Method and system for dynamically adjusting green wave band of arterial traffic

Abstract

The invention relates to the technical field of traffic control, in particular to an arterial traffic green wave band dynamic adjustment method and system, and the method comprises the steps: constructing an arterial traffic data multi-source recognition network, and collecting and transmitting real-time arterial traffic data to a signal control center; constructing a single-intersection control scheme of each intersection on the trunk line based on the real-time trunk line traffic data; based on each single intersection control scheme, a trunk line coordination control scheme is constructed by taking a trunk line as an axis and taking districts as a network; according to the multi-source identification network, carrying out real-time judgment on emergencies, and according to the emergencies, constructing a predictive emergency scheduling scheme; according to multi-source real-time trunk traffic data, on the basis of actual traffic scene suitability, trunk green wave signals can be adaptively adjusted, intersection traffic flow changes are accurately matched, green wave control is effectively formed, temporary congestion scenes can be quickly responded through real-time emergency judgment and predictive scheduling, local congestion is prevented from being diffused to other intersections of a trunk, and the traffic safety of the trunk is improved. And the road passing efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of traffic control technology, and in particular to a method and system for dynamically adjusting green wave traffic on main roads. Background Technology

[0002] With the acceleration of urbanization, the growth rate of urban traffic flow and the trend of population concentration in cities are becoming increasingly significant, leading to increasingly serious urban traffic congestion problems. Traditional traffic signal control systems mostly adopt timed control mode, that is, to realize the flow and stopping of vehicles in different lanes at different intersections through fixed intervals and fixed direction lanes. However, the above methods are not very adaptable to real-time changes in traffic flow and cannot meet the current traffic conditions.

[0003] The proposed technologies are based on long-term historical data of the target intersection, and implement variable lane settings and green wave control to alleviate traffic congestion. However, while this approach is effective for regular traffic conditions, it cannot effectively respond to temporary emergencies such as traffic accidents or temporary road construction. It also cannot be adjusted according to the actual operation of each intersection on the main road, which can easily lead to local congestion and even affect the traffic flow at other main road intersections.

[0004] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the present invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0005] The technical problem to be solved by this invention is to provide a method and system for dynamic adjustment of green wave signals on main roads, which adapts to actual traffic scenarios and adaptively adjusts the green wave signals on main roads to improve road traffic efficiency.

[0006] To achieve the above objectives, the technical solution adopted by this invention is: a method for dynamically adjusting green wave traffic flow on main roads, comprising:

[0007] Construct a multi-source identification network for trunk line traffic data, and collect and transmit real-time trunk line traffic data to the signal control center;

[0008] Based on the real-time trunk line traffic data, a single-intersection control scheme is constructed for each intersection on the trunk line.

[0009] Based on each of the single-intersection control schemes, a trunk line coordinated control scheme is constructed with the trunk line as the axis and the area as the network.

[0010] The multi-source identification network is used to make real-time judgments on emergencies, and a predictive emergency dispatch plan is constructed based on the emergencies.

[0011] Furthermore, the construction of a multi-source identification network for trunk traffic data and the collection and transmission of real-time trunk traffic data to the signal control center include the following steps:

[0012] Real-time traffic data for each intersection is obtained sequentially;

[0013] The real-time traffic data of each intersection on the main road is compiled to form the real-time main road traffic data.

[0014] The real-time trunk traffic data is processed in a unified format and a timestamp is added.

[0015] Several sets of real-time trunk traffic data obtained at consecutive time intervals are projected into a simulated map space to form a dynamic correspondence.

[0016] Furthermore, the step of sequentially acquiring real-time intersection traffic data for each intersection includes the following steps:

[0017] Select any intersection on the main road and spatially divide the lane area and intersection area of ​​the current intersection;

[0018] Lane flow data is determined in real time by a first data acquisition terminal that is arranged at multiple intervals in the lane area.

[0019] Traffic data at the intersection is determined in real time by a second data acquisition terminal deployed in the intersection area.

[0020] The intersection floating data is determined by retrieving user movement information from the third-party operation platform in real time.

[0021] The lane flow data, intersection traffic data, and intersection floating data are verified and summarized to form the real-time intersection traffic data for the corresponding intersection.

[0022] Furthermore, the construction of a single-intersection control scheme for each intersection on the trunk line based on the real-time trunk line traffic data includes the following steps:

[0023] Based on the real-time traffic data at the intersection, obtain the real-time queue length at the intersection;

[0024] Set corresponding queue length warning thresholds based on the average lane queue length and the lane length at the intersection;

[0025] When the queue length warning threshold is reached, the first green light adjustment duration for the intersection lanes is determined based on the real-time queue length.

[0026] Set a special scenario threshold, and determine the second green light adjustment duration for the intersection lanes based on the special scenario threshold.

[0027] Furthermore, the step of setting a special scenario threshold and determining the second green light adjustment duration for the intersection lanes based on the special scenario threshold includes the following steps:

[0028] Identify the special locations and their distribution, the time distribution of special traffic flows, and the average proportion of traffic flow in each direction at each intersection related to the main road, and construct a list of special scenarios;

[0029] Based on the list of special scenarios, multiple scenario trigger thresholds are set;

[0030] When the scenario trigger threshold is reached, a relevant intersection is selected based on the scenario trigger threshold, and the average proportion P of traffic flow in each direction in the special traffic flow time distribution of the intersection is determined. average-time With real-time traffic flow percentage P real-timei The second green light adjustment duration K is obtained according to the following formula. m :

[0031] K m =k p ×(P real-time -P average-time )+k i ·P i

[0032] Where, k p k i Indicates parameter adjustment; P i This indicates adjustment of the cumulative deviation.

[0033] Furthermore, the construction of a trunk line coordinated control scheme based on each of the single-intersection control schemes, with the trunk line as the axis and the area as the network, includes the following steps:

[0034] Each of the single-intersection control schemes is projected into the simulated map space, and a trunk line spatiotemporal map model is constructed.

[0035] Based on the aforementioned trunk line spatiotemporal diagram model, traffic flow predictions for the trunk line intersection itself and adjacent intersections are obtained.

[0036] To minimize the total travel time in trunk line areas, an optimal flow timing model for each route is constructed.

[0037] Based on the traffic flow prediction of the trunk road intersection itself and adjacent intersections, the trunk road coordinated control scheme is obtained through the optimal traffic flow timing model of the path.

[0038] Furthermore, the step of obtaining traffic flow predictions for the trunk road intersection itself and adjacent intersections based on the trunk road spatiotemporal diagram model includes the following steps:

[0039] Based on the aforementioned trunk line spatiotemporal graph model, the trunk line spatiotemporal graph model is formed with road segment-intersection as nodes, average vehicle speed of road segment and traffic flow of intersection as node features, and vehicle travel time as marginal weight.

[0040] Construct a coupling matrix between adjacent intersections and the intersection itself to determine the traffic flow status of the next intersection:

[0041]

[0042] x j (t)=[q j1 (t),q j2 (t),...,q jn (t)] T n = 1, 2, ..., k

[0043] Where, x j (t+1) represents the traffic flow status at the (j+1)th intersection within the trunk line area, A j Let x be an autoregressive matrix, representing the time-series variation of traffic flow at the j-th intersection. j (t) represents the traffic flow status at the j-th intersection within the trunk line area, B j This represents the control matrix, indicating the impact of green light duration on traffic flow; u j (t) represents the single-intersection control scheme for the j-th intersection, N j C represents the set of adjacent intersections of the j-th intersection; jl Let q be the coupling matrix, representing the traffic flow impact of adjacent intersection l on intersection j; jn (t) represents the traffic flow at the j-th intersection and the n-th approach lane within the trunk line area.

[0044] Furthermore, the process of obtaining the trunk line coordinated control scheme based on the traffic flow prediction of the trunk line intersection itself and adjacent intersections, through the optimal traffic flow timing model, includes the following steps:

[0045] Determine the minimum total delay target within the region, and determine the green light synchronization constraints and intersection cycle coordination constraints for adjacent intersections;

[0046] Multiple selectable paths are identified within the trunk line area. The optimal traffic flow timing model is used to simultaneously optimize the traffic flow allocation ratio of each path and the timing parameters of each intersection, thereby solving for the optimal traffic flow allocation and optimal timing allocation for each path.

[0047] Based on the green light synchronization constraints and intersection cycle coordination constraints, the optimal traffic flow allocation and optimal timing allocation for each path are selected and integrated to form the trunk line coordination control scheme.

[0048] Furthermore, the step of making real-time determinations of emergencies based on the multi-source identification network and constructing predictive emergency dispatch schemes based on the emergencies includes the following steps:

[0049] The multi-source identification network is used to identify the features of sudden events to determine whether an abnormal event has occurred, and the current abnormal event is assigned a type based on the feature identification results.

[0050] Based on the real-time trunk traffic data of the abnormal event, the single-intersection control scheme of the associated intersection, and the congestion spread dataset of similar historical events, the level of congestion spread in the future time period is predicted.

[0051] Based on the type of the abnormal event and the level of congestion spread, a multi-level predictive emergency dispatch scheme is constructed.

[0052] This invention also provides a dynamic adjustment system for green wave traffic on main roads, comprising:

[0053] A multi-source identification network is used to collect and transmit real-time trunk traffic data;

[0054] A signal control center is used to receive the real-time trunk traffic data, analyze the real-time trunk traffic data, and output a corresponding signal control scheme, including:

[0055] The single-scheme processing module is used to construct a single-intersection control scheme for each intersection on the trunk line based on the real-time trunk line traffic data.

[0056] The coordination scheme processing module is used to construct a trunk line coordination control scheme based on each single-intersection control scheme, with the trunk line as the axis and the area as the network.

[0057] The emergency response processing module is used to make real-time judgments on emergencies based on the multi-source identification network and to construct predictive emergency dispatch plans based on the emergencies.

[0058] The beneficial effects of this invention are as follows: This invention acquires real-time trunk road traffic data through a multi-source identification network, and constructs individual control and coordinated control schemes for each intersection on the trunk road based on this data. It can adaptively adjust the green wave signal of the trunk road according to the adaptability of the actual traffic scenario based on the multi-source real-time trunk road traffic data, accurately match the changes in traffic flow at intersections, and construct a coordinated control scheme with the main trunk road as the axis and the area as the network, effectively forming green wave control. Furthermore, through real-time emergency event judgment and predictive scheduling, it can quickly respond to temporary congestion scenarios, prevent local congestion from spreading to other intersections on the trunk road, and improve road traffic efficiency. Attached Figure Description

[0059] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1 This is a flowchart illustrating the dynamic adjustment method of green wave traffic on trunk lines in an embodiment of the present invention.

[0061] Figure 2 This is a schematic diagram illustrating the process of collecting and transmitting real-time trunk traffic data in an embodiment of the present invention;

[0062] Figure 3 This is a schematic diagram of the process for obtaining real-time intersection traffic data in an embodiment of the present invention;

[0063] Figure 4 This is a flowchart illustrating the construction of a single-intersection control scheme in an embodiment of the present invention;

[0064] Figure 5 This is a schematic diagram of the process for determining the second green light adjustment duration in an embodiment of the present invention;

[0065] Figure 6 This is a schematic diagram of the process for constructing a trunk line coordinated control scheme in an embodiment of the present invention;

[0066] Figure 7 This is a schematic diagram of the process for obtaining trunk line intersection traffic prediction in an embodiment of the present invention;

[0067] Figure 8 This is a schematic diagram of the process for obtaining the trunk line coordination control scheme in an embodiment of the present invention;

[0068] Figure 9 This is a schematic diagram illustrating the process of constructing a predictive emergency dispatch scheme in an embodiment of the present invention;

[0069] Figure 10 This is a schematic diagram of the structure of the green wave dynamic adjustment system for trunk traffic in an embodiment of the present invention. Detailed Implementation

[0070] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0071] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly attached to the other element or there may be an intervening element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or there may be an intervening element. The terms "vertical," "horizontal," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementation.

[0072] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0073] like Figures 1 to 9 The method for dynamically adjusting the green wave traffic flow on main roads, as shown, includes the following steps:

[0074] Construct a multi-source identification network for trunk traffic data, and collect and transmit real-time trunk traffic data to the signal control center; collect real-time trunk traffic data, such as traffic flow, vehicle speed, and queue length, through multi-dimensional devices, and transmit them to the signal control center with low latency, providing accurate data support for subsequent single-intersection schemes and trunk coordination schemes;

[0075] Based on real-time trunk line traffic data, a single-intersection control scheme is constructed for each intersection on the trunk line; that is, for each intersection on the trunk line, its phase duration and phase sequence are optimized based on real-time data to ensure that the traffic efficiency of a single intersection is maximized and to avoid the impact of congestion at a single intersection on the normal operation of the entire trunk line.

[0076] Based on the control scheme for each individual intersection, a coordinated control scheme for trunk lines is constructed with the main road as the axis and the area as the network. This means integrating the individual intersection schemes on the trunk line into a collaborative scheme, optimizing the green wave traffic between intersections to achieve green wave traffic on the main road, while also taking into account the needs of non-main road intersections within the area and the different travel routes for the same destination, effectively forming diversion path guidance and avoiding the spread of congestion.

[0077] The system uses a multi-source identification network to make real-time judgments on emergencies and builds predictive emergency dispatch plans based on these emergencies. This allows for the rapid identification of abnormal events and adjustment of signal schemes, further predicting the path of congestion spread and preventing local events from escalating into overall trunk line congestion.

[0078] This invention acquires real-time arterial traffic data through a multi-source identification network and constructs individual and coordinated control schemes for each intersection on the arterial road based on this data. It can adaptively adjust the green wave signal of the arterial road according to the adaptability of the multi-source real-time arterial traffic data to the actual traffic scenario, accurately match the changes in traffic flow at intersections, and construct a coordinated control scheme with the main arterial road as the axis and the area as the network, effectively forming green wave control. Furthermore, through real-time emergency event judgment and predictive scheduling, it can quickly respond to temporary congestion scenarios, prevent local congestion from spreading to other intersections on the arterial road, and improve road traffic efficiency.

[0079] Based on the above embodiments, constructing a multi-source identification network for trunk traffic data and collecting and transmitting real-time trunk traffic data to the signal control center includes the following steps:

[0080] Real-time traffic data for each intersection is acquired sequentially; specifically, for each independent intersection on the main road, data from all sensing devices within that intersection is integrated to ensure that the traffic status of a single intersection is captured comprehensively.

[0081] Real-time traffic data from each intersection on the trunk line is aggregated to form real-time trunk line traffic data. The collected data from all intersections on the trunk line is centrally aggregated, so that the real-time trunk line traffic data can cover the entire trunk line, providing global data support for subsequent trunk line coordination and control. Specifically, data can be collected by aggregating data within edge node areas and by aggregating data at different trunk line levels. The transmission process can use fiber optic transmission combined with gateway reception, which can simultaneously receive data streams from multiple trunk line areas, ensuring that there is no data loss or delay during the aggregation process.

[0082] Real-time trunk traffic data is processed in a unified format and timestamps are added to achieve data standardization. The data is then transformed into associative standard data by adding timestamps to ensure the convenience of subsequent data retrieval. During the timestamp addition process, multiple additions can be used, such as collecting timestamps and receiving timestamps, to facilitate subsequent analysis of data transmission delays.

[0083] Several real-time arterial traffic data obtained at continuous time intervals are projected onto a simulated map space to form a dynamic correspondence. By spatializing and dynamizing continuous real-time data through the simulated map, signal control personnel can intuitively see changes in arterial traffic flow, while providing a data and spatial correlation basis for the construction of different subsequent schemes. Specifically, the processing generally involves steps such as simulated map construction, dynamic data projection rule construction, dynamic display of time series, and abnormal data prompts. After the construction is completed, the simulated map space can provide an API interface to access the signal control center, realizing dynamic control between subsequent single-intersection control schemes, arterial coordinated control schemes, predictive emergency dispatch schemes, and the simulated map.

[0084] Based on the above embodiments, obtaining real-time intersection traffic data for each intersection sequentially includes the following steps:

[0085] Select any intersection on the main road and spatially divide the lane area and intersection area of ​​the current intersection. Through clear area division, determine the deployment range and collection targets of different data collection devices to ensure that subsequent data can accurately correspond to the specific spatial location of the intersection. Specifically, the lane area can be set as the lane segment 50-100 meters behind the stop line of the approach lane, or it can be adjusted according to the number of lanes at the intersection. For 3 lanes or more in one direction, take 100 meters. The intersection area can be set as the area in front of the stop line of all approach lanes to the stop line of the exit lane within the intersection, that is, the core area of ​​vehicle crossing. In some special intersections, such as roundabouts and T-junctions, the lane area can be divided according to the direction of the approach lane + lane function, such as the left-turn lane and right-turn lane of the east approach lane of the roundabout. The intersection area is divided according to the main traffic flow direction, such as the inner circle traffic area and the outer circle merging area of ​​the roundabout, to avoid incomplete area coverage due to the special shape of the intersection.

[0086] Lane flow data is determined in real time by multiple data acquisition terminals arranged at intervals in the lane area. By deploying equipment at intervals in the lane area, data such as flow, queuing, and vehicle speed of each lane are acquired in real time, providing a basis for lane-level phase optimization in subsequent single-intersection control schemes. Specifically, the first data acquisition terminal can be set up as a combination of geomagnetic sensor and millimeter-wave radar, taking into account both data accuracy and environmental adaptability. For example, a geomagnetic sensor can be buried 5 meters, 20 meters, and 50 meters behind the stop line of each lane to form segmented perception. A millimeter-wave radar is installed on a roadside pole 30 meters behind the stop line of each lane. The pole is 6-8 meters high and the radar has a downward angle of 15°-20°, covering a range of 20-80 meters for that lane.

[0087] Real-time traffic data at intersections is determined by deploying a second data acquisition terminal in the intersection area. By covering the intersection area with equipment, data such as vehicle turning conflicts, pedestrian crossings, and intersection efficiency are acquired, avoiding the neglect of overall intersection safety and efficiency due to focusing solely on lane data. Specifically, the second data acquisition terminal can be configured as a combination of high-definition cameras and pedestrian detection sensors to achieve full-element perception of vehicles, pedestrians, and non-motorized vehicles in the intersection area. For example, one high-definition camera and one infrared beam sensor can be installed on each of the four corner poles at the intersection to identify vehicle turning behavior, the number of turning conflicts, and the intersection clearing time in real time. Simultaneously, the number of non-motorized vehicles is identified to avoid decreased traffic efficiency caused by non-motorized vehicle interference. The status of pedestrians entering / leaving the crosswalk is detected in real time, and data such as pedestrian crossing flow and crossing time are calculated.

[0088] By retrieving user movement information from third-party operating platforms in real time, the floating data of the intersection is determined; by retrieving user movement information from third-party platforms, such as vehicle GPS and mobile phone positioning, floating data such as traffic flow speed and travel time of the road sections around the intersection are obtained, which makes up for the limitation that fixed equipment can only perceive the inside of the intersection and more comprehensively reflects the impact of the intersection on the surrounding traffic.

[0089] Verify and summarize lane flow data, intersection traffic data, and intersection floating data to form real-time intersection traffic data for the corresponding intersections. Integrate the three types of data into a structured single-intersection real-time data packet to ensure data accuracy and completeness, which can be directly used for subsequent real-time trunk traffic data aggregation. During data aggregation and processing, data verification is required, such as internal consistency verification, external rationality verification, and abnormal data handling.

[0090] Based on the above embodiments, and using real-time trunk line traffic data, constructing a single-intersection control scheme for each intersection on the trunk line includes the following steps:

[0091] Based on real-time traffic data at intersections, the real-time queue length at intersections is obtained. The real-time queue length can directly reflect the current congestion level of the lanes, providing a quantitative benchmark for subsequent threshold settings and green light adjustments.

[0092] Set corresponding queue length warning thresholds based on the average lane queue length and the lane length at the intersection; by combining the lane length of the intersection itself and the real-time average queue length, set differentiated warning thresholds to avoid false triggering or missed triggering caused by uniform thresholds, such as triggering adjustment when the queue of a short lane is slightly longer or not adjusting when the queue of a long lane is too long.

[0093] When the queue length warning threshold is reached, the first green light adjustment duration for the intersection lanes is determined based on the real-time queue length. The first green light adjustment duration is calculated based on the difference between the real-time queue length and the warning threshold. This ensures that the increase in green light duration is positively correlated with the degree of queue congestion by obtaining the normal adjustment duration, thus alleviating congestion without wasting signal resources. Specifically, the first green light adjustment duration is constrained by the total cycle duration and the coordination constraint of the opposite lanes during the calculation to ensure the balance of intersection signals.

[0094] Set thresholds for special scenarios and determine the second green light adjustment duration for intersection lanes based on these thresholds. For special scenarios such as surrounding schools, hospitals, and shopping malls, set independent thresholds to trigger the second green light adjustment duration, thus achieving a closed loop of automatic scenario recognition and automatic signal scheme triggering. This solves the problem of poor adaptability to special scenarios and avoids safety hazards or traffic efficiency losses caused by only considering vehicle queuing.

[0095] Based on the above embodiments, setting a special scenario threshold and determining the second green light adjustment duration for intersection lanes based on the special scenario threshold includes the following steps:

[0096] Identify special locations related to trunk roads and their location distribution, special traffic flow time distribution, and the average proportion of traffic flow in each direction at each intersection to construct a list of special scenarios; specifically, samples of scenarios such as school dismissal, shopping mall promotions, and temporary accidents around the target intersection can be collected, with ≥500 samples for each scenario, covering sunny days, rainy days, daytime, and nighttime conditions;

[0097] Based on a list of special scenarios, multiple scenario trigger thresholds are set. The special scenario list is input into the learning model for training, and different trigger thresholds are set for different scenarios during recognition to ensure the accuracy of scenario recognition. For example, the trigger conditions for the school scenario are: time 16:30-17:00 + number of pedestrians in the main road area where the school gate is located ≥ 20.

[0098] When the scene trigger threshold is reached, the relevant intersection is selected based on the scene trigger threshold, and the average proportion P of traffic flow in each direction in the special traffic flow time distribution of the intersection is determined. average-time With real-time traffic flow percentage P real-timei The second green light adjustment duration K is obtained according to the following formula. m :

[0099] K m =k p ×(P real-time -P average-time )+k i ·P i

[0100] Where, k p k i Indicates parameter adjustment; P i This indicates adjustment of the cumulative deviation.

[0101] The above formula automatically calculates the second green light adjustment duration in special scenarios, ensuring a stable adjustment range and avoiding frequent fluctuations in traffic lights. At the same time, it enables the system to automatically adjust under special conditions, prioritizing the operation of special traffic flows at individual intersections.

[0102] Based on the above embodiments, and based on the control scheme for each single intersection, constructing a trunk line coordinated control scheme with the trunk line as the axis and the area as the network includes the following steps:

[0103] Each single-intersection control scheme is projected onto the simulated map space, and a trunk line spatiotemporal map model is constructed. The dispersed single-intersection control schemes are projected onto the simulated map to form a dynamic and visualized panoramic view of the trunk line signal, avoiding isolated schemes and providing a spatial and temporal data foundation for subsequent traffic prediction and timing optimization.

[0104] Based on the trunk line spatiotemporal diagram model, traffic flow predictions for the trunk line intersection itself and adjacent intersections are obtained; a traffic flow transmission relationship between the intersection itself and adjacent intersections is constructed, which can predict the traffic flow changes of the intersection itself and adjacent intersections within a set time period in the future, ensuring that subsequent timing plans can adapt to traffic flow trends in advance, rather than passively responding to the current state. Specifically, the prediction range can be set to the intersection itself, upstream intersections and downstream intersections according to different needs, and the prediction events can be set to short-term and long-term predictions, without specific limitations here;

[0105] With the goal of minimizing the total travel time in the trunk line area, an optimal traffic flow timing model is constructed. The optimal traffic flow timing model is used to quantify the relationship between signal timing and transit time, while incorporating constraints on trunk lines and areas, as well as traffic efficiency and safety, to ensure that the solution is scientific and controllable.

[0106] Based on traffic flow predictions at trunk road intersections and adjacent intersections, a trunk road coordination control scheme is obtained through a path-optimal traffic flow timing model. This ensures that the output trunk road coordination control scheme guarantees the smoothness of the green wave on the main trunk road while also taking into account the balance of traffic flow in the area, ultimately minimizing the total travel time in the trunk road area.

[0107] Based on the above embodiments, according to the trunk road spatiotemporal diagram model, the traffic flow prediction for the trunk road intersection itself and adjacent intersections is obtained, including the following steps:

[0108] Based on the trunk line spatiotemporal graph model, with road segment-intersection as nodes, average vehicle speed of road segment and traffic flow of intersection as node features, and vehicle travel time as marginal weight, a trunk line spatiotemporal graph model is formed; ensuring that the trunk line spatiotemporal graph model can accurately reflect the continuous movement process of traffic flow from road segment to intersection and then to the next road segment.

[0109] Construct a coupling matrix between adjacent intersections and the intersection itself to determine the traffic flow status of the next intersection:

[0110]

[0111] x j (t)=[q j1 (t),q j2 (t),...,q jn (t)] T n = 1, 2, ..., k

[0112] Where, x j (t+1) represents the traffic flow status at the (j+1)th intersection within the trunk line area, A j Let x be an autoregressive matrix, representing the time-series variation of traffic flow at the j-th intersection. j (t) represents the traffic flow status at the j-th intersection within the trunk line area, B jThis represents the control matrix, indicating the impact of green light duration on traffic flow; u j (t) represents the single-intersection control scheme for the j-th intersection, N j C represents the set of adjacent intersections of the j-th intersection; jl Let q be the coupling matrix, representing the traffic flow impact of adjacent intersection l on intersection j; jn (t) represents the traffic flow at the j-th intersection and the n-th approach lane within the trunk line area.

[0113] Through the above steps, the traffic flow prediction results were quantitatively calculated, taking into account both the physical constraints of the road network topology and the human intervention constraints of signal control, to ensure that the prediction results are more in line with the actual traffic scenario.

[0114] Based on the above embodiments, the trunk line coordinated control scheme is obtained through the optimal traffic timing model by predicting the traffic flow at the trunk line intersection itself and adjacent intersections, including the following steps:

[0115] Determine the minimum total delay target within the region, and determine the green light synchronization constraints and intersection cycle coordination constraints for adjacent intersections. The green light synchronization constraints for adjacent intersections are determined based on the travel time of vehicles between the two intersections to avoid vehicles arriving at the next intersection when the light is red. The intersection cycle coordination can be determined based on the signal light duration of the associated intersections to avoid timing disorder caused by excessive differences in the cycles of adjacent intersections.

[0116] Multiple selectable routes within the trunk line area are identified. The traffic allocation ratio of each route and the timing parameters of each intersection are simultaneously optimized using the optimal traffic flow timing model to obtain the optimal traffic flow allocation and optimal timing allocation for each route. When there are multiple routes in the area, such as from intersection A to intersection D, which can be reached via intersection B or intersection C, traffic flow allocation is needed to guide vehicles to choose the optimal route. At the same time, the timing of each intersection on the route is adjusted in a coordinated manner to achieve bidirectional optimization of route traffic efficiency and signal timing.

[0117] Based on green light synchronization constraints and intersection cycle coordination constraints, the optimal traffic flow allocation and optimal timing allocation for each path are selected and integrated to form a trunk line coordinated control scheme. That is, the optimization results of each path are verified through constraints, infeasible schemes are eliminated, feasible results are integrated, and a trunk line control scheme covering all paths and satisfying global coordination is formed.

[0118] Based on the above embodiments, the real-time determination of emergencies using a multi-source identification network and the construction of a predictive emergency dispatch plan based on the emergencies include the following steps:

[0119] The multi-source identification network is used to identify the features of emergencies to determine whether an abnormal event has occurred. Based on the feature identification results, the current abnormal event is classified into three categories: traffic disruption (such as accidents or debris), priority passage (such as emergency vehicle passage), and traffic flow change (such as temporary increase in traffic flow). This provides a classification basis for subsequent dispatching plans.

[0120] Based on real-time arterial traffic data of abnormal events, single-intersection control schemes for related intersections, and congestion diffusion datasets of similar historical events, the level of congestion diffusion in the future time period can be predicted. A predictive model can be built based on the congestion diffusion dataset of similar historical events to simulate traffic flow changes at different time points, outputting predicted indicators such as queue length at each intersection, average vehicle speed on road segments, and the number of intersections covered by congestion. Simultaneously, the simulation results are weighted and corrected by incorporating the diffusion speed of similar historical events to improve prediction accuracy. The level of congestion diffusion can include localized slight diffusion, localized moderate diffusion, localized diffusion on arterial roads, and comprehensive diffusion on arterial roads, which are set based on different predicted indicators and are not specifically limited here.

[0121] Based on the type of abnormal event and the level of congestion spread, a multi-level predictive emergency dispatch scheme is constructed. That is, for different types of abnormal events, a precise response to different types and degrees of impact of emergencies is achieved. A hierarchical dispatch scheme with graded response and precise dispatch is constructed to ensure rapid handling of events while minimizing the impact on regular traffic, maximizing the efficiency of trunk line traffic, and realizing rapid handling of emergencies on trunk lines.

[0122] like Figure 10 As shown, the present invention also provides a dynamic adjustment system for green wave traffic on main roads, comprising:

[0123] A multi-source identification network is used to collect and transmit real-time trunk traffic data;

[0124] The signal control center receives real-time arterial traffic data, analyzes the data, and outputs corresponding signal control schemes, including:

[0125] The single-scheme processing module is used to construct a single-intersection control scheme for each intersection on the trunk line based on real-time trunk line traffic data.

[0126] The coordination scheme processing module is used to construct a trunk line coordination control scheme based on each single intersection control scheme, with the trunk line as the axis and the area as the network.

[0127] The emergency response module is used to make real-time judgments on emergencies based on the multi-source identification network and to build predictive emergency dispatch plans based on the emergencies.

[0128] This invention acquires real-time arterial traffic data through a multi-source identification network and constructs individual and coordinated control schemes for each intersection on the arterial road based on this data. It can adaptively adjust the green wave signal of the arterial road according to the adaptability of the multi-source real-time arterial traffic data to the actual traffic scenario, accurately match the changes in traffic flow at intersections, and construct a coordinated control scheme with the main arterial road as the axis and the area as the network, effectively forming green wave control. Furthermore, through real-time emergency event judgment and predictive scheduling, it can quickly respond to temporary congestion scenarios, prevent local congestion from spreading to other intersections on the arterial road, and improve road traffic efficiency.

[0129] The specific working process of the above system has been explained in the above embodiments, and will not be repeated here.

[0130] Those skilled in the art should understand that this invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to this invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for dynamically adjusting green wave traffic flow on main roads, characterized in that, include: Construct a multi-source identification network for trunk line traffic data, and collect and transmit real-time trunk line traffic data to the signal control center; Based on the real-time trunk line traffic data, a single-intersection control scheme is constructed for each intersection on the trunk line. Based on each of the single-intersection control schemes, a trunk line coordinated control scheme is constructed with the trunk line as the axis and the area as the network. The multi-source identification network is used to make real-time judgments on emergencies, and a predictive emergency dispatch plan is constructed based on the emergencies.

2. The method for dynamic adjustment of green wave traffic on trunk lines according to claim 1, characterized in that, The construction of a multi-source identification network for trunk traffic data and the collection and transmission of real-time trunk traffic data to the signal control center include the following steps: Real-time traffic data for each intersection is obtained sequentially; The real-time traffic data of each intersection on the main road is aggregated to form the real-time main road traffic data. The real-time trunk traffic data is processed in a unified format and a timestamp is added. Several sets of real-time trunk traffic data obtained at consecutive time intervals are projected into a simulated map space to form a dynamic correspondence.

3. The method for dynamic adjustment of green wave traffic on trunk lines according to claim 2, characterized in that, The process of sequentially acquiring real-time traffic data for each intersection includes the following steps: Select any intersection on the main road and spatially divide the lane area and intersection area of ​​the current intersection; Lane flow data is determined in real time by a first data acquisition terminal that is arranged at multiple intervals in the lane area. Traffic data at the intersection is determined in real time by a second data acquisition terminal deployed in the intersection area. The intersection floating data is determined by retrieving user movement information from the third-party operation platform in real time. The lane flow data, intersection traffic data, and intersection floating data are verified and summarized to form the real-time intersection traffic data for the corresponding intersection.

4. The method for dynamic adjustment of green wave traffic on trunk lines according to claim 1, characterized in that, The process of constructing a single-intersection control scheme for each intersection on the trunk line based on the real-time trunk line traffic data includes the following steps: Based on the real-time traffic data at the intersection, obtain the real-time queue length at the intersection; Set corresponding queue length warning thresholds based on the average lane queue length and the lane length at the intersection; When the queue length warning threshold is reached, the first green light adjustment duration for the intersection lanes is determined based on the real-time queue length. Set a special scenario threshold, and determine the second green light adjustment duration for the intersection lanes based on the special scenario threshold.

5. The method for dynamic adjustment of green wave traffic on trunk lines according to claim 4, characterized in that, The process of setting a special scenario threshold and determining the second green light adjustment duration for the intersection lanes based on the special scenario threshold includes the following steps: Identify the special locations and their distribution, the time distribution of special traffic flows, and the average proportion of traffic flow in each direction at each intersection related to the main road, and construct a list of special scenarios; Based on the list of special scenarios, multiple scenario trigger thresholds are set; When the scenario trigger threshold is reached, a relevant intersection is selected based on the scenario trigger threshold, and the average proportion P of traffic flow in each direction in the special traffic flow time distribution of the intersection is determined. average-time With real-time traffic flow percentage P real-timei The second green light adjustment duration K is obtained according to the following formula. m : K m =k p ×(P real-time -P average-time )+k i ·P i Where, k p k i Indicates parameter adjustment; P i This indicates adjustment of the cumulative deviation.

6. The method for dynamic adjustment of green wave traffic on trunk lines according to claim 1, characterized in that, The construction of a trunk line coordinated control scheme based on each of the single-intersection control schemes, with the trunk line as the axis and the area as the network, includes the following steps: Each of the single-intersection control schemes is projected into the simulated map space, and a trunk line spatiotemporal map model is constructed. Based on the trunk line spatiotemporal diagram model, traffic flow predictions for the trunk line intersection itself and adjacent intersections are obtained; To minimize the total travel time in trunk line areas, an optimal flow timing model for each route is constructed. Based on the traffic flow prediction of the trunk road intersection itself and adjacent intersections, the trunk road coordinated control scheme is obtained through the optimal traffic flow timing model of the path.

7. The method for dynamic adjustment of green wave traffic on trunk lines according to claim 6, characterized in that, The step of obtaining traffic flow predictions for the trunk road intersection itself and adjacent intersections based on the trunk road spatiotemporal diagram model includes the following steps: Based on the aforementioned trunk line spatiotemporal graph model, the trunk line spatiotemporal graph model is formed with road segment-intersection as nodes, average vehicle speed of road segment and traffic flow of intersection as node features, and vehicle travel time as marginal weight. Construct a coupling matrix between adjacent intersections and the intersection itself to determine the traffic flow status of the next intersection: x j (t+1)=A j x j (t)+B j u j (t)+∑ l∈Nj C jl x l (t) x j (t)=[q j1 (t),q j2 (t),...,q jn (t)] T ,n=1,2,...,k Where, x j (t+1) represents the traffic flow status at the (j+1)th intersection within the trunk line area, A j Let x be an autoregressive matrix, representing the time-series variation of traffic flow at the j-th intersection. j (t) represents the traffic flow status at the j-th intersection within the trunk line area, B j This represents the control matrix, indicating the impact of green light duration on traffic flow; u j (t) represents the single-intersection control scheme for the j-th intersection, N j C represents the set of adjacent intersections of the j-th intersection; jl Let q be the coupling matrix, representing the traffic flow impact of adjacent intersection l on intersection j; jn (t) represents the traffic flow at the j-th intersection and the n-th approach lane within the trunk line area.

8. The method for dynamic adjustment of green wave traffic on trunk lines according to claim 7, characterized in that, The process of obtaining the trunk line coordinated control scheme based on the traffic flow prediction of the trunk line intersection itself and adjacent intersections, and through the optimal traffic flow timing model of the path, includes the following steps: Determine the minimum total delay target within the region, and determine the green light synchronization constraints and intersection cycle coordination constraints for adjacent intersections; Multiple selectable paths are identified within the trunk line area. The optimal traffic flow timing model is used to simultaneously optimize the traffic flow allocation ratio of each path and the timing parameters of each intersection, thereby solving for the optimal traffic flow allocation and optimal timing allocation for each path. Based on the green light synchronization constraints and intersection cycle coordination constraints, the optimal traffic flow allocation and optimal timing allocation for each path are selected and integrated to form the trunk line coordination control scheme.

9. The method for dynamic adjustment of green wave traffic on trunk lines according to claim 1, characterized in that, The step of making real-time determinations of emergencies based on the multi-source identification network and constructing predictive emergency dispatch schemes based on the emergencies includes the following steps: The multi-source identification network is used to identify the features of sudden events to determine whether an abnormal event has occurred, and the current abnormal event is assigned a type based on the feature identification results. Based on the real-time trunk traffic data of the abnormal event, the single-intersection control scheme of the associated intersection, and the congestion spread dataset of similar historical events, the level of congestion spread in the future time period is predicted. Based on the type of the abnormal event and the level of congestion spread, a multi-level predictive emergency dispatch scheme is constructed.

10. A dynamic adjustment system for green wave traffic on main roads, characterized in that, include: A multi-source identification network is used to collect and transmit real-time trunk traffic data; A signal control center is used to receive the real-time trunk traffic data, analyze the real-time trunk traffic data, and output a corresponding signal control scheme, including: The single-scheme processing module is used to construct a single-intersection control scheme for each intersection on the trunk line based on the real-time trunk line traffic data. The coordination scheme processing module is used to construct a trunk line coordination control scheme based on each single-intersection control scheme, with the trunk line as the axis and the area as the network. The emergency response processing module is used to make real-time judgments on emergencies based on the multi-source identification network and to construct predictive emergency dispatch plans based on the emergencies.

Images