A traffic signal control method, device and electronic equipment

Abstract

The application provides a traffic signal control method, device and electronic equipment. The method obtains multi-scale traffic state data, including the import flow of each import in each intersection, the vehicle flow turning ratio of each turning in each intersection and the turning traffic capacity. The turning flow of each turning in each intersection is obtained by calculation according to the import flow and the vehicle flow turning ratio. The real-time saturation of each turning in each intersection is obtained by calculation based on the turning flow and the turning traffic capacity. The network average saturation is obtained by calculation according to the real-time saturation. The target control model is determined based on the comparison result of the network average saturation and the preset saturation. The signal timing scheme is generated according to the control strategy corresponding to the target control model. The traffic signal is controlled based on the signal timing scheme. The above method provides a more practical and accurate traffic signal control method.

Description

Technical Field

[0001] This application relates to the field of traffic management technology, and in particular to a traffic signal control method, device and electronic equipment. Background Technology

[0002] With the evolution of technologies such as video detection and intelligent connectivity, and the large-scale deployment of roadside vehicle detection equipment, traffic perception technology has achieved a leap from single-section-based flow statistics to path-level operational mechanism understanding. Among these advancements, high-definition checkpoint detectors with vehicle identification capabilities are widely installed at key urban road nodes. The multi-dimensional data they collect (including license plates, timestamps, lane positions, and vehicle speeds) lays the data foundation for in-depth road network-level traffic analysis. Based on this multi-dimensional data, key traffic flow parameters such as flow rate, speed, and travel time can be aggregated and statistically generated to characterize the travel characteristics of motor vehicles in the regional road network and support dynamic source analysis of traffic congestion.

[0003] However, relevant traffic management technologies generally suffer from problems such as insufficient data utilization, inadequate perception accuracy, and unsatisfactory control effects. Specifically, at the data acquisition level, traditional detectors are mostly deployed at intersections and main roads, with limited coverage and difficulty in obtaining refined flow information such as lane-specific and entrance-specific data. This results in significant shortcomings in both spatial granularity and completeness of traffic state perception. Simultaneously, these systems often rely on data from single detectors, lacking integrated analysis of multi-source heterogeneous information, leading to low data utilization and difficulty in supporting a complete depiction of regional road network traffic conditions. At the data fusion level, because relevant schemes only involve single-scale data, cross-scale information fusion is difficult, hindering a comprehensive understanding from cross-sectional flow to road network operational status. At the control application level, relevant control strategies are mostly based on single-scale data or preset thresholds, lacking adaptive response capabilities to dynamic changes in real-time traffic conditions. This results in low matching between control schemes and actual traffic demand, making it difficult to achieve refined balance control of regional traffic flow. Clearly, a new traffic signal control method is urgently needed to address at least one of the above problems.

[0004] It should be noted that the above content only provides background information related to this application and does not necessarily constitute prior art. Summary of the Invention

[0005] This application provides a traffic signal control method, device, and electronic device to solve the technical problems of insufficient data utilization, insufficient perception accuracy, and poor traffic signal control effect in related technologies.

[0006] Other features and advantages of this application will become apparent from the following detailed description, or may be learned in part from practice of this application.

[0007] This application provides a traffic signal control method, comprising: acquiring multi-scale traffic state data, wherein the multi-scale traffic state data includes the inlet flow of each entrance at each intersection, the traffic flow turning ratio of each turn at each intersection, and the turning capacity; calculating the turning flow of each turn at each intersection based on the inlet flow and the traffic flow turning ratio; calculating the real-time saturation of each turn at each intersection based on the turning flow and the turning capacity; calculating the network average saturation based on the real-time saturation; determining a target control model based on the comparison result between the network average saturation and a preset saturation; generating a signal timing scheme according to the control strategy corresponding to the target control model; and controlling the traffic signal based on the signal timing scheme.

[0008] In one embodiment of this application, based on the aforementioned scheme, multi-scale traffic state data is obtained, including:

[0009] Acquire vehicle passage data, target area road network data, and checkpoint equipment information, wherein the vehicle passage data includes checkpoint equipment number and lane number; establish a mapping relationship between checkpoint equipment and traffic flow based on the target area road network data and the checkpoint equipment information; group vehicles according to the checkpoint equipment number and the lane number, and count the number of vehicles in each group; combine the mapping relationship to map the number of vehicles in each group to each lane in each intersection to obtain the lane flow rate of each lane in each intersection; and count the lane flow rate according to the intersection name and the approach lane to obtain the approach flow rate of each approach lane in each intersection.

[0010] In one embodiment of this application, based on the aforementioned scheme, after obtaining vehicle passage data, the method further includes: traversing the vehicle passage data, grouping it by license plate number, calculating the time difference between two adjacent vehicle passage data for the same vehicle, wherein the vehicle passage data includes the license plate number; if the time difference is greater than a preset duration, then determining that the two adjacent vehicle passage data are not on the same path; if the time difference is less than or equal to the preset duration, then determining that the two adjacent vehicle passage data are on the same path; until the vehicle passage data traversal is completed, vehicle trajectory data is obtained, and the path of each vehicle is determined, wherein the path is determined by if... The system consists of: a dry trajectory point system; based on the checkpoint device number and the lane number, and using the mapping relationship, the representation of each trajectory point is transformed to obtain a path with a preset representation; all vehicle paths are traversed, paths with the same trajectory point sequence are grouped into one category, and the frequency of each category is counted to obtain the path flow; for each category of path, the travel time of all vehicles on the path is collected, outliers are removed using the quartile method, and the average value is calculated to obtain the average path travel time; a regional path table is obtained based on the path, the path flow, and the average path travel time.

[0011] In one embodiment of this application, after obtaining the regional path table based on the aforementioned scheme, the method further includes: constructing a regional road network based on the target regional road network data, and constructing a weighted directed graph, wherein the nodes of the weighted directed graph are key points on the road, the edges of the weighted directed graph are roads, and the weight of the edge is the distance between two key points; determining a target path from the regional path table, starting from a first trajectory point, traversing each trajectory point of the target path; determining the current node corresponding to the first trajectory point from the weighted directed graph, and determining the next node of the current node; if it is determined that the next trajectory point of the first trajectory point corresponds to the next node, then it is determined that there is no gap between the first trajectory point and the next trajectory point. Missing trajectory points; if it is determined that the next trajectory point does not correspond to the next node, then the next node is used as the starting point and the point corresponding to the next trajectory point is used as the ending point. The shortest path is searched in the weighted directed graph, and all nodes in the shortest path are identified as missing trajectory points; a complete regional path table is obtained based on the missing trajectory points and the regional path table, and the regional traffic flow is statistically obtained based on the complete regional path table; the direction of traffic flow is determined based on the positional relationship between the current trajectory point and the next trajectory point in the complete regional path table, thereby obtaining the traffic flow turning ratio of each turn in each intersection; the number of turning lanes is determined based on the complete regional path table, and the turning capacity of each turn in each intersection is determined based on the number of turning lanes and the saturation flow rate.

[0012] In one embodiment of this application, before generating a signal timing scheme based on the aforementioned scheme and the control strategy corresponding to the target control model, the method further includes: obtaining the first segment traffic flow of the target road segment in a first preset time period and the second segment traffic flow in a second preset time period, wherein the first preset time period and the second preset time period differ by a preset duration; calculating the segment traffic flow change rate based on the first segment traffic flow and the second segment traffic flow; if the segment traffic flow change rate is greater than a preset traffic flow change threshold, or if a forced update cycle is reached, then regenerating the signal timing scheme according to the control strategy corresponding to the target control model.

[0013] In one embodiment of this application, based on the aforementioned scheme, a signal timing scheme is generated according to the control strategy corresponding to the target control model, including: if the target control model is a maximum pressure control model, determining the pressure weight of each phase based on the complete regional path table; allocating green light duration according to the preset objective function of the maximum pressure control model, the pressure weight, and the first preset constraint condition to generate the signal timing scheme.

[0014] In one embodiment of this application, based on the aforementioned scheme and according to the control strategy corresponding to the target control model, a signal timing scheme is generated, which further includes: if the target control model is a Webster delay minimization model, calculating the signal period based on the total loss time and the real-time saturation; allocating the green light duration based on the turning flow, the traffic flow turning ratio, the signal period, the total loss time, and the second preset constraint condition to generate the signal timing scheme.

[0015] In one embodiment of this application, after controlling traffic signals based on the aforementioned scheme and the signal timing scheme, the method further includes: calculating a path balance index based on path flow and average path flow; calculating a regional capacity improvement based on the capacity of a first region before implementing the signal timing scheme and the capacity of a second region before implementing the signal timing scheme; calculating a key road segment saturation variance based on the real-time saturation of key road segments and the average network saturation; and evaluating the control effect of controlling traffic signals based on the signal timing scheme based on one or more of the path balance index, the regional capacity improvement, and the key road segment saturation variance.

[0016] This application also provides a traffic signal control device, comprising: a data acquisition module for acquiring multi-scale traffic state data, the multi-scale traffic state data including the inlet flow of each entrance at each intersection, the traffic flow turning ratio of each turn at each intersection, and the turning capacity; a first calculation module for calculating the turning flow of each turn at each intersection based on the inlet flow and the traffic flow turning ratio; a second calculation module for calculating the real-time saturation of each turn at each intersection based on the turning flow and the turning capacity; a model determination module for calculating the network average saturation based on the real-time saturation, and determining a target control model based on the comparison result of the network average saturation and a preset saturation; and a signal control module for generating a signal timing scheme based on the control strategy corresponding to the target control model, and controlling the traffic signal based on the signal timing scheme.

[0017] This application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the traffic signal control method as described in any of the above embodiments.

[0018] The beneficial effects of this application are as follows: The core of this application lies in constructing a new traffic management system that can comprehensively perceive the regional traffic status and achieve dynamic balance control. By integrating traffic data at different spatial and temporal scales, it achieves comprehensive perception from lane flow of each lane at each intersection to the regional road network status. The target control model is determined by calculating the average network saturation. The target control model generates a signal timing scheme based on the traffic flow balance strategy. By executing the signal timing scheme, it achieves adaptive control of traffic signals based on traffic flow balance, providing a data foundation for precise traffic management and control, thereby improving the level of urban traffic management and alleviating traffic congestion.

[0019] Furthermore, this application fully utilizes the vehicle passage data from existing road checkpoints to refine the traffic status of the target area's road network. This includes the detailed acquisition of lane flow and inlet flow at each intersection within the target area at the micro level, vehicle travel path identification and critical path localization at the meso level, and path flow estimation and road network visualization at the macro level. Through multi-scale traffic status analysis at the micro, meso, and macro levels, a seamless connection between status perception and control decision-making is achieved, which helps to implement more effective regional traffic flow balance control.

[0020] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0022] In the attached diagram: Figure 1 This is a schematic flowchart illustrating a traffic signal control method according to an exemplary embodiment of this application; Figure 2 This is a schematic diagram illustrating the path recognition process of a traffic signal control method according to an exemplary embodiment of this application; Figure 3 This is a schematic diagram illustrating the path reconstruction process of a traffic signal control method according to an exemplary embodiment of this application; Figure 4 This is an exemplary embodiment of the traffic signal control method shown in this application, illustrating an adaptive signal control framework. Figure 5 This is a flowchart illustrating a traffic signal control method according to another exemplary embodiment of this application; Figure 6 This is a block diagram illustrating a traffic signal control device in an exemplary embodiment of this application; Figure 7 A schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application is shown. Detailed Implementation

[0023] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0024] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. The drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the shape, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0025] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the present application. However, it will be apparent to those skilled in the art that embodiments of the present application may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the present application.

[0026] First, it's important to clarify that origin and destination points, or OD points for short, are points where "origin" represents the starting point and "destination" represents the ending point. In traffic network analysis, each traffic area or location can be considered a potential origin or destination point. Origin and destination points are used to describe and analyze the starting and ending points of traffic trips and are important concepts in fields such as traffic planning, traffic flow forecasting, and traffic management.

[0027] SUMO (Simulation of Urban Mobility) is an open-source, microscopic, multimodal traffic simulation software that supports the entire process from road network construction to traffic flow simulation. It is widely used in research on intelligent transportation, autonomous driving, and other fields.

[0028] Dijkstra's algorithm is a classic greedy algorithm for calculating the single-source shortest path and is applicable to non-negative weighted graphs.

[0029] Figure 1 This is a schematic flowchart illustrating a traffic signal control method according to an exemplary embodiment of this application. The traffic signal control method can be executed by a computing processing device, which can be a computer. (Refer to...) Figure 1 As shown, the traffic signal control method includes at least steps S110 to S140, which are described in detail below: In step S110, multi-scale traffic state data is acquired.

[0030] Among them, multi-scale traffic condition data includes at least the inbound traffic flow at each entrance of each intersection, the turning ratio of traffic flow at each turn of each intersection, and the turning capacity.

[0031] In one embodiment of this application, acquiring multi-scale traffic state data includes: acquiring vehicle passage data, target area road network data, and checkpoint equipment information, wherein the vehicle passage data includes checkpoint equipment number and lane number; establishing a mapping relationship between checkpoint equipment and traffic flow based on the target area road network data and checkpoint equipment information; grouping vehicles according to checkpoint equipment number and lane number, and counting the number of vehicles in each group; mapping the number of vehicles in each group to each lane in each intersection based on the mapping relationship to obtain the lane flow of each lane in each intersection; and counting the lane flow according to the intersection name and approach lane to obtain the approach flow of each approach lane in each intersection.

[0032] In one embodiment of this application, after obtaining vehicle passage data, the method further includes: traversing the vehicle passage data, grouping it by license plate number, calculating the time difference between two adjacent vehicle passage data points for the same vehicle, where the vehicle passage data includes the license plate number; if the time difference is greater than a preset duration, then it is determined that the two adjacent vehicle passage data points do not share the same path; if the time difference is less than or equal to the preset duration, then it is determined that the two adjacent vehicle passage data points share the same path; this process continues until the vehicle passage data traversal is complete, resulting in vehicle trajectory data and determining the path of each vehicle. The vehicle trajectory data is stored using a dictionary structure, where the key of the dictionary structure is the license plate number. The value represents the vehicle's path, which consists of several trajectory points. Based on the checkpoint device number and lane number, and using a mapping relationship, the representation of each trajectory point is transformed to obtain the path in a preset representation. All vehicle paths are traversed, and paths with the same sequence of trajectory points are grouped into one category. The frequency of occurrence of each category of paths is counted to obtain the path flow. For each category of paths, the travel time of all vehicles on the path is collected. After removing outliers using the quartile method, the average value is calculated to obtain the average travel time of the path. Based on the path, path flow, and average travel time of the path, a regional path table is obtained.

[0033] In one embodiment of this application, after obtaining the regional path table, the method further includes: constructing a regional road network based on the target regional road network data, and building a weighted directed graph, wherein the nodes of the weighted directed graph are key points on the road, the edges of the weighted directed graph are roads, and the weight of the edge is the distance between two key points; determining the target path from the regional path table, starting from the first trajectory point, traversing each trajectory point of the target path; determining the current node corresponding to the first trajectory point from the weighted directed graph, and determining the next node of the current node; if it is determined that the next trajectory point of the first trajectory point corresponds to the next node, then it is determined that there are no missing trajectory points between the first trajectory point and the next trajectory point; If the next trajectory point does not correspond to the next node, then the next node is used as the starting point and the point corresponding to the next trajectory point is used as the ending point. The shortest path is searched in the weighted directed graph, and all nodes in the shortest path are identified as missing trajectory points. Based on the missing trajectory points and the regional path table, a complete regional path table is obtained. The regional traffic flow is calculated based on the complete regional path table. Based on the positional relationship between the current trajectory point and the next trajectory point in the complete regional path table, the direction of traffic flow is determined, thereby obtaining the traffic flow turning ratio for each turn at each intersection. The number of turning lanes is determined based on the complete regional path table, and the turning capacity for each turn at each intersection is determined based on the number of turning lanes and the saturation flow rate.

[0034] In some embodiments, data acquisition and processing include the following steps: (1) Data Acquisition. Obtain the checkpoint equipment information table of the target area road network. The main contents include checkpoint equipment name, checkpoint equipment number, and checkpoint equipment status. This is a dynamic table and will be updated as the road network or equipment changes. Obtain the target area road network information based on the Open Street Map (OSM), assign numbers to each intersection, and adjust according to the channelization layout of the actual intersections to obtain the target area road network data. Obtain the vehicle passage record table using the traffic police platform interface. This table includes information such as the detected vehicle passage record time, license plate number, checkpoint equipment number, driving direction, lane number, vehicle speed, and vehicle type. This is a dynamic table and will increase as the number of detected vehicles increases over time.

[0035] It should be noted that the vehicle passage record data and other user data obtained in the embodiments of this application are all obtained with the user's consent, or actively submitted after the user's relevant instructions, or inevitably uploaded by the user when using the corresponding application through the client, webpage, etc. In the technical solution disclosed in this application, the collection, storage, use, processing, transmission, provision, disclosure, and application of user personal information all comply with the provisions of relevant laws and regulations, necessary confidentiality measures have been taken, and they do not violate public order and good morals.

[0036] (2) Checkpoint Equipment Information Table Processing. Based on the checkpoint equipment names in the checkpoint equipment information table, the intersection name, approach lane, and checkpoint number are identified and stored in the table. For example, "University Town South Road and Fuxing Road Westbound Electronic Police 1" is identified as "University Town South Road and Fuxing Road" intersection - "West" approach - "1" checkpoint. Finally, an updated checkpoint equipment information table is obtained, including one or more of the following: intersection name, approach lane, checkpoint number, and checkpoint equipment number. Checkpoint equipment information is obtained based on the updated checkpoint equipment information table. Checkpoint equipment information includes, but is not limited to, intersection name, approach lane, checkpoint number, and checkpoint equipment number.

[0037] (3) Vehicle passage record table processing. Due to detection errors, communication errors, and analysis requirements, the vehicle passage data in the vehicle passage record table needs to be cleaned. The cleaning steps include: 1) Read the vehicle record table and convert it into a DataFrame format. DataFrame is a core data structure in Pandas, similar to a two-dimensional table or a data table in a database.

[0038] 2) Filter useful fields, format timestamps, and observe data issues.

[0039] 3) Remove completely duplicate data.

[0040] 4) Remove duplicate data within a short time window (i.e., within a preset time period). For example, only the first data entry of the same vehicle detected at the same checkpoint within 5 minutes will be recorded.

[0041] 5) Remove data with missing key fields, such as missing license plate data.

[0042] 6) Sort by vehicle passage time from smallest to largest.

[0043] 7) Store the cleaned vehicle passage data. Each vehicle passage data includes one or more of the following: passage time, license plate number, checkpoint equipment number, lane number, etc.

[0044] The cleaned vehicle passage data is stored to obtain a vehicle passage record dataset.

[0045] In some embodiments, intersection traffic flow statistics based on vehicle passage record datasets include the following steps: (1) Establishing a mapping relationship between checkpoint equipment and traffic flow: First, establish a mapping relationship between intersection names and intersection numbers. Through the target area road network data and the updated checkpoint equipment information table, establish a mapping relationship table between checkpoint data lane numbers and traffic flow. Traffic flow includes straight-through traffic flow, left-turning traffic flow, right-turning traffic flow, etc. The table includes intersection name, intersection number, approach lane, checkpoint serial number, checkpoint equipment number, number of lanes, lane number, and traffic flow. It can be understood that this embodiment establishes a mapping relationship between checkpoint equipment and traffic flow through the mapping relationship table between checkpoint data lane numbers and traffic flow. It can also be understood that the target area road network data is associated with the updated checkpoint equipment information table through the intersection name, and then the actual lane number is combined to assign traffic flow turning attributes, finally forming a complete mapping relationship table between checkpoint data lane numbers and traffic flow.

[0046] (2) Calculate the lane flow of each lane in each intersection: Import the vehicle record dataset, group them according to the checkpoint device number and lane number, and then count the number of vehicles in each group. At the same time, through the checkpoint data lane number and traffic flow mapping relationship table, map the number of vehicles to the specific intersection name-entry lane-traffic flow to obtain the lane flow of each lane in each intersection. The turning ratio of traffic flow at each intersection can also be calculated based on this lane flow.

[0047] (3) Calculate the inflow of each entrance in each intersection: Collect the lane flow of each lane in each intersection according to the intersection name and the entrance lane to obtain the inflow of each entrance in each intersection. At the same time, the inflow is also the cross-sectional flow of the road segment.

[0048] In some embodiments, travel route identification based on vehicle pass record datasets includes the following steps: (1) Match and identify the path at the checkpoint and collect the path traffic. Figure 2 This is a schematic diagram illustrating the path recognition process of a traffic signal control method according to an exemplary embodiment of this application, as shown below. Figure 2 As shown, import the vehicle passage record dataset. Starting from the first row, iterate through each vehicle passage record, determining if it is the first trajectory point for that license plate number. If so, store the combination of the vehicle passage record time, checkpoint device number, and lane number as the trajectory point information for the vehicle's first path, and then iterate through the next row of vehicle passage data. If not, check again if the time difference with the previous trajectory point is greater than a preset duration (e.g., 30 minutes). If it does not exceed the preset duration, continue storing the current path; otherwise, store a new path, and then iterate through the next row of vehicle passage data. Repeat the above steps until the last row of vehicle passage data is processed, ultimately obtaining a dictionary-formatted vehicle trajectory dataset. A dictionary is a data structure used to store key-value pairs, where the key is the license plate number and the value is all the paths for that vehicle. Each path is a list composed of different "vehicle passage record time, checkpoint device number, and lane number", such as: "Carplate:[[(Timestamp ('2025-06-17 07:00:06'), '50019559051218100489', 1)], [(Timestamp ('2025-06-17 08:30:48'), '50019559051218100149', 2), (Timestamp ('2025-06-17 08:34:42'), '50019559051218100201', 1)]]".

[0049] The system iterates through the vehicle trajectory dataset, calculates and stores the travel time and traffic flow for every two trajectory points, and also calculates and stores the travel time and traffic flow for the entire path. Based on the checkpoint device number and lane number data, and using the mapping table between lane number and traffic flow in the checkpoint data, the original trajectory points are converted into strings of "intersection number-entry lane-traffic flow" as trajectory points (e.g., 101 North Straight). The system outputs the total traffic flow and the average travel time based on the quartile method for each path (the quartile method can be described as determining a reasonable range of data based on the quartiles, filtering out valid data, and then using it to calculate the mean). If a path contains only one trajectory point, the travel time is 0. Finally, the system obtains a regional path table, which includes the path, path traffic flow, and average path travel time.

[0050] (2) Reconstructing missing paths between checkpoints. Since electronic police checkpoints are sparsely and unevenly distributed in the regional road network, the distance between adjacent checkpoints is large and there are multiple feasible paths. Therefore, it is necessary to reconstruct the missing paths between checkpoints to obtain the complete vehicle travel path.

[0051] First, a road network topology G=(N,A) is constructed based on the road network data of the target area, where N is the set of intersections and A is the set of road segments. The data comes from OSM and is combined with actual channelization calibration. The specific steps include: 1) Based on SUMO simulation software, establish and calibrate the simulation environment of the region according to the static road network data of the target area.

[0052] 2) Construct a weighted directed graph G_SUMO using the road network file and construct a function to obtain the shortest path distance through the SUMO interface.

[0053] 3) Construct a weighted directed graph G of key nodes (i.e., key points) based on the static road network data of the target area and the traffic flow turning information at its intersections, ignoring nodes other than those at intersections. It can be understood that key nodes are those located at intersections.

[0054] 4) Extract the distance between nodes from the weighted directed graph G_SUMO as the weight of edge G to obtain the complete road network topology graph G.

[0055] Next, the missing road network paths are reconstructed based on the regional route table. The specific steps include: 1) Parse the string of each path trajectory point in the regional path table and iterate through each trajectory point of each path.

[0056] 2) Construct the next node acquisition function based on the road network topology diagram G, that is, determine the next node of the known trajectory point based on the trajectory point, the direction of the entrance lane and the direction of the traffic flow.

[0057] 3) Determine whether the next node is a trajectory point recorded by the checkpoint (i.e., a recorded trajectory point in the area path table). If it is, save this trajectory point. Otherwise, based on the distance weight of the edge, use Dijkstra's algorithm to find the shortest path between the next node and the trajectory point recorded by the next checkpoint, and save it as the missing path to be restored.

[0058] Continue the above steps until the last path has been processed, ultimately obtaining a complete regional path table. By sorting these paths, the path with the highest traffic volume or the greatest delay (the difference between actual travel time and free-flow travel time) can be directly identified; this is the critical path of the regional road network, providing guidance for subsequent road network management and serving as a key focus. (Reference) Figure 3 As shown, Figure 3 This is a schematic diagram illustrating the path restoration process of a traffic signal control method according to an exemplary embodiment of this application.

[0059] In addition, by taking the first trajectory point of each path in the complete regional path table as the starting point O and the last trajectory point as the ending point D, the OD flow (i.e., regional flow) of the region can be obtained.

[0060] In some embodiments of this application, the process of road network traffic estimation and visualization includes the following steps: Based on a complete regional route table, the traffic flow between each pair of trajectory points, i.e., the road segment traffic flow between two intersections, can be statistically analyzed and the road network traffic flow can be visualized. A route map is drawn, and the traffic flow of each road segment in the road network is visually displayed using different line colors, for example, from green to red, indicating that the traffic flow increases. In addition, based on the positional relationship between the current trajectory point intersection and the next trajectory point intersection in the path, the turning direction of the traffic flow in the path can be determined, thereby statistically analyzing the traffic flow of each intersection without checkpoint equipment, and obtaining the traffic flow turning ratio of each turn at each intersection in the regional road network, which can provide more refined data for traffic signal control.

[0061] In step S120, the turning flow rate of each turn at each intersection is calculated based on the inlet flow rate and the turning ratio of the traffic flow; the real-time saturation of each turn at each intersection is calculated based on the turning flow rate and the turning capacity.

[0062] It is understandable that an intersection can have one or more lanes, and each lane corresponds to one or more turns. In other words, an intersection can have one or more turns. This is just an example for the sake of understanding, referring to the actual road layout.

[0063] In some embodiments, the turning flow rate of each turn at each intersection is obtained based on the complete regional route table. Among them, turning flow Import flow Traffic flow turning ratio The product is determined, that is Among them, import flow It is the traffic flow entering intersection i from road segment a within time period t, obtained from the vehicle passage records at the checkpoint; traffic flow turning ratio. The proportion of vehicles that turn from import j to exit j′ is obtained statistically by the path reconstruction algorithm and satisfies the following conditions. .

[0064] In some embodiments, the real-time saturation of each turn at each intersection ,in, To redirect traffic, The turning capacity of each turn at each intersection. Number of turning lanes Determined by the saturation flow rate s: ; The number of lanes for a specific turn is given by data from channelization drawings or on-site measurements. s is the saturation flow rate of a single lane, determined with reference to traffic engineering standards (e.g., 1800 vehicles / hour / lane is commonly used for urban arterial roads).

[0065] In step S130, the average network saturation is calculated based on the real-time saturation, and the target control model is determined based on the comparison between the average network saturation and the preset saturation.

[0066] In some embodiments, the process of adaptive signal control optimization based on multi-scale traffic conditions can be referred to Figure 4 , Figure 4This is an exemplary embodiment of the adaptive traffic signal control framework diagram of the traffic signal control method shown in this application. Adaptive traffic signal control is based on the multi-scale traffic state calculated in the aforementioned steps. It dynamically extracts and establishes flow rate, saturation, queuing, and network indicators that can be used for real-time decision-making through control parameters. Subsequently, it determines whether the control strategy needs to be updated based on a mechanism combining event-triggered and periodic-triggered mechanisms. Based on the network congestion level (i.e., average network saturation), it selects a target control model (e.g., a maximum pressure control model or a Webster delay minimization model) for timing optimization to generate a signal timing scheme, which is then sent to the signal controller for execution. The control effect after execution is evaluated using indicators such as delay, balance, and load distribution on key road segments, and fed back for the next round of strategy updates, thus forming a closed-loop adaptive control.

[0067] In this embodiment, the process of dynamically extracting control parameters includes the following steps: Phase structure generation. For each intersection i∈N, where N is the set of intersections, the conflict matrix is ​​extracted based on the lane number and traffic flow mapping table established in the previous steps: Equation (1) in, It is the conflict matrix (or conflict set) of intersection i. From the import channel Drive towards the exit lane Traffic flow (i.e., a specific turning action, such as a left turn at the north entrance or a straight turn at the east entrance). From the import channel Drive towards the exit lane Traffic flow, This indicates a turning pair (conflict pair), meaning that if two traffic flows receive a green light at the same time, their paths will intersect (conflict point) inside the intersection, and therefore cannot be placed in the same signal phase.

[0068] Generate a feasible phase set Each phase It is a set of conflict-free redirects.

[0069] The following content is provided for the calculation of key parameters: Real-time saturation: Equation (2) Cumulative queue length: Equation (3) Path impedance: ,in Equation (4) in, The real-time saturation of each turn at each intersection is defined as the ratio of turning flow to turning capacity, and is used to measure the degree of congestion. This represents the turning flow rate at each turn within each intersection; The turning capacity of each turn at each intersection; The cumulative queue length at intersection i and entrance lane j at time t is calculated on a rolling basis based on the difference between historical queue length and inbound / outbound flow. For the previous moment Queue length, It is the update step size (e.g., 5 seconds, 1 signal cycle, etc.); For the time period Within, the arrival flow of vehicles at intersection i and entrance lane j; Let J be the saturated outbound flow rate of the inbound lane, which represents the maximum number of vehicles (i.e., traffic capacity) that can be allowed to pass through the lane if it is always green. Let represent the percentage of green light time (green light ratio) at intersection i at time t. This refers to the green light duration (effective green light time) corresponding to this entrance lane. The signal period; Indicates in The actual number of vehicles that were able to exit entrance lane j during the specified time period; The travel impedance of path p is determined by the travel time of each segment along that path. Accumulated; Given the real-time travel time of road segment a, the average travel time of vehicles passing through this road segment is calculated by matching vehicle passage records at checkpoints. The free-flow travel time of road segment a is determined by the ratio of the segment length to the design speed. This represents the real-time traffic flow of road segment a during time period t. The traffic capacity of road segment a is a static parameter. and These are parameters for the BPR function, used to describe the nonlinear characteristics of travel time as a function of traffic flow. They are typically set based on empirical values ​​from traffic engineering.

[0070] In one embodiment of this application, before generating a signal timing scheme according to the control strategy corresponding to the target control model, the method further includes: obtaining the first segment traffic flow of the target road segment in a first preset time period and the second segment traffic flow in a second preset time period, wherein the first preset time period and the second preset time period differ by a preset duration; calculating the segment traffic flow change rate based on the first segment traffic flow and the second segment traffic flow; if the segment traffic flow change rate is greater than a preset traffic flow change threshold, or if a mandatory update cycle is reached, then regenerating the signal timing scheme according to the control strategy corresponding to the target control model.

[0071] In this embodiment, the adaptive strategy execution mechanism adopts an adaptive control strategy that combines event-driven and period-driven approaches, dynamically adjusting signal control parameters according to real-time traffic conditions. The specific execution mechanism is as follows: Taking road segment a as the target road segment as an example, the control policy update is triggered when any of the following conditions are met: Equation (5) Equation (6) in, Let q be the rate of change of traffic flow in road segment a during time period t (i.e., the first preset time period), describing the sensitivity index to traffic flow fluctuations; a (t) represents the real-time traffic flow of road segment a during time period t; For road segment a during the time period Traffic flow on the road segment (i.e., the second preset time period) The preset duration can also be understood as the second preset time period differing from the first preset time period by a preset duration T; η is a preset flow rate change threshold, used to determine whether the control strategy needs to be updated immediately; T update This is a mandatory update cycle, representing the maximum validity period of the control strategy in the absence of sudden traffic changes.

[0072] In this embodiment, the average network saturation can be calculated using the following formula: Equation (7) in, The average network saturation reflects the overall network congestion level of the target area's road network. Turning at the intersection Real-time saturation, |I i | represents the number of approach lanes at intersection i, and N represents the set of intersections; , Let i be the set of approach lanes and exit lanes, describing the physical composition of the intersection. The data comes from the channelization design drawing.

[0073] Taking a preset traffic change threshold of 0.85 as an example, based on the average network saturation... Dynamically select control strategy: when When the network congestion level is >0.85, the maximum pressure control model is adopted to achieve network-level optimization through path-level traffic information, and priority is given to relieving pressure on critical paths.

[0074] when When the network latency is ≤0.85 (network smooth / slow-moving state), the Webster delay minimization model is adopted to achieve local optimization at the intersection level and reduce vehicle delay.

[0075] It is understood that the preset traffic change threshold of 0.85 in this embodiment is only for illustrative purposes. In actual applications, the value of the preset traffic change threshold can be set according to the needs, and this application does not impose any restrictions on it.

[0076] In one embodiment of this application, a signal timing scheme is generated according to the control strategy corresponding to the target control model, including: if the target control model is a maximum pressure control model, determining the pressure weight of each phase based on the complete regional path table; allocating the green light duration according to the preset objective function, pressure weight and first preset constraint of the maximum pressure control model to generate the signal timing scheme.

[0077] In this embodiment, if the target control model is a maximum pressure control model (Max-Pressure), the expression of its preset objective function is as follows: Equation (8) in, The pressure weight for the k-th phase at intersection i reflects the traffic priority of that phase. The green light duration for the k-th phase at intersection i is dynamically calculated by the control model based on real-time traffic demand. To represent the k-th phase of intersection i, it can be understood that a phase refers to a set of conflict-free turning traffic flows that simultaneously obtain the right-of-way at a signalized intersection. This represents the set of feasible phases; for explanations of other parameters, please refer to the corresponding explanations above, and they will not be repeated here.

[0078] Among them, the pressure weight of each phase in each intersection Determined by the complete region path table: Equation (9) in, is the pressure weight of the k-th phase at intersection i, reflecting the traffic priority of that phase; p is the target path; P is the set of all possible OD paths in the area, generated by the path identification and reconstruction algorithm; This represents the real-time traffic flow of path p during time period t, i.e., the path traffic flow of the target path. It reflects the traffic distribution at the meso-level and is obtained through statistics. For explanations of other parameters, please refer to the corresponding explanations above, which will not be repeated here.

[0079] The first preset constraint condition is shown in the following expression: Equation (10) in, Let be the number of phases at intersection i, determined by the conflict matrix and the feasible phase generation algorithm, derived from the feasible phase set. Size; Let be the green light duration for the k-th phase at intersection i, which is dynamically calculated by the control model based on real-time traffic demand. The time loss for the k-th phase of intersection i includes start-up loss and clearing loss, and the reference standard is usually 3-5 seconds; Let be the signal period of intersection i, which is dynamically calculated by the optimization model and satisfies the minimum and maximum period constraints; N is the set of intersections. The minimum green light time for safety is a mandatory constraint set based on pedestrian crossing safety or vehicle clearance requirements; The upper and lower limits are set according to the intersection level and management requirements to constrain the signal cycle. Ψ is the signal period of intersection j; Ψ is the set of coordinated intersection pairs, defining adjacent intersections that need to be coordinated in phase difference or synchronized in period; δ is the period coordination tolerance coefficient, which limits the maximum deviation ratio of the periods between coordinated intersections; other parameters are explained in the corresponding explanations above, and will not be repeated here.

[0080] In some embodiments, the maximum pressure control model can be solved through the following steps: 1. Extract the turning flow matrix: Obtain the turning flow of each turn at each intersection based on the complete regional route table. .

[0081] 2. Calculate pressure weights: Calculate the pressure weight for each phase at each intersection based on path flow and topology. .

[0082] 3. Greedy phase selection: Select the feasible phase combination with the highest pressure weight.

[0083] 4. Green light duration allocation: Green light duration is allocated according to the pressure weight ratio, while also satisfying the minimum green light time constraint.

[0084] In one embodiment of this application, generating a signal timing scheme according to the control strategy corresponding to the target control model further includes: if the target control model is a Webster delay minimization model, calculating the signal cycle based on the total loss time and real-time saturation; allocating the green light duration based on the turning flow, traffic flow turning ratio, signal cycle, total loss time, and a second preset constraint condition to generate the signal timing scheme.

[0085] In this embodiment, if the target control model is the Webster delay minimization model, the signal period calculation based on flow rate can refer to the following formula: Equation (11) in, The optimal signal period for intersection i in time period t is the theoretically optimal signal period calculated based on the Webster delay minimization model, which aims to minimize the total delay of the intersection. For the total lost time, This represents the real-time saturation; explanations of other parameters are provided above and will not be repeated here. When the maximum saturation is greater than or equal to 0.9, the maximum period C is used directly. max .

[0086] Using statistical traffic flow turning ratio The green light duration is dynamically allocated, ensuring that the green light duration is not less than the minimum green light time. The green light duration allocation can be referenced as follows: Equation (12) in, Let be the duration of the green light for the k-th phase at intersection i; Turning at intersection i Turning flow; Let t be the signal period at intersection i during time period t; This represents the total lost time; other parameters are explained in the corresponding explanations above, and will not be repeated here.

[0087] Performance index calculation: Average delay: Equation (13) Total network delay: Equation (14) in, Let be the average delay of intersection i in time period t (i.e., the average waiting time for each vehicle). Let i be the vehicle arrival flow rate at intersection i, phase k, during time period t. Let be the total traffic flow at intersection i during time period t, where That is, the sum of all phase flows at intersection i during time period t; Let t be the signal period at intersection i during time period t; Let be the duration of the green light for the k-th phase at intersection i; Let i be the saturation (flow rate ratio) of phase k at intersection i during time period t. ,in This is the saturation flow rate for that phase (the sum of the saturation flow rates of each turning lane). Total network delay; The average delay of approach lane j at intersection i during time period t; This represents the approximate flow rate of approach lane j at intersection i during time period t, which usually refers to the actual arrival flow rate of that approach lane; other parameters are explained in the corresponding explanations above, and will not be repeated here.

[0088] Micro-level diagnosis and real-time optimization using average delays help adjust intersection-level signals; macro-level evaluation and decision support using total network delays help assess overall control effectiveness.

[0089] In step S140, a signal timing scheme is generated according to the control strategy corresponding to the target control model, and traffic signals are controlled based on the signal timing scheme.

[0090] In one embodiment, the signal timing scheme is sent to the traffic signal controller so that the traffic signal controller can execute the signal timing scheme.

[0091] In one embodiment of this application, after controlling traffic signals based on a signal timing scheme, the method further includes: calculating a path balance index based on path flow and average path flow; calculating a regional capacity improvement based on the capacity of a first region before implementing the signal timing scheme and the capacity of a second region before implementing the signal timing scheme; calculating a key road segment saturation variance based on the real-time saturation of key road segments and the average network saturation; and evaluating the control effect of controlling traffic signals based on the signal timing scheme based on one or more of the path balance index, regional capacity improvement, and key road segment saturation variance.

[0092] In this embodiment, the control effect evaluation index is calculated with reference to the following: Path equilibrium index: Equation (15) Improved regional traffic capacity: Equation (16) Variance of saturation in key road sections: Equation (17) in, The path balance index reflects the evenness of the distribution of OD traffic across multiple available paths; This represents the real-time traffic (i.e., path traffic) for path p. Average path traffic; To assess the difference in throughput capacity before and after the implementation of the control strategy, the amount of regional traffic capacity improvement is determined. The travel demand flow of OD pair (o,d) in time period t is obtained by path identification and OD estimation algorithm; The total OD flow after implementation (i.e., the first area's traffic capacity); The total OD flow (i.e., the second zone capacity) before implementation. The saturation variance of key road sections reflects the dispersion of road network load distribution; This represents the total number of road segments; This represents the real-time saturation of road segment a. This represents the average saturation of the network; other parameters are explained in the corresponding explanations above, and will not be repeated here.

[0093] The path balance index quantifies the balance of OD traffic distribution, and the variance of saturation in key road segments reflects the load balancing effect of the road network.

[0094] In one embodiment of this application, the execution flow of the traffic signal control method includes the following steps: 1. Obtain the latest multi-scale traffic status data, including the inbound traffic flow at each intersection, the turning ratio of traffic flow at each turn at each intersection, and the path flow.

[0095] 2. Calculate the turning flow rate for each turn at each intersection. .

[0096] 3. Calculate the real-time saturation of each turn at each intersection. .

[0097] 4. Calculate the average saturation of the network. .

[0098] 5. Select the appropriate control model based on the network congestion level.

[0099] 6. Solve the control model to generate a signal timing scheme to ensure that the green light duration is not less than the minimum green light time.

[0100] 7. Send the signal timing plan to the traffic signal controller for execution.

[0101] 8. Monitor the control effect in real time and provide feedback to the road network traffic visualization module.

[0102] This embodiment and the traffic signal control method provided in the above embodiments belong to the same concept. The specific way each step is executed has been described in detail in the foregoing embodiments, and will not be repeated here.

[0103] The control effect is fed back to the road network traffic visualization module in real time. The difference in traffic distribution before and after control is displayed intuitively through color changes (e.g., from green to red), forming a complete closed loop of "state perception - signal control - effect evaluation".

[0104] In other embodiments, a segment-related pressure method can be used, which calculates pressure weights using only the flow difference between adjacent segments, thereby significantly reducing computational complexity. Specifically, for intersections... The Each phase contains the turning traffic flow. Its pressure weight (equivalent to the aforementioned) This is the pressure weighting used for the table estimation. The expression can be simplified to the difference between the queue length of the upstream entrance lane and the available capacity of the downstream road segment. The upstream entrance lane refers to the entrance lane j of the current intersection i (i.e., upstream refers to the direction in which vehicles enter the intersection), and the downstream road segment refers to the road segment from intersection i along the turning lane j→j′ (i.e., the road segment corresponding to the exit lane j′, leading to the next intersection).

[0105] Equation (18) in, For the estimated pressure weights, For the intersection Import channel The cumulative queue length, The measured flow rate of the downstream road segment is determined by the inlet flow rate of each inlet at each intersection. For the traffic capacity of downstream road sections, This represents the maximum number of vehicles that can be stored in the downstream section. This is an adjustable weighting coefficient (a value of 0.6-0.8 is recommended).

[0106] In this embodiment, the method for determining pressure weight does not require traversing all OD paths. It only requires statistics on the inbound traffic flow at each intersection and calculations at each intersection. Import channel The cumulative queue length can be used to calculate the pressure weight. This method of calculating pressure weight is mainly used when it is difficult to obtain overall traffic flow data and topology deconstruction data. It aims to significantly improve calculation efficiency while ensuring control effect, especially for large road networks with a large number of intersections.

[0107] In other embodiments, a historical trend-weighted method can be used to predict the traffic flow turning ratio in the short term, achieving a more forward-looking green light allocation. First, using statistical historical turning ratio data (including traffic flow turning ratios), an exponentially weighted moving average (EWMA) method is employed to predict future... The turning ratio for each time period is given by the following formula: Equation (19) in, The predicted turning point ratio for future periods. This represents the actual steering ratio (i.e., traffic flow steering ratio) measured during the current time period. This is the average turning ratio for the same period in history. This is a smoothing coefficient (recommended value 0.3-0.5). Based on the predicted steering ratio, recalculate the predicted steering flow rate: Equation (20) in, Turning at intersection i Predicted shift traffic, Let be the inflow rate of traffic entering intersection i through road segment a within time period t. This represents the predicted shift ratio for future time periods.

[0108] Predicting turning traffic Substituting the green light duration allocation formula (i.e., formula (12)) from the aforementioned embodiments, and replacing the original real-time turning traffic flow... This approach, while preserving the complete mathematical form of the Webster model, effectively addresses scenarios with drastic fluctuations in steering ratio by introducing a lightweight prediction mechanism, thereby improving the smoothness of signal control. Compared to the solutions in the aforementioned embodiments, it only adds an exponential smoothing calculation step, maintains completely identical data requirements (both dependent on the vehicle flow steering ratio), and has extremely low engineering implementation costs.

[0109] In one embodiment of this application, a complete closed-loop architecture of "perception-balance-control" is constructed to achieve refined management and dynamic balance control of regional traffic flow. (Reference) Figure 5 , Figure 5 This is a flowchart illustrating a traffic signal control method according to another exemplary embodiment of this application, the core steps of which are as follows: S1. Multi-scale traffic condition perception. Vehicle passage records from the traffic police platform are obtained through an interface and processed to obtain a vehicle passage record dataset. A mapping relationship between checkpoint equipment and traffic flow is established, and the vehicle passage record dataset is analyzed to obtain lane flow and inlet flow at each intersection (micro-level perception). Based on the vehicle passage record dataset, vehicle travel paths are identified through matching and path flow is aggregated to obtain a regional path table. A road network topology is constructed, and missing paths are reconstructed from the regional path table to obtain a complete regional path table (meso-level perception). The complete regional path table is analyzed and processed to obtain road network flow and visualize it in charts. Simultaneously, regional OD flow is statistically analyzed (macro-level perception).

[0110] S2. Adaptive signal control optimization based on multi-scale traffic conditions. Based on the multi-scale traffic condition data obtained in step S1, control parameters are dynamically extracted, and an appropriate target control model (such as the maximum pressure control model or the Webster delay minimization model) is selected according to the network congestion level. A signal timing scheme is generated, and the control strategy is dynamically updated through a triggering mechanism, forming a complete closed loop of "state perception - signal control - effect evaluation".

[0111] To facilitate understanding of the overall implementation path of this application, as follows: Figure 5As shown, this application considers traffic state perception and traffic flow balancing as one set of integrated functional modules, and adaptive signal control and execution feedback as another set of integrated functional modules. The two sets of modules exchange information through key state variables such as path flow, OD demand, and road segment / network load, and form a feedback loop through control effect evaluation, thereby achieving dynamic balance and refined management of the regional road network.

[0112] It should be noted that the traffic signal control method provided in this embodiment is based on the same concept as the traffic signal control method provided in the above embodiments. The specific methods of each step, module and unit operation have been described in detail in the foregoing embodiments and will not be repeated here.

[0113] The regional traffic flow balance adaptive traffic signal control method based on multi-scale state perception provided in this application has the following significant advantages: First, the multi-scale state perception capability is significantly improved. Based on the vehicle passage record data at checkpoints, multi-scale traffic state information such as lane flow, vehicle travel paths, and origin-destination (OD) flow can be fully extracted from each lane at each intersection in the road network, achieving comprehensive perception from lane flow at intersections to the overall state of the regional road network. At the micro level, through refined mining of checkpoint data, real-time flow and turning information of intersection entrances and lanes can be obtained, providing a data foundation for precise signal control. At the meso level, through vehicle travel path identification and reconstruction, the driving trajectories and critical paths of vehicles within the region can be grasped, providing a basis for identifying road network bottlenecks. At the macro level, through OD flow estimation and road network visualization, the overall distribution characteristics of regional traffic flow can be grasped, providing support for regional traffic flow balance control.

[0114] Secondly, the adaptive control strategy achieves dynamic optimization. This invention employs a triggering mechanism combining event-driven and periodic-driven approaches to dynamically adjust signal control parameters based on real-time traffic conditions, thus solving the problems of lag in control parameter adjustment and poor system stability in related methods. Simultaneously, the control model selection strategy based on network congestion levels achieves an organic combination of network-level and intersection-level optimization, resulting in good control performance under various traffic conditions.

[0115] Furthermore, the regional traffic flow balancing effect has been significantly improved. Through network-level optimization of the maximum pressure control model, a balanced distribution of regional traffic flow has been achieved, effectively alleviating congestion on key paths and bottleneck sections. Simultaneously, multi-dimensional control effectiveness evaluation indicators allow for a comprehensive assessment of the control effect, providing a basis for continuous optimization of control strategies and forming a complete closed loop of "perception-balancing-control".

[0116] Furthermore, this application also boasts advantages such as high data utilization, low system maintenance costs, and strong robustness. By fully utilizing vehicle passage data from existing road checkpoints, it eliminates the need for a large number of new detectors, reducing system construction and maintenance costs. Simultaneously, the fusion and analysis of multi-source data enhances the system's robustness, maintaining good control performance even when some detectors fail.

[0117] Finally, the implementation of this application will significantly improve the level of urban traffic management, alleviate traffic congestion, enhance road network capacity, reduce vehicle delays and emissions, and have significant socio-economic benefits.

[0118] Figure 6 This is a block diagram illustrating a traffic signal control device according to an exemplary embodiment of this application. The device can be configured in a computer device or other devices; this embodiment does not limit the implementation environment to which the device is applicable.

[0119] like Figure 6 As shown, the exemplary traffic signal control device includes: a data acquisition module 610, a first calculation module 620, a second calculation module 630, a model determination module 640, and a signal control module 650.

[0120] The system includes: a data acquisition module 610 for acquiring multi-scale traffic state data, including the inlet flow rate of each entrance at each intersection, the turning ratio of each turn at each intersection, and the turning capacity; a first calculation module 620 for calculating the turning flow rate of each turn at each intersection based on the inlet flow rate and the turning ratio; a second calculation module 630 for calculating the real-time saturation of each turn at each intersection based on the turning flow rate and the turning capacity; a model determination module 640 for calculating the network average saturation based on the real-time saturation and determining the target control model based on the comparison between the network average saturation and the preset saturation; and a signal control module 650 for generating a signal timing scheme based on the control strategy corresponding to the target control model and controlling the traffic signals based on the signal timing scheme.

[0121] It should be noted that the traffic signal control device and the traffic signal control method provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the traffic signal control device provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0122] Embodiments of this application also provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the traffic signal control method as described in any of the above embodiments.

[0123] Figure 7 A schematic diagram of a computer system suitable for implementing the embodiments of this application is shown. It should be noted that... Figure 7 The computer system 700 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of the embodiments of this application.

[0124] like Figure 7 As shown, the computer system 700 includes a Central Processing Unit (CPU) 701, which can perform various appropriate actions and processes, such as executing the methods provided in the various embodiments described above, based on a program stored in Read-Only Memory (ROM) 702 or a program loaded from storage portion 708 into Random Access Memory (RAM) 703. The RAM 703 also stores various programs and data required for system operation. The CPU 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0125] The following components are connected to I / O interface 705: an input section 706 including a keyboard, mouse, etc.; an output section 707 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to I / O interface 705 as needed. A removable medium 711, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 710 as needed so that computer programs read from it can be installed into storage section 708 as needed.

[0126] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 709, and / or installed from removable medium 711. When the computer program is executed by central processing unit (CPU) 701, it performs various functions defined in the system of this application.

[0127] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0128] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0129] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0130] Another aspect of this application provides a computer-readable storage medium storing a computer program that is executed by a processor to implement the traffic signal control method as described in any of the above embodiments. This computer-readable storage medium may be included in the electronic device described in the above embodiments, or it may exist independently and not incorporated into the electronic device.

[0131] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0132] Another aspect of this application provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the traffic signal control methods provided in the various embodiments described above.

[0133] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the method according to the embodiments of this application.

[0134] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the embodiments disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0135] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A traffic signal control method, characterized in that, include: Acquire multi-scale traffic state data, which includes the inlet flow of each entrance at each intersection, the turning ratio of traffic flow at each turn at each intersection, and the turning capacity. The turning flow rate at each intersection is calculated based on the inlet flow rate and the turning ratio of the traffic flow. The real-time saturation of each turn at each intersection is calculated based on the turning flow rate and the turning capacity. The average network saturation is calculated based on the real-time saturation. The target control model is determined based on the comparison between the average network saturation and the preset saturation. Based on the control strategy corresponding to the target control model, a signal timing scheme is generated, and traffic signals are controlled based on the signal timing scheme.

2. The traffic signal control method according to claim 1, characterized in that, Acquire multi-scale traffic state data, including: Acquire vehicle passage data, as well as target area road network data and checkpoint equipment information, wherein the vehicle passage data includes checkpoint equipment number and lane number; Based on the road network data of the target area and the checkpoint equipment information, establish a mapping relationship between checkpoint equipment and traffic flow; The vehicles are grouped according to the checkpoint device number and the lane number, and the number of vehicles in each group is counted. Based on the mapping relationship, the number of vehicles in each group is mapped to each lane in each intersection to obtain the lane flow of each lane in each intersection. The lane flow is statistically analyzed according to the intersection name and the approach lane to obtain the approach flow of each approach lane at each intersection.

3. The traffic signal control method according to claim 2, characterized in that, After acquiring vehicle data, the method further includes: The vehicle passage data is traversed, grouped by license plate number, and the time difference between two adjacent vehicle passage data for the same vehicle is calculated. The vehicle passage data includes the license plate number. If the time difference is greater than the preset duration, then it is determined that the two adjacent vehicle passage data are not from the same path; If the time difference is less than or equal to the preset duration, then the two adjacent vehicle passage data are determined to be on the same path; The process continues until the vehicle data is traversed, vehicle trajectory data is obtained, and the path of each vehicle is determined, wherein the path consists of several trajectory points. Based on the checkpoint device number and the lane number, and using the mapping relationship, the representation of each trajectory point is transformed to obtain a path in a preset representation form; Traverse all vehicle paths, group paths with the same sequence of trajectory points into one category, count the number of occurrences of each category of paths, and obtain the path traffic. For each type of route, the travel time of all vehicles on the route is collected, outliers are removed using the quartile method, and the average value is calculated to obtain the average travel time of the route. A regional path table is obtained based on the path, the path traffic, and the average travel time of the path.

4. The traffic signal control method according to claim 3, characterized in that, After obtaining the regional path table, the method further includes: A regional road network is constructed based on the road network data of the target area, and a weighted directed graph is built. The nodes of the weighted directed graph are key points on the road, the edges of the weighted directed graph are roads, and the weight of the edge is the distance between two key points. The target path is determined from the regional path table, and each trajectory point of the target path is traversed starting from the first trajectory point; the current node corresponding to the first trajectory point is determined from the weighted directed graph, and the next node of the current node is determined; If it is determined that the next trajectory point of the first trajectory point corresponds to the next node, then it is determined that there is no missing trajectory point between the first trajectory point and the next trajectory point; if it is determined that the next trajectory point does not correspond to the next node, then the next node is used as the starting point and the point corresponding to the next trajectory point is used as the ending point to find the shortest path in the weighted directed graph, and all nodes in the shortest path are determined as missing trajectory points. Based on the missing trajectory points and the regional path table, a complete regional path table is obtained, and the regional traffic is statistically calculated based on the complete regional path table. Based on the positional relationship between the current trajectory point and the next trajectory point in the complete regional path table, the direction of traffic flow is determined, thereby obtaining the traffic flow turning ratio for each turn at each intersection; The number of turning lanes is determined based on the complete regional route table, and the turning capacity of each turn at each intersection is determined based on the number of turning lanes and the saturation flow rate.

5. The traffic signal control method according to any one of claims 1 to 4, characterized in that, Before generating a signal timing scheme based on the control strategy corresponding to the target control model, the method further includes: Obtain the traffic flow of the target road segment in the first road segment during a first preset time period and the traffic flow of the second road segment during a second preset time period, wherein the first preset time period and the second preset time period differ by a preset duration; The rate of change of traffic flow in the road segment is calculated based on the traffic flow in the first road segment and the traffic flow in the second road segment. If the traffic flow change rate of the road segment is greater than the preset traffic flow change threshold, or if the forced update cycle is reached, the signal timing scheme will be regenerated according to the control strategy corresponding to the target control model.

6. The traffic signal control method according to any one of claims 1 to 4, characterized in that, Based on the control strategy corresponding to the target control model, a signal timing scheme is generated, including: If the target control model is a maximum pressure control model, the pressure weight of each phase is determined based on the complete regional path table; The green light duration is allocated according to the preset objective function of the maximum pressure control model, the pressure weight, and the first preset constraint condition to generate the signal timing scheme.

7. The traffic signal control method according to any one of claims 1 to 4, characterized in that, Based on the control strategy corresponding to the target control model, a signal timing scheme is generated, which further includes: If the target control model is the Webster delay minimization model, the signal period is calculated based on the total loss time and the real-time saturation. The green light duration is allocated based on the turning flow rate, the traffic flow turning ratio, the signal cycle, the total lost time, and the second preset constraint to generate the signal timing scheme.

8. The traffic signal control method according to any one of claims 1 to 4, characterized in that, After controlling traffic signals based on the aforementioned signal timing scheme, the method further includes: The path balance index is calculated based on path flow and average path flow. Based on the first area traffic capacity before the implementation of the signal timing scheme and the second area traffic capacity before the implementation of the signal timing scheme, the area traffic capacity improvement is calculated. Based on the real-time saturation of key road segments and the average saturation of the network, the variance of saturation of key road segments is calculated. The control effect of traffic signal control based on the signal timing scheme is evaluated based on one or more of the path balance index, the regional capacity improvement, and the key road segment saturation variance.

9. A traffic signal control device, characterized in that, include: The data acquisition module is used to acquire multi-scale traffic state data, which includes the inlet flow of each entrance at each intersection, the turning ratio of traffic flow at each turn at each intersection, and the turning capacity. The first calculation module is used to calculate the turning flow of each turn at each intersection based on the inlet flow and the vehicle turning ratio. The second calculation module is used to calculate the real-time saturation of each turn at each intersection based on the turning flow and the turning capacity. The model determination module is used to calculate the average network saturation based on the real-time saturation, and determine the target control model based on the comparison result between the average network saturation and the preset saturation. The signal control module is used to generate a signal timing scheme based on the control strategy corresponding to the target control model, and to control traffic signals based on the signal timing scheme.

10. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the traffic signal control method as described in any one of claims 1 to 8.

Images