data mining and optimization, generating a signal lamp control optimization suggestion, and dynamically regulating the intelligent traffic signal lamp.

[0004] This application also provides an intelligent road traffic signal lamp dynamic regulation method based on artificial intelligence, including: constructing a traffic state feature vector based on traffic flow data; performing real-time analysis on the traffic state feature vector through a deep learning model, predicting the traffic flow change trend within a future time window, and generating a prediction result; generating a dynamic signal lamp regulation strategy according to the prediction result; sending the dynamic signal lamp regulation strategy to a signal lamp control unit, and adjusting the signal lamp state in real time through the signal lamp control unit to generate a regulation result; storing the regulation result and the traffic flow data in a traffic management database for data mining and optimization, generating a signal lamp control optimization suggestion, and dynamically regulating the intelligent traffic signal lamp.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0006] The intelligent road traffic signal dynamic regulation system and method based on artificial intelligence provided in this application relate to the technical field of signal lamp regulation, solve the technical problem that traditional signal lamps are difficult to flexibly adjust according to real-time traffic flow, and achieve the technical effect of dynamically adjusting the signal lamp timing to realize intelligent timing, reducing congestion and the accident incidence rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions of the embodiments of this application, the drawings of the embodiments of this application will be briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the operations in the front or below do not necessarily need to be executed precisely in sequence. On the contrary, according to needs, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0008] Figure 1 It is a schematic structural diagram of the intelligent road traffic signal dynamic regulation system based on artificial intelligence provided for the embodiments of this application;

[0009] Figure 2 It is a schematic flowchart of the intelligent road traffic signal dynamic regulation method based on artificial intelligence provided for the embodiments of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0010] The above description is only an overview of the technical solutions of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the following specifically gives the detailed description of the embodiments of this application.

[0011] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the drawings. The described embodiments should not be regarded as limitations of this application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

[0012] In the following description, "some embodiments" are involved, which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "first / second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application.

[0013] The embodiments of this application provide an intelligent road traffic signal dynamic regulation method based on artificial intelligence. The method is applied to an intelligent road traffic signal dynamic regulation system based on artificial intelligence, as Figure 1 shown. The system includes:

[0014] The intelligent road traffic signal dynamic regulation system based on artificial intelligence according to the embodiments of this application is used to solve the technical problem that traditional traffic lights are difficult to flexibly adjust according to real-time traffic flow, and achieve the technical effect of dynamically adjusting the signal timing to realize intelligent timing, reducing congestion phenomena and the accident incidence rate. The intelligent road traffic signal dynamic regulation system based on artificial intelligence includes: a vector construction module 10, which is used to construct a traffic state feature vector based on traffic flow data. Next, the specific configuration of the vector construction module 10 will be described in detail. As described above, the vector construction module 10 may further include: a first calculation unit, which is used to extract the historical traffic flow data of the target intersection and calculate the average flow rate, vehicle speed standard deviation, and queue length change rate at different time periods; a weight construction unit, which is used to identify abnormal event features and construct event influence weights; a first prediction unit, which is used to predict congestion propagation based on historical congestion patterns and obtain a congestion prediction result; a data fusion unit, which is used to fuse the average flow rate, the vehicle speed standard deviation, the queue length change rate, the event influence weights, and the congestion prediction result to obtain a traffic state feature vector.

[0015] First, historical traffic flow data, including original data such as vehicle flow, speed distribution, and queue length, need to be extracted from the traffic monitoring system, inductive loop, or video detection equipment at the target intersection. For different time periods (such as morning rush hour, off-peak, evening rush hour), the data is segmented and aggregated according to a time window (for example, in units of 15 minutes): the average flow is obtained by calculating the mean value of the total number of vehicles passing through the intersection per unit time within each time period; the standard deviation of speed is calculated based on the dispersion degree of the instantaneous speeds of all vehicles within each time period, reflecting the speed volatility; the queue length change rate is obtained by statistically calculating the increase or decrease amplitude of the number of queuing vehicles within each time window (such as the growth rate of the queue length from the initial value to the peak value) and normalizing it. Abnormal event features can be identified through event detection algorithms (such as computer vision-based traffic accident recognition or weather sensor data): for traffic accidents, parameters such as the location of the event (whether it is in the core area of the intersection), duration (such as more than 30 minutes), and the number of affected lanes (such as occupying more than 2 lanes) are extracted; for severe weather (heavy rain, fog), indicators such as visibility (such as less than 100 meters) and precipitation intensity (such as the precipitation per hour exceeding 20 mm) are quantified. Based on the event type and parameters, the analytic hierarchy process (AHP) is used to construct the event impact weight. For example, the weight of a traffic accident is 0.7 (if it affects more than 50% of the lanes), and the weight of heavy rain is 0.5 (if the visibility < 100 meters), and the weight value is corrected through the expert scoring method. Further, a congestion propagation model is constructed based on historical congestion data: using a graph neural network (GNN) to simulate the congestion diffusion path of the upstream and downstream sections of the intersection, inputting the historical congestion start time, duration, and diffusion range (such as the time when the congestion spreads from the intersection to the adjacent section), and outputting the probability of the area and duration that the congestion may cover within the next 30 minutes (such as the probability of spreading to the downstream 500-meter area is 80%).

[0016] Finally, the above indicators are fused into a traffic state feature vector: the average flow, standard deviation of speed, and queue length change rate are standardized by Z-score; the event impact weight and congestion prediction results (such as the probability of the diffusion range) are normalized by Min-Max. A feature vector containing 5 dimensions is generated through feature concatenation (such as [flow mean = 1.2, speed standard deviation = 0.8, queue change rate = 0.6, event weight = 0.7, congestion probability = 0.85]), and is input into a deep learning model for traffic state classification and generation of control strategies.

[0017] The real-time analysis module 20 is used to perform real-time analysis on the traffic state feature vectors through a deep learning model, predict the traffic flow change trend within a future time window, and generate a prediction result. Next, the specific configuration of the real-time analysis module 20 will be described in detail. As described above, the real-time analysis module 20 may further include: a data training unit for collecting traffic flow data of a target intersection for N consecutive days, constructing a training data set, where the training data set includes the traffic state feature vectors and the corresponding actual traffic flow changes, and N is an integer greater than 2; a feature construction unit for using a convolutional neural network to extract the spatio-temporal features of the traffic flow data and generate high-dimensional features; a data modeling unit for performing time series modeling on the high-dimensional features through a long short-term memory network to obtain the deep learning model.

[0018] Collect traffic flow data of a target intersection for N consecutive days (N > 2). This data should include traffic state feature vectors in multiple dimensions, such as traffic volume, vehicle speed, lane occupancy rate, etc., and the corresponding actual traffic flow changes of these feature vectors. These raw data constitute the training data set, providing a basis for the model to learn the internal laws of traffic flow changes.

[0019] Next, use a convolutional neural network (CNN) to process the training data set. CNN can efficiently extract the spatio-temporal features in the data. By sliding the convolutional kernel in the time and space dimensions, it captures the local patterns and change trends of the traffic flow data, thereby generating a high-dimensional feature representation. The high-dimensional features not only contain the spatial distribution information of the traffic flow but also the dynamic changes in time. After obtaining the high-dimensional features, input the high-dimensional features into a long short-term memory network (LSTM) for time series modeling. The LSTM network captures the long-term dependencies in the time series data and effectively remembers and forgets information according to the gating mechanism, thereby accurately predicting future traffic flow changes. By training the LSTM network, the model learns the mapping relationship from high-dimensional features to actual traffic flow changes, and thus obtains a deep learning model that can accurately predict traffic flow. This not only improves the prediction accuracy of the model but also enhances its adaptability to complex traffic flow changes.

[0020] Furthermore, apply the deep learning model to the analysis of real-time traffic state feature vectors. The feature vectors are collected in real time by sensors deployed at the target intersection and continuously input into the model. The model first uses a convolutional neural network (CNN) to quickly process the input feature vectors and extract high-dimensional features containing spatio-temporal information. Subsequently, these high-dimensional features are fed into a long short-term memory network (LSTM) for time series analysis.

[0021] The LSTM network captures the long-term dependencies hidden in the feature vectors, thus accurately predicting the changing trend of traffic flow within a future time window. This prediction process is online and real-time, capable of quickly responding to changes in traffic conditions. Finally, the model generates prediction results, which include the expected values of traffic flow and their changing trends over a period of time in the future, providing valuable decision-making support for traffic management departments.

[0022] By leveraging the prediction ability of the deep learning model through real-time analysis processes, accurate prediction of traffic flow is achieved, contributing to the smooth operation and efficient management of urban traffic.

[0023] The regulation analysis module 30 is used to generate a dynamic signal light regulation strategy according to the prediction results; below, the specific configuration of the regulation analysis module 30 will be described in detail. As described above, the regulation analysis module 30 may further include: a second calculation unit for extracting the predicted trend of traffic flow changes according to the prediction results, calculating the optimal green light duration for each phase, and obtaining the optimal green light durations for multiple phases; a data detection unit for identifying high-priority phases, determining the phase switching priority, and when an abnormal event is detected, starting an emergency response mode to forcibly extend the green light duration of relevant phases and shorten the durations of other phases through the emergency response mode; a data integration unit for integrating the optimal green light durations for multiple phases, the phase switching priority, and the emergency response mode into the dynamic signal light regulation strategy.

[0024] First, extract the predicted trend of traffic flow changes, which is used to characterize the dynamic changes of vehicle flows in each direction over a period of time in the future. Combine with the traffic flow theory model to calculate the optimal green light duration for each phase (i.e., the time period when vehicle flows in different directions pass). This calculation process aims to maximize the traffic efficiency at the intersection, reduce vehicle waiting time, and alleviate traffic congestion.

[0025] After obtaining the optimal green light durations for multiple phases, further identify the high-priority phases, which usually correspond to the directions with large traffic flow or high traffic pressure. Determine the priority order of phase switching based on the priority information to ensure that the direction with the largest traffic flow can obtain more passing opportunities within a limited time.

[0026] In addition, respond to traffic abnormal events, such as traffic accidents and vehicle breakdowns, through the emergency response mode. When such abnormal events are detected, immediately start the emergency response mode to ensure that emergency vehicles (such as ambulances and fire trucks) can quickly pass through the intersection by forcibly extending the green light duration of relevant phases and shortening the durations of other phases, and at the same time provide necessary passing convenience for affected vehicles.

[0027] Finally, the optimal green light durations, phase switching priorities, and emergency response modes of multiple phases are integrated into a complete set of dynamic signal control strategies. The dynamic signal control strategy can flexibly adjust signal control according to the real-time traffic conditions, effectively improving the traffic capacity and safety of intersections.

[0028] Next, the specific configuration of the regulation analysis module 30 will be described in detail. As described above, the regulation analysis module 30 may further include: a first event generation unit for generating a temporary road closure event according to the road construction plan cycle of the target road; a second event generation unit for generating a traffic accident event according to the accident impact analysis of the target road; a third event generation unit for generating a bad weather event according to the weather visibility analysis of the target road; and a fourth event generation unit for adding the temporary road closure event, the traffic accident event, and the bad weather event to the abnormal event.

[0029] To comprehensively cover abnormal events that may affect traffic flow, first, a multi-dimensional analysis of the target road is carried out. A temporary road closure event is planned and generated according to the road construction plan cycle. The temporary road closure event may include the specific start and end times of construction, the specific section of the road closure, and the estimated traffic impact range, providing an important basis for subsequent traffic management.

[0030] Secondly, an accident impact analysis is carried out on the target road. By analyzing historical accident data and road characteristics, possible traffic accident events are predicted and generated. The traffic accident event details the possible location, time, and potential impact degree of the accident, helping to formulate countermeasures in a timely manner.

[0031] In addition, an in-depth analysis of the weather visibility of the target road is also carried out. By combining meteorological data and road environment, bad weather events are identified and generated. The bad weather event may include weather conditions such as haze, heavy rain, and blizzard that may affect traffic safety, as well as their potential impact on road traffic capacity.

[0032] Finally, these temporary road closure events, traffic accident events, and bad weather events are uniformly added to the abnormal event library. The abnormal event library provides comprehensive abnormal event information support for the formulation of the dynamic signal control strategy, ensuring that the strategy can make timely and effective adjustments for various emergencies.

[0033] The real-time adjustment module 40 is used to send the dynamic signal control strategy to the signal control unit, and the signal control unit adjusts the signal state in real time to generate a regulation result.

[0034] After the formulation of the dynamic signal light control strategy is completed, it is promptly sent to the signal light control unit. This step can be achieved through a dedicated communication channel, ensuring the rapid transmission and accurate reception of the control strategy. The signal light control unit, as a core component of the traffic management system, is responsible for adjusting the status of the signal lights in real time according to the received control strategy.

[0035] The issuance of the control strategy triggers the automatic response mechanism of the signal light control unit. The control unit first analyzes the key information in the control strategy, including the optimal green light duration of each phase, the phase switching priority, and the emergency response mode, etc. Subsequently, it precisely adjusts the signal light timing plan according to this information to ensure the efficient and orderly management of the traffic flow at the intersection.

[0036] During the implementation of the control strategy, the signal light control unit also continuously monitors the traffic conditions at the intersection and generates control results in real time. The control results are used to characterize key indicators such as the traffic flow change, vehicle waiting time, and intersection passing efficiency after the implementation of the strategy. By comparing the data before and after the control, the effectiveness of the dynamic signal light control strategy can be evaluated, and necessary adjustments and optimizations can be made in a timely manner.

[0037] Finally, through the real-time adjustment of the signal light control unit, the dynamic signal light control strategy is transformed into actual traffic management effects. It not only improves the passing capacity of the intersection but also enhances the flexibility and adaptability of the traffic system, providing a strong guarantee for the smooth operation of urban traffic.

[0038] The mining and optimization module 50 is used to store the control results and the traffic flow data in the traffic management database for data mining and optimization, generate signal light control optimization suggestions, and dynamically control the intelligent traffic signal lights.

[0039] Next, the specific configuration of the mining and optimization module 50 will be described in detail. As described above, the mining and optimization module 50 can further include: a data identification unit for counting the control results and the traffic flow data of the target road for consecutive M months, identifying high-frequency congestion periods and inefficient phases, where M is an integer greater than 1; a scheme adjustment unit for adjusting the default signal light timing plan based on the high-frequency congestion periods to generate a first timing plan; a switching optimization unit for optimizing the phase switching based on the inefficient phases to determine the phase switching data; and a data mining unit for storing the first timing plan and the phase switching data in the traffic management database for data mining and optimization to generate the signal light control optimization suggestions.

[0040] In order to continuously optimize the traffic light control strategy, the control results and traffic flow data of the target road for M consecutive months (M>1) can be counted. This long-term data accumulation can provide rich traffic condition information and help identify high-frequency congestion periods and inefficient phases.

[0041] First, we use statistical analysis methods to filter out the periods of frequent congestion from a large amount of data. These periods usually correspond to critical moments with heavy traffic and low traffic efficiency. Based on these high-frequency congestion periods, we make targeted adjustments to the default timing plan of traffic lights, generating a first timing plan that is more suitable for actual traffic needs.

[0042] Secondly, we conducted in-depth phase switching optimization for the identified inefficient phases. By analyzing the changes in traffic flow during the phase switching process, we determined the optimal phase switching timing and sequence, thereby generating phase switching data. These data provide an important basis for the refined adjustment of traffic light control.

[0043] Finally, the first timing scheme and phase switching data are stored in the traffic management database, and further analyzed and optimized using data mining technology. Traffic light control optimization suggestions are generated by mining the correlation and regularity between data. These suggestions are aimed at further improving the traffic efficiency of intersections, reducing traffic congestion, and providing more scientific and effective decision support for urban traffic management. It not only reflects the data-driven management concept, but also demonstrates the huge potential of intelligent transportation systems in urban traffic optimization.

[0044] The specific configuration of the mining optimization module 50 will be described in detail below. As described above, the mining optimization module 50 may further include: an indicator formulation unit, which is used to implement the first timing scheme and the phase switching data at the target intersection, and continuously monitor the traffic flow data to formulate a vehicle delay index; a data determination unit, which is used to calculate the vehicle delay improvement rate according to the vehicle delay index, and determine whether the vehicle delay improvement rate is lower than the preset improvement rate; a data updating unit, which is used to update the first timing scheme and the phase switching data until the preset improvement rate is met when the vehicle delay improvement rate is lower than the preset improvement rate, and generate the i-th timing scheme and the j-th phase switching data, where i and j are positive integers not equal to 0; a data storage unit, which is used to store the i-th timing scheme and the j-th phase switching data in the traffic management database.

[0045] After implementing the first timing plan and phase switching data at the target intersection, the traffic flow data is continuously monitored, and vehicle delay indicators are formulated to evaluate the control effect. By calculating the average delay time of vehicles passing through the intersection, the impact of the control strategy on traffic flow is intuitively reflected.

[0046] Subsequently, the vehicle delay improvement rate was calculated based on the vehicle delay index and compared with the preset improvement rate. If the improvement rate is lower than the preset value, it indicates that there is still room for optimization in the current regulation strategy. The first timing plan and phase switching data were iteratively updated by adjusting parameters such as the green light duration and phase switching sequence in order to achieve higher traffic efficiency.

[0047] This update process continues until the vehicle delay improvement rate reaches or exceeds the preset improvement rate. At this time, the i-th timing plan and the j-th phase switching data that meet the requirements (i and j are positive integers not equal to 0) are obtained, and these optimized plans are stored in the traffic management database. This not only improves the traffic capacity of the intersection but also provides strong support for subsequent optimization work.

[0048] In the above text, reference is made to Figure 1 described in detail the intelligent road traffic signal dynamic regulation system based on artificial intelligence according to the embodiments of the present application. Next, reference will be made to Figure 2 describe the intelligent road traffic signal dynamic regulation method based on artificial intelligence according to the embodiments of the present application.

[0049] Step A100, constructing a traffic state feature vector based on traffic flow data; in a possible implementation manner, step A100 further includes step A110, extracting historical traffic flow data of the target intersection, calculating the average flow, vehicle speed standard deviation, and queue length change rate at different time periods; performing step A120, identifying abnormal event features to construct an event impact weight; performing step A130, predicting congestion propagation based on historical congestion patterns to obtain a congestion prediction result; performing step A140, fusing the average flow, the vehicle speed standard deviation, the queue length change rate, the event impact weight, and the congestion prediction result to obtain a traffic state feature vector.

[0050] Performing step A200, performing real-time analysis on the traffic state feature vector through a deep learning model, predicting the traffic flow change trend within a future time window, and generating a prediction result; in a possible implementation manner, step A200 further includes step A210, collecting traffic flow data of the target intersection for N consecutive days, constructing a training data set, the training data set including the traffic state feature vector and the corresponding actual traffic flow change, N is an integer greater than 2; performing step A220, using a convolutional neural network to extract the spatio-temporal features of the traffic flow data to generate high-dimensional features; performing step A230, performing time series modeling on the high-dimensional features through a long short-term memory network to obtain the deep learning model.

[0051] Execute step A300 to generate a dynamic signal control strategy according to the prediction result. In a possible implementation, step A300 further includes step A310 to extract the predicted trend of traffic flow changes according to the prediction result, calculate the optimal green light duration for each phase, and obtain the optimal green light durations for multiple phases; execute step A320 to identify high-priority phases, determine the phase switching priority, and when an abnormal event is detected, activate the emergency response mode to forcibly extend the green light duration of relevant phases and shorten the durations of other phases through the emergency response mode; execute step A330 to integrate the optimal green light durations for multiple phases, the phase switching priority, and the emergency response mode into the dynamic signal control strategy.

[0052] In a possible implementation, step A320 further includes step A signal control to analyze the road construction plan cycle based on the target road and generate a temporary road closure event; execute step A322 to analyze the accident impact based on the target road and generate a traffic accident event; execute step A323 to analyze the weather visibility based on the target road and generate a severe weather event; execute step A324 to add the temporary road closure event, the traffic accident event, and the severe weather event to the abnormal event.

[0053] Execute step A400 to send the dynamic signal control strategy to the signal control unit, and adjust the signal status in real time through the signal control unit to generate a control result; execute step A500 to store the control result and the traffic flow data in the traffic management database for data mining and optimization, and generate a signal control optimization suggestion to dynamically control the intelligent traffic signal.

[0054] In a possible implementation, step A500 further includes step A510 to count the control results and the traffic flow data of the target road for consecutive M months, identify the high-frequency congestion periods and inefficient phases, where M is an integer greater than 1; execute step A520 to adjust the default signal timing plan based on the high-frequency congestion periods to generate a first timing plan; execute step A530 to optimize the phase switching based on the inefficient phases and determine the phase switching data; execute step A540 to store the first timing plan and the phase switching data in the traffic management database for data mining and optimization, and generate the signal control optimization suggestion.

[0055] In a possible implementation, step A540 further includes step A541, implementing the first timing plan and the phase switching data at the target intersection, continuously monitoring traffic flow data, and formulating a vehicle delay index; performing step A542, calculating a vehicle delay improvement rate according to the vehicle delay index, and determining whether the vehicle delay improvement rate is lower than a preset improvement rate; performing step A543, when the vehicle delay improvement rate is lower than the preset improvement rate, updating the first timing plan and the phase switching data until the preset improvement rate is met, generating the i-th timing plan and the j-th phase switching data, where i and j are positive integers not equal to 0; performing step A544, storing the i-th timing plan and the j-th phase switching data in the traffic management database.

[0056] The embodiments of the present application solve the technical problem that traditional signal lights are difficult to flexibly adjust according to real-time traffic flow, and achieve the technical effects of dynamically adjusting the signal light timing to realize intelligent timing, reducing congestion and the accident rate.

[0057] The intelligent road traffic signal light dynamic regulation system based on artificial intelligence provided by the embodiments of the present application can execute the intelligent road traffic signal light dynamic regulation method provided by any embodiment of the present application, and has the corresponding functional modules and beneficial effects for executing the method.

[0058] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or the server. The included individual units and modules are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present application.

[0059] The above specific implementation manners do not constitute a limitation to the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present application shall be included within the protection scope of the present application.

Claims

1. The intelligent road traffic light dynamic control system based on artificial intelligence is characterized by: The system comprises: A vector construction module is used to construct a traffic state feature vector based on traffic flow data; A real-time analysis module, used to perform real-time analysis on the traffic state feature vector through a deep learning model, predict the traffic flow change trend in a future time window, and generate a prediction result; A control analysis module, used to generate a dynamic traffic light control strategy according to the prediction results; A real-time adjustment module, used to send the dynamic traffic light control strategy to the traffic light control unit, adjust the traffic light state in real time through the traffic light control unit, and generate a control result; The mining and optimization module is used to store the control results and the traffic flow data in the traffic management database for data mining and optimization, generate traffic light control optimization suggestions, and dynamically control the intelligent traffic lights.

2. The artificial intelligence-based intelligent road traffic signal light dynamic control system as claimed in claim 1, characterized in that: Vector building blocks include: The first calculation unit is used to extract the historical traffic flow data of the target intersection and calculate the average flow rate, the standard deviation of vehicle speed and the change rate of queue length in different time periods; A weight building unit, used to identify abnormal event features and build event impact weights; A first prediction unit, configured to perform congestion propagation prediction based on historical congestion patterns to obtain a congestion prediction result; The data fusion unit is used to fuse the average flow, the vehicle speed standard deviation, the queue length change rate, the event impact weight, and the congestion prediction result to obtain a traffic state feature vector.

3. The artificial intelligence-based intelligent road traffic signal light dynamic control system as claimed in claim 1, characterized in that: Real-time analysis modules include: A data training unit, used to collect traffic flow data of a target intersection for N consecutive days to construct a training data set, wherein the training data set includes the traffic state feature vector and the corresponding actual traffic flow change, where N is an integer greater than 2; A feature construction unit, used for extracting the spatiotemporal features of the traffic flow data using a convolutional neural network to generate high-dimensional features; A data modeling unit is used to perform time series modeling on the high-dimensional features through a long short-term memory network to obtain the deep learning model.

4. The artificial intelligence-based intelligent road traffic light dynamic control system as claimed in claim 1, characterized in that: The regulatory analysis module includes: A second calculation unit is used to extract the traffic flow change prediction trend according to the prediction result, calculate the optimal green light duration of each phase, and obtain the optimal green light duration of multiple phases; A data detection unit, used to identify high-priority phases, determine phase switching priorities, and when an abnormal event is detected, start an emergency response mode, through which the green light duration of the relevant phase is forcibly extended and the duration of other phases is shortened; A data integration unit is used to integrate the multiple phase-optimal green light durations, the phase switching priorities, and the emergency response mode into the dynamic traffic light control strategy.

5. The artificial intelligence-based intelligent road traffic signal light dynamic control system as claimed in claim 4, characterized in that: The regulatory analysis module includes: A first event generating unit is used to analyze the road construction plan cycle according to the target road and generate a temporary road closure event; A second event generating unit, used for performing accident impact analysis according to a target road and generating a traffic accident event; a third event generating unit, configured to perform weather visibility analysis on a target road and generate a severe weather event; The fourth event generating unit is used to add the temporary road closure event, the traffic accident event, and the severe weather event to the abnormal events.

6. The artificial intelligence-based intelligent road traffic signal light dynamic control system as claimed in claim 1, characterized in that: Mining optimization modules include: A data identification unit, used to collect statistics on the control results of the target road for M consecutive months and the traffic flow data, and identify high-frequency congestion periods and inefficient phases, where M is an integer greater than 1; A scheme adjustment unit, configured to adjust a default timing scheme of the traffic light based on the high-frequency congestion period to generate a first timing scheme; A switching optimization unit, configured to perform phase switching optimization based on the inefficient phase and determine phase switching data; A data mining unit is used to store the first timing scheme and the phase switching data in the traffic management database for data mining optimization, and generate the traffic light control optimization suggestion.

7. The artificial intelligence-based intelligent road traffic signal light dynamic control system as claimed in claim 6, characterized in that: Mining optimization modules include: An indicator formulation unit, configured to implement the first timing scheme and the phase switching data at a target intersection, and continuously monitor traffic flow data to formulate a vehicle delay indicator; A data determination unit, configured to calculate a vehicle delay improvement rate according to the vehicle delay index, and determine whether the vehicle delay improvement rate is lower than a preset improvement rate; A data updating unit, configured to update the first timing scheme and the phase switching data until the preset improvement rate is satisfied when the vehicle delay improvement rate is lower than the preset improvement rate, and generate an i-th timing scheme and a j-th phase switching data, where i and j are positive integers not equal to 0; A data storage unit is used to store the i-th timing scheme and the j-th phase switching data in the traffic management database.

8. The method for dynamic control of intelligent road traffic lights based on artificial intelligence is characterized by: The method is used to implement the artificial intelligence-based intelligent road traffic signal light dynamic control system according to any one of claims 1 to 7, and the method comprises: Based on traffic flow data, construct traffic state feature vector; By using a deep learning model, the traffic state feature vector is analyzed in real time to predict the traffic flow change trend in a future time window and generate a prediction result; generating a dynamic traffic light control strategy according to the prediction result; The dynamic signal light control strategy is sent to the signal light control unit, and the signal light state is adjusted in real time by the signal light control unit to generate a control result; The control result and the traffic flow data are stored in a traffic management database for data mining optimization, and traffic light control optimization suggestions are generated to dynamically control the intelligent traffic lights.

Images