An urban road network dynamic traffic coordination control method, system and medium
By acquiring and processing urban road network traffic data, and using LSTM models for prediction and compensation adjustment, a dynamic signal control scheme is generated, which solves the problem of slow response in traditional traffic control systems and improves traffic control efficiency and smoothness.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- AI SUPER EYE TECH CO LTD
- Filing Date
- 2026-03-12
- Publication Date
- 2026-07-14
Smart Images
Figure CN122392330A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent transportation technology, specifically to a method, system, and medium for dynamic traffic coordination and control of urban road networks. Background Technology
[0002] In urban traffic management, traditional traffic control systems often face the problem of slow response speed. Due to the complexity of traffic flow and road condition changes, traditional methods rely on fixed models and preset rules, which often cannot adapt to sudden traffic events or changes in a timely manner, such as traffic accidents or weather changes. These systems usually require a long time to update and adjust control strategies, resulting in delayed traffic signal responses and thus failing to effectively alleviate traffic congestion or improve road efficiency. Summary of the Invention
[0003] This application provides a method, system, and medium for dynamic traffic coordination control of urban road networks, which addresses the technical problem of slow response in traffic prediction and control in the prior art.
[0004] In view of the above problems, this application provides a method, system and medium for dynamic traffic coordination control of urban road networks.
[0005] A first aspect of this application provides a method for dynamic traffic coordination and control of urban road networks, the method comprising:
[0006] Multiple traffic operation data points from multiple intersections in the urban road network are acquired, and synchronous processing and validity screening are performed to form multiple standardized traffic operation data points. Based on these standardized traffic operation data points, the traffic demand of the urban road network within a future preset time window is predicted and analyzed to obtain a basic control demand prediction vector. An initial signal control scheme is generated based on the basic control demand prediction vector, wherein the initial signal control scheme describes the cycle settings, release ratios, and traffic coordination relationships of each intersection within the time window. The initial signal control scheme is then subject to targeted compensation adjustment to obtain a target signal control scheme, which is then sent to the signal control equipment for execution.
[0007] A second aspect of this application provides a dynamic traffic coordination and control system for urban road networks, the system comprising:
[0008] The data acquisition module is used to acquire multiple traffic operation data from multiple intersections in the urban road network, perform synchronous processing and validity filtering, and form multiple standardized traffic operation data. The predictive analysis module is used to predict and analyze the traffic demand of the urban road network within a future preset time window based on the multiple standardized traffic operation data, and obtain a basic control demand prediction vector. The scheme generation module is used to generate an initial signal control scheme based on the basic control demand prediction vector, wherein the initial signal control scheme describes the cycle setting, release ratio and traffic coordination relationship of each intersection within the time window. The execution module is used to perform targeted compensation adjustment on the initial signal control scheme to obtain a target signal control scheme, and send the target signal control scheme to the signal control equipment for execution.
[0009] A third aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements a dynamic traffic coordination control method for urban road networks provided in this application.
[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0011] This application acquires multiple traffic operation data points from multiple intersections in an urban road network, performs synchronous processing and validity screening to form multiple standardized traffic operation data points. Based on these standardized traffic operation data points, it predicts and analyzes the traffic demand of the urban road network within a future preset time window to obtain a basic control demand prediction vector. An initial signal control scheme is generated based on the basic control demand prediction vector, wherein the initial signal control scheme describes the cycle settings, release ratios, and traffic coordination relationships of each intersection within the time window. The initial signal control scheme is then subject to targeted compensation adjustment to obtain a target signal control scheme, which is then sent to the signal control equipment for execution. This invention solves the technical problem of slow response in traffic prediction and control in the prior art, achieving the technical effects of improving traffic control response speed and optimizing traffic flow through traffic data acquisition and real-time signal control adjustment. Attached Figure Description
[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0013] Figure 1 A schematic flowchart of a dynamic traffic coordination and control method for urban road networks is provided for an embodiment of this application;
[0014] Figure 2 This is a schematic diagram of a dynamic traffic coordination and control system for urban road networks provided in an embodiment of this application.
[0015] Figure labeling: Data acquisition module 11, predictive analysis module 12, scheme generation module 13, execution module 14. Detailed Implementation
[0016] This application provides a method, system, and medium for dynamic traffic coordination control of urban road networks. It addresses the technical problem of slow response in traffic prediction and control in existing technologies by acquiring traffic data and adjusting real-time signal control, thereby improving the response speed of traffic control and optimizing traffic flow.
[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.
[0018] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.
[0019] Example 1, as Figure 1 As shown, this application provides a method for dynamic traffic coordination and control of urban road networks, the method comprising:
[0020] Step S100: Obtain multiple traffic operation data from multiple intersections in the urban road network, perform synchronous processing and validity screening, and form multiple standardized traffic operation data.
[0021] In this embodiment, traffic operation data from multiple intersections in the urban road network, including traffic flow, vehicle speed, queue length, and traffic signal status, are first acquired using roadside sensors, surveillance cameras, and vehicle-mounted GPS devices. The collected data is transmitted wirelessly to a data processing platform to form a raw traffic dataset. Next, this data is processed synchronously, with traffic data from each intersection sorted by time using timestamps and aligned using a time synchronization algorithm to ensure that all data is compared and analyzed within the same time window.
[0022] Then, a validity screening process is performed. By setting threshold ranges, the collected traffic data is compared and outliers outside the normal range are removed, such as zero vehicle speed or abnormal traffic flow. Next, statistical methods are used to detect and remove missing or duplicate data caused by equipment malfunction or data loss, ensuring that all retained data meets the validity requirements.
[0023] Finally, data standardization is performed to transform the data collected from different intersections into a unified standard format. Through normalization, data with different dimensions, such as traffic flow and speed, are converted to the same numerical range, enabling effective comparison and analysis of traffic data from different intersections on the same scale. After this series of processes, multiple standardized traffic operation data sets are ultimately generated.
[0024] Step S200: Based on the multiple standardized traffic operation data, predict and analyze the traffic demand of the urban road network within a future preset time window to obtain a basic control demand prediction vector.
[0025] In this embodiment, historical traffic data is first collected, including information such as traffic flow, vehicle speed, queue length, and traffic signal status at multiple intersections. The data originates from traffic monitoring equipment such as roadside sensors, video surveillance, and vehicle-mounted GPS devices. Next, data preprocessing ensures that the acquired data can be compared and analyzed under the same standard. Time synchronization processing aligns data from different intersections and time periods, eliminating biases caused by time differences. Simultaneously, validity screening removes invalid data and outliers, ensuring data quality. At this point, all traffic data is converted into a standardized format, making data from different intersections comparable and providing a reliable basis for predicting future traffic demand.
[0026] Then, based on standardized traffic operation data, traffic demand forecasting analysis is performed within a preset time window. In this process, a Long Short-Term Memory (LSTM) network model is used for training. The training data includes standardized traffic data from multiple intersections over historical time periods, including features such as traffic flow, vehicle speed, and queue length. Simultaneously, the basic control demand prediction vector is also included in the training data as the target output data. The training data consists of historical traffic operation data and corresponding traffic control demands. The input data represents past traffic conditions, such as past traffic flow, vehicle speed, and queue length, while the output data represents the traffic control demands within the corresponding time window, set by technical experts based on historical data and traffic management objectives. During training, the LSTM model learns the relationship between the input and output data, adjusting its internal parameters to capture the long-term dependence and changing patterns of traffic demand. During training, the LSTM model optimizes weights and biases through the backpropagation algorithm, gradually improving its ability to predict changes in traffic demand.
[0027] Finally, after training, the trained LSTM model is fed with multiple standardized traffic operation data. Based on this input data, the model predicts traffic demand within a preset time window and outputs a basic control demand prediction vector. This vector contains traffic demand information for each intersection in the future time period, such as traffic flow, queue length, and other traffic control requirements.
[0028] Step S300: Generate an initial signal control scheme based on the basic control demand prediction vector, wherein the initial signal control scheme is used to describe the cycle settings, release ratios, and traffic coordination relationships of each intersection within the time window.
[0029] In this embodiment, the cycle setting is first calculated based on the traffic demand data for each intersection provided in the basic control demand prediction vector, such as traffic flow and queue length. The cycle setting refers to the signal cycle at each intersection, specifically including the cycle length of the traffic lights and the signal transition duration. When performing this step, the cycle length is first determined based on the predicted traffic flow. Assume the traffic flow at each intersection is Q. i Then, the formula is used to calculate T under the period setting: i =T min +α×Q i Perform the calculation, where T min T is the preset shortest signal cycle for each intersection, and α is a preset adjustment coefficient representing the relationship between traffic volume and cycle length. Intersections with higher traffic volume have longer cycle lengths (Tcycle and Tcycle). i The longer the cycle time T, the better, ensuring the intersection can handle more traffic flow. Conversely, for intersections with low traffic volume, the cycle time T is longer. i It will be shorter.
[0030] Next, the release ratio is calculated based on traffic flow and queue length. The release ratio refers to the proportion of green light time within a signal cycle to the total cycle. The release ratio is calculated based on traffic flow Q. i and queue length L i We apply weighted averages. Assume the allowance ratio P at each intersection. i Through formula Calculations are performed, in which, It is the maximum traffic volume among all intersections. This is the maximum queue length across all intersections. The formula indicates that when both traffic volume and queue length are high, the release ratio should be... This will increase accordingly, thereby extending the green light duration at the intersection and ensuring that more vehicles can pass through.
[0031] Then, traffic coordination is calculated to ensure that traffic lights at multiple intersections can work in coordination, avoiding simultaneous traffic flow at different intersections and causing congestion. In this process, the signal control relationships between adjacent intersections are first analyzed based on the road network topology. Assuming there are two adjacent intersections R1 and R2, with some traffic flow influence between them, the signal switching timing is calculated by analyzing the traffic flow predictions of adjacent intersections and using the principle of maximizing traffic flow. Specifically, for adjacent intersections R1 and R2, assuming their green light cycles are G1 and G2 respectively, their switching relationship can be expressed as G1 + G2 = T. cycle Perform the calculation, where T cycle This refers to the total cycle time between the two intersections. Optimization algorithms, such as the shortest path algorithm, calculate the green light time between adjacent intersections based on the predicted traffic volumes Q1 and Q2, enabling coordinated signal light switching at adjacent intersections and avoiding traffic bottlenecks.
[0032] Through the above steps, an initial signal control scheme is finally generated. This scheme includes specific parameters such as the signal cycle, green light duration, traffic flow ratio, and signal coordination with adjacent intersections for each intersection. For example, for intersection R1 with high traffic volume, its cycle is set to a longer time T1, and its traffic flow ratio is set to a higher ratio P1. At the same time, it coordinates with the adjacent intersection R2 by adjusting the signal switching timing to ensure the efficiency of the road network and the smooth flow of vehicles.
[0033] Step S400: Perform targeted compensation adjustment on the initial signal control scheme to obtain a target signal control scheme, and send the target signal control scheme to the signal control device for execution.
[0034] In this embodiment, when performing targeted compensation adjustment on the initial signal control scheme, firstly, the intersection to be coordinated is screened based on multiple standardized traffic operation data, and the target intersection is determined by analyzing the traffic operation status of multiple intersections; then, multiple neighboring nodes that have traffic correlation with the target intersection are selected, and the standardized operation data of multiple neighboring nodes are jointly correlated by combining vehicle running time and congestion propagation patterns to obtain neighbor correlation characteristics; finally, the initial signal control scheme is compensated and adjusted based on the neighbor correlation characteristics, and the relevant control parameters in the initial signal control scheme are adjusted to determine the target signal control scheme.
[0035] Next, the target signal control scheme is sent to the signal control equipment for execution. Specifically, the control parameters in the target signal control scheme are processed into commands to form corresponding signal control command data. The control parameters include the cycle settings, green light release ratios, and phase connection relationships for each intersection. Subsequently, the signal control command data is sent to the signal control equipment at the corresponding intersection. After receiving the signal control command data, the signal control equipment parses the control parameters and controls the release sequence and duration of each phase of the traffic lights according to the cycle settings, green light release ratios, and phase connection relationships.
[0036] Furthermore, in the method provided in the application embodiments, the method of performing targeted compensation adjustment on the initial signal control scheme to obtain the target signal control scheme further includes:
[0037] Based on the aforementioned standardized traffic operation data, the intersection to be coordinated is screened to determine the target intersection; multiple neighboring nodes with traffic correlation with the target intersection are selected, and the standardized operation data of multiple neighboring nodes are jointly correlated by combining vehicle travel time and congestion propagation patterns to obtain neighbor correlation characteristics; based on the neighbor correlation characteristics, the initial signal control scheme is compensated and adjusted to determine the target signal control scheme.
[0038] In this embodiment of the application, when screening intersections for coordination based on multiple standardized traffic operation data, firstly, multiple intersection operation status indicators are constructed according to the traffic saturation level, queue length occupancy ratio, and traffic delay trend of each intersection in the multiple standardized traffic operation data; then, threshold comparison is performed on the multiple intersection operation status indicator sets, and intersections with indicators exceeding the preset range are selected as candidate intersections to obtain a candidate intersection set; finally, the target intersection is determined by combining the topological location of the candidate intersection set in the road network and its influence range on surrounding intersections.
[0039] Next, when selecting multiple neighboring nodes that have traffic relationships with the target intersection, and performing joint correlation processing on the standardized operational data of multiple neighboring nodes based on vehicle travel time and congestion propagation patterns, the following steps are taken: First, the operational propagation paths from the neighboring nodes to the target intersection are determined according to the road network topology relationship between the target intersection and multiple neighboring nodes, thus obtaining a set of operational propagation paths. Then, the arrival time set of the operational propagation path set is estimated based on historical vehicle travel time, and the standardized operational data of multiple neighboring nodes is time-aligned. Subsequently, feature extraction is performed on the time-aligned standardized operational data of multiple aligned neighboring nodes to generate multiple neighboring feature sets that characterize arrival cluster strength, queue diffusion trend, and traffic fluctuation amplitude. Finally, the multiple neighboring feature sets are jointly correlated to generate neighboring correlation features.
[0040] Finally, the initial signal control scheme is compensated and adjusted based on proximity association features. This process involves first comparing the differences between the proximity association features and the control parameters corresponding to the target intersection in the initial signal control scheme to obtain the control deviation; then, determining the compensation adjustment range based on the control deviation, and adaptively adjusting the cycle parameters, green light allocation ratio, and phase connection relationships in the initial signal control scheme according to the compensation adjustment range, forming a set of compensated and adjusted control parameters; finally, verifying the compensated and adjusted control parameter set to ensure that the minimum clearance time constraint is met, thus generating the target signal control scheme.
[0041] Furthermore, the method provided in the application embodiment, which determines the target intersection by performing intersection coordination screening based on the multiple standardized traffic operation data, further includes:
[0042] Based on the traffic saturation, queue length occupancy ratio, and traffic delay trend of each intersection in the multiple standardized traffic operation data, multiple intersection operation status indicators are constructed; threshold comparison is performed on the multiple intersection operation status indicator sets, and intersections with indicators exceeding the preset range are selected as candidate intersections to obtain a candidate intersection set; the target intersection is determined by combining the topological location of the candidate intersection set in the road network and its influence range on surrounding intersections.
[0043] In this embodiment, multiple intersection operation status indicators are first constructed based on the traffic saturation level, queue length occupancy ratio, and traffic delay trend of each intersection from multiple standardized traffic operation data. Specifically, traffic flow, vehicle speed, and queue length data of each intersection within a continuous time window are extracted from multiple standardized traffic operation data, and the traffic saturation level is calculated. First, the number of vehicles passing the stop line of the intersection within a unit time window is counted as the actual traffic flow. Then, the maximum number of vehicles that can pass through the intersection per unit time under design conditions is obtained as the design capacity of the intersection. Subsequently, the actual traffic flow is divided by the design capacity of the intersection to obtain the traffic saturation level of the intersection within the corresponding time window. Simultaneously, the queue length occupancy ratio is calculated based on the queue detection results. First, the number of queued vehicles is identified through video detection or geomagnetic detection, and the queue length is estimated based on the average vehicle length and vehicle spacing. Then, the effective road length of the approach lane is obtained, and the queue length is divided by the effective length of the approach lane. The system first calculates the average actual travel time of vehicles through the intersection within a time window, and then obtains the free-flow travel time of vehicles through the same road segment under free-flow conditions. The average travel delay within that time window is obtained by subtracting the free-flow travel time from the actual travel time. Multiple travel delay values are calculated sequentially across several consecutive time windows. The difference between the travel delay in the current time window and the travel delay in the previous time window is then calculated and divided by the travel delay in the previous time window to obtain the travel delay change rate, which characterizes the travel delay change trend. Finally, the traffic saturation level, queue length occupancy ratio, and travel delay change trend are combined according to a unified indicator structure to construct multiple intersection operation status indicators.
[0044] Subsequently, threshold comparisons were performed on multiple sets of intersection operation status indicators to filter out intersections whose indicators exceeded preset ranges as candidate intersections. Specifically, the operation status indicators of each intersection were first normalized. This involved first calculating the maximum and minimum values of the corresponding indicators for all intersections, then subtracting the minimum value from the indicator value for each intersection, and finally dividing the result by the difference between the maximum and minimum values, thereby unifying indicators of different dimensions to the same numerical range. Next, thresholds for traffic saturation, queue length occupancy ratio, and traffic delay trend were set, and the indicator values for each intersection were compared with the corresponding thresholds item by item. When any indicator value for an intersection exceeded the corresponding threshold, the intersection was determined to be an intersection with abnormal traffic operation status. Finally, all intersections determined to be in abnormal operation status were aggregated to form a candidate intersection set.
[0045] Finally, the target intersection is determined by combining the topological location of the candidate intersection set in the road network and its influence on surrounding intersections. Specifically, a road network topology is constructed based on the connection relationships between urban roads, where intersections are treated as nodes and roads are treated as edges connecting nodes to form a road network topology graph. Then, the number of directly connected adjacent intersections to each candidate intersection is counted, and this number is taken as the connectivity degree of that intersection in the road network. A corresponding connectivity degree value is obtained for each candidate intersection. Simultaneously, the traffic flow between candidate intersections and adjacent intersections is calculated, and the average travel time of vehicles from a candidate intersection to an adjacent intersection is obtained. Then, the traffic flow on each connecting road is divided by the corresponding vehicle travel time to obtain the traffic propagation influence intensity of the corresponding connecting road. The traffic propagation influence intensity between the candidate intersection and all adjacent intersections is accumulated to obtain the overall traffic influence range value of the candidate intersection. Then, for each candidate intersection in the candidate intersection set, its connectivity value and overall traffic influence range value are recorded. The connectivity value and overall traffic influence range value are then weighted and summed. The connectivity value and overall traffic influence range value are first normalized, then multiplied by a preset weight coefficient and summed to obtain the comprehensive evaluation value of each candidate intersection. Finally, the candidate intersections are sorted from largest to smallest according to their comprehensive evaluation values, and the one with the largest comprehensive evaluation value is selected as the target intersection.
[0046] Furthermore, the method provided in the application embodiment, which selects multiple neighboring nodes that have traffic association relationships with the target intersection, and performs joint association processing on the standardized operational data of multiple neighboring nodes in combination with vehicle travel time and congestion propagation patterns to obtain neighbor association features, also includes:
[0047] Based on the road network topology relationship between the target intersection and the multiple neighboring nodes, the operation propagation path from the neighboring nodes to the target intersection is determined, and an operation propagation path set is obtained. The arrival time set of the operation propagation path set is estimated based on historical vehicle operation time, and time alignment processing is performed on the standardized operation data of the multiple neighboring nodes. Feature extraction is performed on the time-aligned standardized operation data of the multiple aligned neighboring nodes to generate multiple neighboring feature sets characterizing the arrival cluster strength, queuing diffusion trend, and traffic fluctuation amplitude. The multiple neighboring feature sets are jointly associated to generate the neighboring association feature.
[0048] In this embodiment, the propagation path from the neighboring nodes to the target intersection is first determined based on the road network topology relationship between the target intersection and multiple neighboring nodes. Specifically, a road network topology model is constructed based on the connection relationship between urban roads, where each intersection is defined as a topological node, the road connecting two intersections is defined as a topological edge, and an adjacency matrix is used to represent the connection relationship between nodes. Then, multiple neighboring nodes with traffic association with the target intersection are identified, and a reachable path from each neighboring node to the target intersection is searched based on the shortest path search algorithm. This involves traversing the road connection relationship layer by layer to obtain vehicle operation paths composed of multiple road segments. After obtaining each reachable path, the road segments contained in the path are recorded according to the driving order, thereby forming a vehicle operation sequence composed of multiple road segments, and the vehicle operation sequence corresponding to each neighboring node is taken as a propagation path. Finally, the propagation paths corresponding to all neighboring nodes are summarized to obtain a set of propagation paths from neighboring nodes to the target intersection.
[0049] Subsequently, the arrival time set of the propagation path set is estimated based on the historical vehicle running time, and time alignment processing is performed on the standardized running data of multiple neighboring nodes. Specifically, the average travel time of vehicles on each road segment is extracted from historical traffic operation data. This is achieved by statistically analyzing the actual travel time of vehicles from the start to the end of a road segment within a certain time range and averaging the average travel time. Then, for each traffic propagation path in the set of propagation paths, the average travel time corresponding to each road segment in the path is accumulated according to the vehicle's travel order to obtain the path travel time required for a vehicle to reach the target intersection from a neighboring node. This path travel time is then used as the estimated arrival time of the traffic flow from that neighboring node to the target intersection. Subsequently, the corresponding estimated arrival time is calculated for each traffic propagation path, and all estimated arrival times are aggregated to form an arrival time set. After obtaining the arrival time set, time offset processing is performed on the standardized operational data of multiple neighboring nodes based on the estimated arrival time corresponding to each neighboring node. This involves shifting the traffic flow, vehicle speed, and queue length data of neighboring nodes in the original time series towards the target intersection time axis according to the estimated arrival time, thus aligning the traffic operation data of each neighboring node with the same arrival time position on the target intersection time axis, achieving time alignment processing of the standardized operational data of multiple neighboring nodes.
[0050] After completing the time alignment process, feature extraction is performed on the standardized running data of multiple aligned neighboring nodes after time alignment. Specifically, firstly, the time-aligned traffic flow time series data is statistically analyzed according to fixed time windows. By counting the number of vehicles arriving at the target intersection within each time window, and identifying time periods where the number of vehicles is consistently higher than the average for multiple consecutive time windows, the arrival cluster strength is obtained by calculating the relationship between the number of vehicles arriving within this time period and the length of the time window, thus characterizing the degree of concentrated vehicle arrivals at the target intersection. Subsequently, the time-aligned queue length time series data is analyzed for changes. By calculating the difference between queue length values within consecutive time windows, and judging whether the queue is growing or dissipating based on the positive or negative direction of the queue length change, the queue diffusion trend is obtained, which is used to characterize the expansion speed of queued vehicles on the road. At the same time, the time-aligned traffic flow time series data is analyzed for fluctuations. By calculating the deviation between the traffic flow value within consecutive time windows and the average traffic flow for that time period, and statistically analyzing the deviation values, the range of traffic flow changes is obtained, thus obtaining the traffic flow fluctuation amplitude, which is used to characterize the degree of fluctuation in traffic flow over a short period of time. Finally, the arrival cluster strength, queue diffusion trend, and traffic flow fluctuation amplitude are recorded and summarized according to neighboring nodes, thereby generating multiple neighboring feature sets.
[0051] Finally, a joint association process is performed on multiple neighboring feature sets. This process begins by determining the joint association coefficient set based on the proximity of the propagation path sets; then, a weighted average is applied to the multiple neighboring feature sets based on the joint association coefficient set to obtain the neighboring associated features.
[0052] Furthermore, in the method provided in the application embodiments, the joint association processing of the plurality of neighbor feature sets to generate the neighbor association feature further includes:
[0053] Based on the proximity of the set of propagation paths, a joint correlation coefficient set is determined; the multiple neighbor feature sets are weighted based on the joint correlation coefficient set to obtain the neighbor correlation features.
[0054] In this embodiment, the joint correlation coefficient set is first determined based on the proximity of the operational propagation path set. Specifically, the operational propagation path set describes the traffic propagation path relationship between each neighboring node and the target intersection. Each operational propagation path consists of multiple continuous road segments and characterizes the path structure of traffic flow propagating from neighboring nodes to the target intersection. Subsequently, the path length of each operational propagation path in the operational propagation path set is calculated. The path length value is obtained by counting the number of road segments contained in the path or accumulating the road distance. The proximity of neighboring nodes and the target intersection is determined based on the path length value, where a shorter path length indicates a higher proximity. After obtaining the path length corresponding to each neighboring node, all path lengths are normalized, and the corresponding joint correlation coefficient is calculated based on the normalized path length. Neighboring nodes with shorter path lengths correspond to larger joint correlation coefficients, while neighboring nodes with longer path lengths correspond to smaller joint correlation coefficients. Finally, the joint correlation coefficients calculated for all neighboring nodes are summarized to form the joint correlation coefficient set.
[0055] Subsequently, multiple neighbor feature sets are weighted based on the joint correlation coefficient set. Specifically, the multiple neighbor feature sets describe the traffic operation characteristics of each neighbor node, including feature data such as arrival cluster strength, queue diffusion trend, and flow fluctuation amplitude. During the joint correlation processing, the arrival cluster strength, queue diffusion trend, and flow fluctuation amplitude corresponding to each neighbor node are first arranged in the order of neighbor nodes, and a one-to-one correspondence is established with the corresponding joint correlation coefficient in the joint correlation coefficient set. Then, the neighbor feature value corresponding to each neighbor node is multiplied by the corresponding joint correlation coefficient to obtain the weighted feature value. The weighted feature values of all neighbor nodes are then summed to obtain the comprehensive traffic correlation result. Finally, this comprehensive traffic correlation result is used as the neighbor correlation feature to characterize the comprehensive influence of the traffic operation status of multiple neighbor nodes on the traffic operation status of the target intersection.
[0056] Furthermore, in the method provided in the application embodiments, the method for compensating and adjusting the initial signal control scheme based on the proximity association features to determine the target signal control scheme further includes:
[0057] The adjacent correlation features are compared with the control parameters corresponding to the target intersection in the initial signal control scheme to obtain the control deviation. The compensation adjustment range is determined based on the control deviation, and the period parameters, green light allocation ratio, and phase connection relationship in the initial signal control scheme are adaptively adjusted according to the compensation adjustment range to form a set of compensated and adjusted control parameters. The set of compensated and adjusted control parameters is verified to ensure that the minimum clearance time constraint is met, and the target signal control scheme is generated.
[0058] In this embodiment, the differences between the proximity association features and the control parameters corresponding to the target intersection in the initial signal control scheme are first compared. Specifically, the arrival cluster strength, queue diffusion trend, and traffic flow fluctuation amplitude contained in the proximity association features are extracted first, and the values of these three features within the current time window are read. At the same time, the cycle parameters, green light allocation ratio, and phase connection relationship corresponding to the target intersection are extracted from the initial signal control scheme, and the green light time of each signal phase is calculated based on the cycle parameters and green light allocation ratio. Then, the number of vehicles arriving at the target intersection within a unit time window is counted, and the number of vehicles that can pass through the stop line within a unit green light time is calculated based on the green light time of each phase and the average vehicle start interval, thereby obtaining the vehicle release capacity under the current control parameters. Then, the number of concentrated vehicle arrivals, queue growth, and traffic flow changes reflected in the proximity association features are summed to obtain the traffic demand change, and the difference between the traffic demand change and the current vehicle release capacity is calculated to obtain the control deviation.
[0059] Subsequently, the compensation adjustment range is determined based on the control deviation, and the cycle parameters, green light allocation ratio, and phase connection relationships in the initial signal control scheme are adjusted according to the compensation adjustment range to form a set of control parameters after compensation adjustment. Specifically, firstly, the control deviation is multiplied by a preset adjustment coefficient to obtain the compensation adjustment range, and the cycle parameters in the initial signal control scheme are adjusted according to the compensation adjustment range. The compensation adjustment range is increased or decreased based on the original cycle parameters to obtain a new signal cycle. Then, under the new signal cycle conditions, the green light time for each phase is recalculated. Firstly, the number of vehicles arriving in the corresponding lanes for each phase is counted, and the green light time in the signal cycle is redistributed according to the proportion of vehicles arriving in each phase to the total number of vehicles, thus obtaining a new green light allocation ratio. After completing the green light allocation ratio adjustment, the release order of each signal phase is rearranged according to the updated green light time, and the time connection between adjacent phases is adjusted so that the green light times of each phase are arranged sequentially within the signal cycle, thus forming a new phase connection relationship. After completing the above adjustments, the updated cycle parameters, green light allocation ratio, and phase connection relationships are combined to form a set of control parameters after compensation adjustment.
[0060] Finally, the adjusted control parameter set is verified to ensure that the minimum clearance time constraint is met, and the target signal control scheme is generated. Specifically, firstly, the green light time corresponding to each signal phase in the adjusted control parameter set is read, and the green light time of each signal phase is compared with the preset minimum clearance time item by item. When the green light time of a certain signal phase is less than the minimum clearance time, the green light time of that signal phase is increased to the minimum clearance time, and the green light time of other phases is reduced accordingly while keeping the total length of the cycle parameter unchanged. Then, it is checked again whether the green light time of each phase is not less than the minimum clearance time. When the green light time of all phases meets the minimum clearance time constraint, the cycle parameter, green light allocation ratio, and phase connection relationship are used as the final control parameters, thereby generating the target signal control scheme.
[0061] In summary, the embodiments of this application have at least the following technical effects:
[0062] This application acquires multiple traffic operation data points from multiple intersections in an urban road network, performs synchronous processing and validity screening to form multiple standardized traffic operation data points. Based on these standardized traffic operation data points, it predicts and analyzes the traffic demand of the urban road network within a future preset time window to obtain a basic control demand prediction vector. An initial signal control scheme is generated based on the basic control demand prediction vector, wherein the initial signal control scheme describes the cycle settings, release ratios, and traffic coordination relationships of each intersection within the time window. The initial signal control scheme is then subject to targeted compensation adjustment to obtain a target signal control scheme, which is then sent to the signal control equipment for execution. This invention solves the technical problem of slow response in traffic prediction and control in the prior art, achieving the technical effects of improving traffic control response speed and optimizing traffic flow through traffic data acquisition and real-time signal control adjustment.
[0063] Example 2, based on the same inventive concept as the urban road network dynamic traffic coordination control method in the foregoing examples, such as... Figure 2 As shown, this application provides a dynamic traffic coordination and control system for urban road networks. The system and method embodiments in this application are based on the same inventive concept. The system includes:
[0064] The data acquisition module 11 is used to acquire multiple traffic operation data from multiple intersections in the urban road network, perform synchronous processing and validity screening, and form multiple standardized traffic operation data. The prediction and analysis module 12 is used to predict and analyze the traffic demand of the urban road network within a future preset time window based on the multiple standardized traffic operation data, and obtain a basic control demand prediction vector. The scheme generation module 13 is used to generate an initial signal control scheme based on the basic control demand prediction vector, wherein the initial signal control scheme describes the cycle setting, release ratio and traffic coordination relationship of each intersection within the time window. The execution module 14 is used to perform targeted compensation adjustment on the initial signal control scheme to obtain a target signal control scheme, and send the target signal control scheme to the signal control equipment for execution.
[0065] Furthermore, the system is also used to implement the following functions:
[0066] Based on the aforementioned standardized traffic operation data, the intersection to be coordinated is screened to determine the target intersection; multiple neighboring nodes with traffic correlation with the target intersection are selected, and the standardized operation data of multiple neighboring nodes are jointly correlated by combining vehicle travel time and congestion propagation patterns to obtain neighbor correlation characteristics; based on the neighbor correlation characteristics, the initial signal control scheme is compensated and adjusted to determine the target signal control scheme.
[0067] Furthermore, the system is also used to implement the following functions:
[0068] Based on the traffic saturation, queue length occupancy ratio, and traffic delay trend of each intersection in the multiple standardized traffic operation data, multiple intersection operation status indicators are constructed; threshold comparison is performed on the multiple intersection operation status indicator sets, and intersections with indicators exceeding the preset range are selected as candidate intersections to obtain a candidate intersection set; the target intersection is determined by combining the topological location of the candidate intersection set in the road network and its influence range on surrounding intersections.
[0069] Furthermore, the system is also used to implement the following functions:
[0070] Based on the road network topology relationship between the target intersection and the multiple neighboring nodes, the operation propagation path from the neighboring nodes to the target intersection is determined, and an operation propagation path set is obtained. The arrival time set of the operation propagation path set is estimated based on historical vehicle operation time, and time alignment processing is performed on the standardized operation data of the multiple neighboring nodes. Feature extraction is performed on the time-aligned standardized operation data of the multiple aligned neighboring nodes to generate multiple neighboring feature sets characterizing the arrival cluster strength, queuing diffusion trend, and traffic fluctuation amplitude. The multiple neighboring feature sets are jointly associated to generate the neighboring association feature.
[0071] Furthermore, the system is also used to implement the following functions:
[0072] Based on the proximity of the set of propagation paths, a joint correlation coefficient set is determined; the multiple neighbor feature sets are weighted based on the joint correlation coefficient set to obtain the neighbor correlation features.
[0073] Furthermore, the system is also used to implement the following functions:
[0074] The adjacent correlation features are compared with the control parameters corresponding to the target intersection in the initial signal control scheme to obtain the control deviation. The compensation adjustment range is determined based on the control deviation, and the period parameters, green light allocation ratio, and phase connection relationship in the initial signal control scheme are adaptively adjusted according to the compensation adjustment range to form a set of compensated and adjusted control parameters. The set of compensated and adjusted control parameters is verified to ensure that the minimum clearance time constraint is met, and the target signal control scheme is generated.
[0075] In Example 3, based on the same inventive concept as the dynamic traffic coordination and control method for urban road networks in the foregoing embodiments, this application also provides a computer-readable storage medium storing a computer program, which, when executed, implements the steps of any one of the methods described in Example 1 above.
[0076] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.
[0077] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.
Claims
1. A method for dynamic traffic coordination and control of urban road networks, characterized in that, The method includes: Acquire multiple traffic operation data from multiple intersections in the urban road network, perform synchronous processing and validity screening, and generate multiple standardized traffic operation data. Based on the aforementioned standardized traffic operation data, the traffic demand of the urban road network within a future preset time window is predicted and analyzed to obtain a basic control demand prediction vector. An initial signal control scheme is generated based on the basic control demand prediction vector, wherein the initial signal control scheme is used to describe the cycle setting, release ratio and traffic coordination relationship of each intersection within the time window; The initial signal control scheme is adjusted with target compensation to obtain a target signal control scheme, and the target signal control scheme is sent to the signal control equipment for execution.
2. The urban road network dynamic traffic coordination and control method as described in claim 1, characterized in that, The initial signal control scheme is adjusted to achieve target compensation to obtain the target signal control scheme, including: Based on the aforementioned standardized traffic operation data, the intersections to be coordinated are screened to determine the target intersections; Multiple neighboring nodes that have traffic relationships with the target intersection are selected, and the standardized operational data of multiple neighboring nodes are jointly correlated by combining vehicle travel time and congestion propagation patterns to obtain neighboring correlation characteristics. The initial signal control scheme is compensated and adjusted based on the proximity correlation characteristics to determine the target signal control scheme.
3. The urban road network dynamic traffic coordination control method as described in claim 2, characterized in that, Based on the aforementioned standardized traffic operation data, intersection coordination needs are screened to determine target intersections, including: Based on the traffic saturation, queue length occupancy ratio, and traffic delay trend of each intersection in the aforementioned standardized traffic operation data, multiple intersection operation status indicators are constructed. Threshold comparison is performed on the multiple sets of intersection operation status indicators to filter out intersections whose indicators exceed a preset range as candidate intersections, thereby obtaining a set of candidate intersections; The target intersection is determined by combining the topological location of the candidate intersection set in the road network and its influence range on surrounding intersections.
4. The urban road network dynamic traffic coordination and control method as described in claim 2, characterized in that, Multiple neighboring nodes with traffic relationships to the target intersection are selected. Combined with vehicle travel time and congestion propagation patterns, standardized operational data of these neighboring nodes are jointly correlated to obtain proximity association characteristics, including: Based on the road network topology relationship between the target intersection and the multiple neighboring nodes, the operation propagation path from the neighboring nodes to the target intersection is determined, and a set of operation propagation paths is obtained; The arrival time set of the propagation path set is estimated based on the historical vehicle running time, and time alignment processing is performed on the standardized running data of multiple neighboring nodes. Feature extraction is performed on the standardized operational data of multiple aligned neighboring nodes after time alignment to generate multiple neighboring feature sets that characterize the arrival cluster strength, queuing diffusion trend and flow fluctuation amplitude. The multiple sets of neighboring features are subjected to joint association processing to generate the neighboring association features.
5. The urban road network dynamic traffic coordination control method as described in claim 4, characterized in that, The plurality of neighbor feature sets are jointly associated to generate the neighbor association features, including: Based on the proximity of the set of propagation paths, determine the set of joint correlation coefficients; The neighboring feature is obtained by weighting the multiple neighboring feature sets based on the joint correlation coefficient set.
6. The method for dynamic traffic coordination and control of urban road networks as described in claim 2, characterized in that, Based on the proximity correlation features, the initial signal control scheme is compensated and adjusted to determine the target signal control scheme, including: The difference between the proximity association feature and the control parameter corresponding to the target intersection in the initial signal control scheme is compared to obtain the control deviation. The compensation adjustment range is determined based on the control deviation, and the period parameters, green light allocation ratio and phase connection relationship in the initial signal control scheme are adaptively adjusted based on the compensation adjustment range to form a set of control parameters after compensation adjustment. The set of control parameters after compensation and adjustment is verified to ensure that the minimum release time constraint is met, and the target signal control scheme is generated.
7. A dynamic traffic coordination and control system for urban road networks, characterized in that, The system is used to execute the urban road network dynamic traffic coordination control method as described in any one of claims 1-6, and the system includes: The data acquisition module is used to acquire multiple traffic operation data from multiple intersections in the urban road network, perform synchronous processing and validity filtering, and form multiple standardized traffic operation data. The predictive analysis module is used to predict and analyze the traffic demand of the urban road network within a future preset time window based on the multiple standardized traffic operation data, and obtain a basic control demand prediction vector. The scheme generation module is used to generate an initial signal control scheme based on the basic control demand prediction vector, wherein the initial signal control scheme is used to describe the cycle setting, release ratio and traffic coordination relationship of each intersection within the time window; The execution module is used to perform targeted compensation adjustment on the initial signal control scheme to obtain a target signal control scheme, and send the target signal control scheme to the signal control device for execution.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements a dynamic traffic coordination control method for urban road networks as described in any one of claims 1-6.