Traffic signal adaptive control method based on thunder-vision integrated vehicle detector

Abstract

The invention discloses a traffic signal adaptive control method based on a thunder-vision integrated vehicle detector, and relates to the technical field of intelligent traffic, and the method comprises the steps: obtaining the time headway of vehicles in an entrance road section, the number of vehicles entering a target intersection, and the total number of vehicles entering the entrance road section; detecting the number of detained vehicles in each traffic flow direction of the entrance road section, and determining the saturation degree of each traffic flow direction; according to the saturation degree, the effective green light duration of the corresponding phase of each passing flow direction of the target intersection is adjusted, and when it is detected that a congestion flow direction exists, the additionally prolonged effective green light duration is added to the corresponding phase of the congestion flow direction, so that an updated signal timing scheme is obtained; and controlling the traffic lights of the target intersection according to the updated signal timing scheme in the next control period. The signal timing scheme is adjusted in real time according to the change condition of the road traffic flow state, adaptive control of the traffic signal lamp is achieved, and the road congestion problem is effectively relieved.

Description

Technical Field

[0001] The present application relates to the field of intelligent transportation technology, and in particular to a traffic signal adaptive control method based on a radar-visual integrated vehicle detector. Background Art

[0002] Urban road traffic congestion is the most concerned issue in the urban road traffic system, especially in the complex and changeable traffic environment at the intersections of urban roads, where intersection overflow and deadlock are very likely to occur, seriously affecting the overall operation and development of the city.

[0003] Urban road traffic signal control is an effective way to alleviate urban road traffic congestion. At present, urban road traffic signal control mainly relies on pre-designed signal timing logic. However, even if the segmented timing mechanism is introduced, due to the large changes in urban road traffic flow status, unreasonable signal timing often makes it difficult to maximize the utilization of traffic resources, seriously affecting the traffic capacity of road intersections. Summary of the invention

[0004] In view of the above problems and technical requirements, this application proposes a traffic signal adaptive control method based on a radar-visual integrated vehicle detector. The technical solution of this application is as follows:

[0005] A traffic signal adaptive control method based on a radar-visual integrated vehicle detector comprises the following steps:

[0006] The radar-visual integrated vehicle detector arranged at the downstream entrance of each entrance road segment of the target intersection obtains the headway of vehicles in each lane in the entrance road segment and the number of vehicles entering the target intersection at predetermined time intervals, and the radar-visual integrated vehicle detector arranged at the upstream exit of each entrance road segment of the target intersection obtains the total number of vehicles entering the entrance road segment at predetermined time intervals. x ;

[0007] According to the total number of vehicles entering each import road segment C x The number of vehicles entering the target intersection from each lane in the import road section is used to detect the number of stranded vehicles in each traffic flow direction of the import road section; the saturation of each traffic flow direction of the import road section is determined according to the headway in each lane in the import road section combined with the signal timing scheme of the current control cycle;

[0008] According to the saturation of each traffic flow direction of the target intersection, the effective green light duration allocated to the corresponding phase of each traffic flow direction in the basic timing plan of the target intersection is adjusted. The higher the saturation of the traffic flow direction, the longer the effective green light duration of the corresponding phase of the traffic flow direction; and when a congested flow direction is detected according to the number of stranded vehicles in each traffic flow direction of the target intersection, an additional extended effective green light duration is added to the corresponding phase of the congested flow direction to obtain an updated signal timing plan of the target intersection; wherein the congested flow direction is a traffic flow direction that reaches the first level of congestion;

[0009] In the next control cycle, the traffic light at the target intersection is controlled according to the updated signal timing plan of the target intersection.

[0010] A further technical solution is that the traffic signal adaptive control method further includes:

[0011] When it is detected that the congested flow direction reaches the second-level congestion level according to the number of stranded vehicles in each flow direction of the target intersection, an updated signal timing plan of the target intersection is obtained, and the signal timing plan of the upstream intersection connected to the entrance road section where the congested flow direction is located is updated, and the traffic lights of the respective intersections are controlled according to the updated signal timing plans of the target intersection and the upstream intersection in the next control cycle;

[0012] Among them, compared with the signal timing plan of the current control cycle, the updated signal timing plan of the upstream intersection reduces the effective green light duration of the corresponding phase of the traffic flow entering the entrance road section where the congested flow is located.

[0013] A further technical solution is to detect the congestion level according to the number of stranded vehicles in each traffic flow direction of the target intersection, including:

[0014] The number of stranded vehicles C of any traffic flow direction m in each entrance road segment connected to the target intersection y m The number of lanes y included according to the traffic flow direction m m Divide it evenly to get the average number of stranded vehicles in a single lane with traffic direction m The average number of stranded vehicles C in a single lane according to the traffic flow direction m mq Calculate the queue length of a single lane for traffic direction m;

[0015] The congestion degree of traffic direction m is determined according to the queue length of the single lane of traffic direction m. The longer the queue length of the single lane is, the higher the congestion degree is.

[0016] A further technical solution is to add an additional extension of the effective green light duration to the corresponding phase of the congested flow direction, including:

[0017] According to the congestion flow direction, the average number of stranded vehicles in a single lane in the current control cycle is Cmq and the average number of stranded vehicles in a single lane in the congestion flow direction in the previous control cycle C′ mq , determine the traffic flow fluctuation C of the congestion flow direction in the current control cycle relative to the previous control cycle f =C mq -C′ mq ;

[0018] According to the average headway time of vehicles in each lane included in the congestion flow direction in the previous control cycle, T LH Determine the effective green light duration T to be extended additionally for the phase corresponding to the congested flow direction m =C f ×T LH .

[0019] A further technical solution is to adjust the effective green light duration allocated to the corresponding phase of each traffic flow direction in the basic timing scheme of the target intersection according to the saturation of each traffic flow direction of the target intersection, including:

[0020] The effective green light duration of the phase corresponding to the traffic flow with the highest saturation assigned to the target intersection by the basic timing scheme is increased by T, and the effective green light duration of the phase corresponding to other traffic flow directions i assigned to the target intersection by the basic timing scheme is reduced. k is the total number of phases in the current control cycle, X1 is the highest saturation, X i is the saturation of the flow direction i.

[0021] A further technical solution is to increase the effective green light of the corresponding phase of the traffic flow with the highest saturation. X th is the saturation threshold.

[0022] A further technical solution is to determine the saturation of any traffic flow direction of the entrance road section entering the target intersection, including:

[0023] According to the headway time TH of the vehicles in any nth lane included in any traffic flow m in the import road segment n Calculate the wasted green light time W between two adjacent vehicles in the nth lane n =TH n -H n , H is the headway time of the standard lane;

[0024] According to the effective green light duration G of the corresponding phase of any traffic flow direction m in the import road segment in the signal timing scheme of the current control cycle and the wasted green light duration W between two adjacent vehicles in the nth lane included in the traffic flow direction m n , determine the saturation X of the nth lane in the traffic flow direction m n =(GW n) / G, the maximum saturation value of all lanes included in the traffic flow direction m is taken as the saturation of the traffic flow direction m.

[0025] A further technical solution is to detect the number of stranded vehicles in each traffic flow direction in each entrance road section connected to the target intersection, including:

[0026] According to the number of vehicles entering the target intersection in the lanes included in each traffic flow direction in the import road segment, the number of vehicles C entering the target intersection from any traffic flow direction m is determined. m ;

[0027] According to the total number of vehicles entering the import road section C x The number of remaining vehicles in the import road section is determined by the number of vehicles in each traffic flow direction entering the target intersection from the import road section L is the total number of traffic flows from the entrance road segment into the target intersection;

[0028] The number of remaining vehicles C in the import road section y Determine the number of stranded vehicles in any traffic direction m by proportional allocation

[0029] A further technical solution is that the traffic signal adaptive control method further includes:

[0030] The congestion degree of the congested flow direction is displayed through an inductive display screen arranged at a specified distance upstream of the stop line of the upstream intersection connected to the entrance road segment where the congested flow direction is located.

[0031] A further technical solution is that the traffic signal adaptive control method also includes: when it is detected that the congestion level of the congested flow direction in the previous control cycle is lower than the first level congestion level, reducing the additional extended effective green light duration added to the corresponding phase of the congested flow direction.

[0032] The beneficial technical effects of this application are:

[0033] The present application proposes a method for adaptive control of traffic signals based on a radar-visual integrated vehicle detector. The method detects the number of stranded vehicles and the saturation of each traffic flow direction of the entrance road section of the target intersection based on the measurement data of the radar-visual integrated vehicle detector, and adjusts the signal timing scheme of the target intersection in real time. The signal timing scheme is adjusted in time according to the changes in the real-time road traffic flow status, thereby realizing adaptive control of traffic lights and effectively alleviating the problem of road congestion.

[0034] By real-time detecting the number of stranded vehicles in each traffic flow direction of the import road section of the target intersection, the queue length of the import road section can be determined, which can accurately estimate the queue length on a longer section, making up for the defect that the existing sensing equipment can only detect the queue length at a close distance. It ensures that the congestion degree of each flow direction of the import road section can be discovered in time, and different adjustment strategies are set in a targeted manner for different congestion levels to automatically update the signal timing plan to timely divert traffic, avoiding overflow deadlocks and other situations at road intersections. In addition, the full process automatic control no longer requires manual intervention, reducing the investment of manpower and material resources, and significantly improving control efficiency.

[0035] By continuously detecting the congestion level at each intersection and setting up a guidance display screen at the upstream intersection of the congested section to display the congestion situation at the downstream intersection, it is possible to provide guidance before traffic overflow occurs, reduce the continuous influx of vehicles in that direction, and further speed up the resolution of the congestion problem. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a flow chart of the traffic signal adaptive control method.

[0037] Figure 2 It is a schematic diagram of the layout of the radar-visual integrated vehicle detector at the target intersection.

[0038] Figure 3 It is a phase distribution diagram of a control cycle.

[0039] Figure 4 This is a schematic diagram of the congestion level classification of import road sections.

[0040] Figure 5 This is a schematic diagram of the induction display screen layout. DETAILED DESCRIPTION

[0041] The specific implementation of the present application is further described below in conjunction with the accompanying drawings.

[0042] The present application proposes a method for adaptive traffic signal control based on a radar-visual integrated vehicle detector, which is applied to the traffic signal control at intersections of urban road traffic. A traffic signal controller is arranged at each intersection, and all traffic signal controllers are connected to the traffic signal control system. For each target intersection, its entrance road section represents the road section leading to the target intersection. Each entrance road section connects two adjacent upstream and downstream intersections. A radar-visual integrated vehicle detector is arranged at the downstream entrance of each entrance road section, and a radar-visual integrated vehicle detector is arranged at the exit of the upstream intersection of each entrance road section. Taking a cross intersection as an example, the target intersection contains four entrance lanes, and the arrangement positions of the radar-visual integrated vehicle detectors at each entrance lane are as follows: Figure 1 shown.

[0043] The traffic signal adaptive control method of this application is used to control the traffic signal controller in real time. Please refer to Figure 2 The flowchart shown in the figure has the following specific steps:

[0044] Step 1: Obtain the headway of vehicles in each lane of the entrance road segment and the number of vehicles entering the target intersection through the integrated radar and vision vehicle detector arranged at the downstream entrance of each entrance road segment of the target intersection at a predetermined time interval, and obtain the total number of vehicles C entering the entrance road segment through the integrated radar and vision vehicle detector arranged at the upstream exit of each entrance road segment of the target intersection at a predetermined time interval. x .

[0045] The integrated vehicle detector of the present application is arranged at a position 50 meters away from the exit of the upstream intersection of the import road section, and the vehicle flow data of the section 50 meters away from the exit of the upstream intersection can be obtained through the integrated vehicle detector of the radar and vision. Among them, the sampling time of the integrated vehicle detector of the radar and vision arranged at the downstream entrance of the import road section is consistent with the sampling time of the integrated vehicle detector of the radar and vision arranged at the upstream exit of the import road section. The sampling time interval can be customized according to the actual situation. Please measure the data every 10 seconds based on the actual traffic engineering experience.

[0046] Step 2: Based on the total number of vehicles C entering each import road segment x The number of vehicles entering the target intersection in each lane of the import road section is used to detect the number of stranded vehicles in each traffic flow direction of the import road section. The saturation of each traffic flow direction of the import road section is determined according to the headway time in each lane of the import road section combined with the signal timing scheme of the current control cycle.

[0047] Saturation refers to the ratio of the actual traffic flow of vehicles to the saturated capacity at a signalized intersection, which reflects the road service level. The traditional method calculates saturation based on the traffic flow of each lane, but this method has low inaccuracy in practical applications. Therefore, based on a large amount of practical engineering experience, this application proposes a method to calculate saturation based on headway and signal timing scheme. The specific process of calculating the saturation of each traffic flow direction of the entrance road section is as follows:

[0048] (1) Based on the headway time TH of vehicles in any nth lane included in any traffic flow direction m in the entrance road segment n Calculate the wasted green light time W between two adjacent vehicles in the nth lane n =TH n -H n , H is the headway time of the standard lane; the headway time indicates the time difference between the front ends of the two vehicles passing the same place, and this data can be directly obtained through the integrated radar and vision vehicle detector.

[0049] (2) Based on the effective green light duration G of the corresponding phase of any traffic flow direction m in the entrance road segment in the signal timing scheme of the current control cycle and the wasted green light duration W between two adjacent vehicles in the nth lane included in the traffic flow direction m, n , determine the saturation X of the nth lane in the traffic flow direction m n =(GW n ) / G, the maximum saturation value of all lanes included in the traffic flow direction m is taken as the saturation of the traffic flow direction m.

[0050] Because some vehicles in the lane fail to pass through the entrance road section on time due to some reasons, they are stranded in the entrance road section, which is an important cause of traffic congestion. Therefore, it is necessary to use the data obtained by the integrated vehicle detector of radar and vision to detect the number of stranded vehicles in each traffic flow direction in the entrance road section. In one embodiment, the specific method for detecting the number of stranded vehicles in each traffic flow direction in each entrance road section connected to the target intersection is:

[0051] (1) Determine the number of vehicles C entering the target intersection from any traffic direction m according to the number of vehicles entering the target intersection in the lanes included in each traffic direction in the entrance road segment. m .

[0052] (2) Based on the total number of vehicles entering the import road section C x The number of remaining vehicles in the import road section is determined by the number of vehicles in each traffic flow direction entering the target intersection from the import road section L is the total number of traffic flows from the entrance road segment into the target intersection.

[0053] (3) The number of remaining vehicles C in the import road section y Determine the number of stranded vehicles in any traffic direction m by proportional allocation

[0054] Taking a cross intersection as an example, the traffic flow entering the target intersection includes straight flow, left turn flow and right turn flow. The number of vehicles entering the target intersection in the straight flow within 10 seconds is determined as C s , the number of vehicles turning left and entering the target intersection is C l , the number of vehicles entering the target intersection in the right-turn direction is C r According to the total number of vehicles C entering the import road section within 10 seconds x Calculate the number of remaining vehicles C in the import road section y =C x -C l -C r -C s .

[0055] Determine the number of stranded vehicles in the left-turn direction based on the proportion of vehicles entering the target intersection in each traffic flow direction Number of stranded vehicles in the through flow direction Number of stranded vehicles in right-turn flow

[0056] Step 3: According to the saturation of each traffic flow direction of the target intersection, the effective green light duration allocated to the corresponding phase of each traffic flow direction in the basic timing plan of the target intersection is adjusted. The higher the saturation of the traffic flow direction, the longer the effective green light duration allocated to the corresponding phase of the traffic flow direction; and when a congested traffic flow direction is detected according to the number of stranded vehicles in each traffic flow direction of the target intersection, an additional extended effective green light duration is added to the corresponding phase of the congested traffic flow direction to obtain an updated signal timing plan for the target intersection. The congested traffic direction is a traffic flow direction that reaches the first level of congestion.

[0057] First, according to the saturation calculated in step 2, the effective green light duration of each phase corresponding to the traffic flow direction in the existing basic timing scheme is adjusted. The specific method is: the effective green light duration of the phase corresponding to the traffic flow direction with the highest saturation assigned to the basic timing scheme of the target intersection is increased by T, and the effective green light duration of the phase corresponding to other traffic flow directions i assigned to the basic timing scheme of the target intersection is reduced. k is the total number of phases in the current control cycle, X1 is the highest saturation, X i is the saturation of the flow direction i.

[0058] Among them, the time T added to the effective green light duration of the phase corresponding to the traffic flow with the highest saturation can be customized according to the actual traffic conditions, or calculated through the data collected by the sensing device. This application summarizes the time T added to the effective green light duration of the phase corresponding to the traffic flow with the highest saturation based on a large amount of actual traffic engineering experience. X th is the saturation threshold, which is set to 0.6 according to the conventional traffic saturation classification.

[0059] Take the four-phase and four-stage signal timing scheme of north-south straight, north-south left turn, east-west straight, and east-west left turn as an example. Figure 3 As shown in the figure, a control cycle includes four phases, each of which includes the effective green light duration and the green light interval duration. Each traffic flow direction corresponds to a phase, and the saturation of the four traffic flow directions calculated is sorted from large to small as X1, X2, X3 and X4. When the highest saturation X1 exceeds 0.6 and the saturation of the other three traffic flow directions does not exceed 0.6, the effective green light duration of the phase corresponding to the traffic flow direction with the highest saturation is extended. The effective green light duration of the phase corresponding to the traffic flow direction with saturation of X2 is shortened The effective green light duration of the phase corresponding to the traffic flow direction with saturation of X3 is shortened The effective green light duration of the phase corresponding to the traffic flow direction with saturation of X4 is shortened Among them, T=T2+T3+T4.

[0060] The above method can effectively optimize traffic flow while keeping the total duration of the basic timing plan unchanged.

[0061] At the same time, the signal timing scheme of the target intersection can also be adjusted according to the number of stranded vehicles in each traffic flow direction of the target intersection detected in real time. When the congestion level of the traffic flow direction is detected to exceed the first-level congestion level, the traffic flow direction is congested at this time, and the additional effective green light duration is added to the corresponding phase of the congested flow direction to obtain the updated signal timing scheme of the target intersection. The updated signal timing scheme has a total duration increase compared to the basic timing scheme, and the increased duration is the effective green light duration extended for the corresponding phase of the congested flow direction.

[0062] In one embodiment, adding an additional extension of the effective green light duration to the corresponding phase of the congested flow direction includes:

[0063] According to the congestion flow direction, the average number of stranded vehicles in a single lane in the current control cycle is C mq and the average number of stranded vehicles in a single lane in the congestion flow direction in the previous control cycle C′ mq , determine the traffic flow fluctuation C of the congestion flow direction in the current control cycle relative to the previous control cycle f =C mq -C′ mq ;

[0064] According to the average headway time of vehicles in each lane included in the congestion flow direction in the previous control cycle, T LH Determine the effective green light duration T to be extended additionally for the phase corresponding to the congested flow direction m =C f ×T LH .

[0065] In one embodiment, a specific method for detecting the congestion degree according to the number of stranded vehicles in each traffic flow direction of the target intersection is:

[0066] The number of stranded vehicles in any traffic flow direction m in each entrance road segment connected to the target intersection The number of lanes y included according to the traffic flow direction m m Divide it evenly to get the average number of stranded vehicles in a single lane with traffic direction m The average number of stranded vehicles C in a single lane according to the traffic flow direction m mqCalculate the queue length of a single lane for traffic direction m. This application calculates the queue length of a single lane by multiplying the average number of stranded vehicles in a single lane by the reserved distance between two vehicles. Determine the congestion level of traffic direction m based on the queue length of the single lane for traffic direction m. The longer the queue length of the single lane, the higher the congestion level.

[0067] Taking the three traffic flows at the intersection as an example, the number of stranded vehicles in the left-turn flow calculated in step 2 is Calculate the average number of stranded vehicles in a single lane with left-turn flow Then the queue length L of the single lane in the left-turn direction is calculated l =C lq ×5; according to the number of stranded vehicles in the straight flow direction Calculate the average number of stranded vehicles in a single lane in the straight flow direction Then the queue length L of the single lane in the straight flow direction is calculated s =C sq ×5; according to the number of stranded vehicles in the right-turn flow direction Calculate the average number of stranded vehicles in a single lane with right-turn flow Then calculate the queue length L of the single lane for right-turn flow r =C rq ×5;y l is the total number of lanes with left-turn flow, y r is the total number of lanes with right-turn flow, y s It is the total number of lanes for straight flow. The reserved distance between two vehicles is set to 5 meters based on the average vehicle length and the safe distance between two vehicles.

[0068] Since the maximum detection distance of the integrated vehicle detector is 200 meters, if the calculated queue length does not exceed 200 meters, the queue length detected by the integrated vehicle detector is used to calibrate the queue length calculation result to ensure the accuracy of the data.

[0069] Step 4: Control the traffic lights at the target intersection according to the updated signal timing plan of the target intersection in the next control cycle.

[0070] When the congestion level continues to increase and only adjusting the signal timing scheme of the target intersection can no longer alleviate the congestion, it is necessary to further adjust the signal timing scheme of the upstream intersection of the entrance road section of the target intersection. In another embodiment, when the congested flow direction is detected to reach the second-level congestion level according to the number of stranded vehicles in each traffic flow direction of the target intersection, the updated signal timing scheme of the target intersection is obtained, and the signal timing scheme of the upstream intersection connected to the entrance road section where the congested flow direction is located is updated, and the traffic lights of the respective intersections are controlled according to the updated signal timing schemes of the target intersection and the upstream intersection in the next control cycle; wherein, compared with the signal timing scheme of the current control cycle, the updated signal timing scheme of the upstream intersection reduces the effective green light duration of the corresponding phase of the traffic flow direction entering the entrance road section where the congested flow direction is located.

[0071] The congestion level can be set according to the total length R of the import road segment. This application sets three congestion thresholds R1, R2 and R3, where R1 = R-200, R2 = R-100, R3 = R-50. The congestion thresholds of the import road segment are divided as follows: Figure 4 shown. Figure 4 The downstream intersection in is the target intersection, and the road section from the upstream intersection to the downstream intersection is the current entrance road section.

[0072] When the queue length reaches the first-level congestion threshold R1, only the signal timing plan of the target intersection is adjusted; when the queue length reaches the second-level congestion threshold R2, while the target intersection is adjusted at the first-level congestion level, the effective green light duration of the three traffic flow object phases entering the entrance road section from the upstream intersection is shortened to the minimum green light duration. The minimum green light duration is determined according to the length of the zebra crossing for pedestrians and the speed of non-motor vehicles at the upstream intersection according to local traffic management related standards; when the queue length reaches the third-level congestion threshold R3, while the target intersection is adjusted at the first-level congestion level, the green light duration of the corresponding phases of the three traffic flow directions entering the entrance road section from the upstream intersection is shortened to 0, and the traffic entering the entrance road section from the upstream intersection is cut off and no more traffic is released until the queue length is less than the third-level congestion threshold.

[0073] When the congestion level gradually decreases, the adjusted signal timing plan needs to be adjusted back to the basic signal timing plan in time. In one embodiment, when it is detected that the congestion level of the congested flow direction in the previous control cycle is lower than the third or second level congestion level, the green light duration of the corresponding phases of the three traffic flow directions entering the entrance road section from the upstream intersection is increased, and the increased duration is the same as the previously reduced duration; when it is detected that the congestion level of the congested flow direction in the previous control cycle is lower than the first level congestion level, the additional extended effective green light duration added to the corresponding phase of the congested flow direction is reduced, and the reduced effective green light duration is calculated the same as the above-mentioned method for increasing the effective green light duration.

[0074] In order to further improve the efficiency of guiding the congestion degree of the entrance road section, the congestion degree of the congestion flow direction is displayed by setting up an induction display screen at a specified distance upstream of the stop line of the upstream intersection connected to the entrance road section where the congested flow direction is located. The layout position of the induction display screen is as follows: Figure 5 The specific distance of the induction display screen upstream of the stop line of the upstream intersection can be customized, for example, it can be set 200 meters upstream of the stop line of the upstream intersection.

[0075] The induction display screen displays the congestion of the road ahead, which can be distinguished by different colors. When the congestion level of the import road section is level 1, the road section is displayed in yellow, when the congestion level of the import road section is level 2, the road section is displayed in orange, and when the congestion level of the import road section is level 3, the road section is displayed in red. The induction display screen displays congestion information of at least 6 roads, with no less than 3 roads displayed horizontally and vertically. When the congestion level of a certain road section reaches level 3, the induction display screen prompts the driver heading to the section to re-plan the route.

[0076] The above is only a preferred embodiment of the present application, and the present application is not limited to the above embodiments. It is understood that other improvements and changes directly derived or associated by those skilled in the art without departing from the spirit and concept of the present application should be considered to be included in the protection scope of the present application.

Claims

1. A traffic signal adaptive control method based on a radar-visual integrated vehicle detector, characterized in that: The traffic signal adaptive control method comprises: The headway of vehicles in each lane in the entrance road segment and the number of vehicles entering the target intersection are obtained at predetermined intervals through the integrated radar and visual vehicle detector arranged at the downstream entrance of each entrance road segment of the target intersection, and the total number of vehicles entering the entrance road segment is obtained at predetermined intervals through the integrated radar and visual vehicle detector arranged at the upstream exit of each entrance road segment of the target intersection. x ; According to the total number of vehicles entering each import road segment C x and the number of vehicles entering the target intersection in each lane of the entrance road segment to detect the number of stranded vehicles in each traffic flow direction of the entrance road segment; determine the saturation of each traffic flow direction of the entrance road segment according to the headway in each lane of the entrance road segment combined with the signal timing scheme of the current control cycle; According to the saturation of each traffic flow direction of the target intersection, the effective green light duration allocated to the corresponding phase of each traffic flow direction in the basic timing plan of the target intersection is adjusted, and the higher the saturation of the traffic flow direction, the longer the effective green light duration of the corresponding phase of the traffic flow direction; and when a congested flow direction is detected according to the number of stranded vehicles in each traffic flow direction of the target intersection, an additional extended effective green light duration is added to the corresponding phase of the congested flow direction to obtain an updated signal timing plan of the target intersection; wherein the congested flow direction is a traffic flow direction that reaches the first level of congestion; In the next control cycle, the traffic light at the target intersection is controlled according to the updated signal timing scheme of the target intersection.

2. The traffic signal adaptive control method according to claim 1, characterized in that: The traffic signal adaptive control method further includes: When it is detected that the congested flow direction reaches the second-level congestion level according to the number of stranded vehicles in each traffic flow direction of the target intersection, an updated signal timing plan of the target intersection is obtained, and a signal timing plan of an upstream intersection connected to the entrance road section where the congested flow direction is located is updated, and in the next control cycle, the traffic lights of the respective intersections are controlled according to the updated signal timing plans of the target intersection and the upstream intersection; Among them, compared with the signal timing plan of the current control cycle, the updated signal timing plan of the upstream intersection reduces the effective green light duration of the corresponding phase of the traffic flow entering the entrance road section where the congested flow is located.

3. The traffic signal adaptive control method according to claim 2, characterized in that: Detecting the congestion degree according to the number of stranded vehicles in each traffic flow direction of the target intersection includes: The number of stranded vehicles C of any traffic flow direction m in each entrance road segment connected to the target intersection y m The number of lanes y included according to the traffic flow direction m m Divide it evenly to get the average number of stranded vehicles in a single lane with traffic direction m The average number of stranded vehicles C in a single lane according to the traffic flow direction m mq Calculate the queue length of a single lane for traffic direction m; The congestion degree of traffic direction m is determined according to the queue length of the single lane of traffic direction m. The longer the queue length of the single lane is, the higher the congestion degree is.

4. The traffic signal adaptive control method according to claim 3, characterized in that: The additional effective green light duration for the corresponding phase of the congested flow direction includes: According to the congestion flow direction, the average number of stranded vehicles in a single lane in the current control cycle is C mq and the average number of stranded vehicles C′ in the congested flow direction in the previous control cycle mq , determine the traffic flow fluctuation C of the congestion flow direction in the current control cycle relative to the previous control cycle f =C mq -C′ mq ; According to the average headway time of vehicles in each lane included in the congestion flow direction in the previous control cycle, T LH Determine the effective green light duration T to be extended additionally for the phase corresponding to the congested flow direction m =C f ×T LH .

5. The traffic signal adaptive control method according to claim 1, characterized in that: Adjusting the effective green light duration allocated to the corresponding phase of each traffic flow direction in the basic timing scheme of the target intersection according to the saturation of each traffic flow direction of the target intersection includes: The effective green light duration of the phase corresponding to the traffic flow with the highest saturation assigned to the target intersection by the basic timing scheme is increased by T, and the effective green light duration of the phase corresponding to other traffic flow directions i assigned to the target intersection by the basic timing scheme is reduced by k is the total number of phases in the current control cycle, X1 is the highest saturation, X i is the saturation of the flow direction i.

6. The traffic signal adaptive control method according to claim 5, characterized in that: The effective green light of the corresponding phase of the highest saturation traffic flow is increased by X th is the saturation threshold.

7. The traffic signal adaptive control method according to claim 1, characterized in that: Determine the saturation of any traffic flow direction of the entrance road segment into the target intersection including: According to the headway time TH of the vehicles in any nth lane included in any traffic flow m in the import road segment n Calculate the wasted green light time W between two adjacent vehicles in the nth lane n =TH n -H n , H is the headway time of the standard lane; According to the effective green light duration G of the corresponding phase of any traffic flow direction m in the import road segment in the signal timing scheme of the current control cycle and the wasted green light duration W between two adjacent vehicles in the nth lane included in the traffic flow direction m n , determine the saturation X of the nth lane in the traffic flow direction m n =(GW n ) / G, the maximum saturation value of all lanes included in the traffic flow direction m is taken as the saturation of the traffic flow direction m.

8. The traffic signal adaptive control method according to claim 1, characterized in that: Detecting the number of stranded vehicles in each traffic flow direction in each entrance road segment connected to the target intersection includes: The number of vehicles C entering the target intersection from any traffic flow direction m is determined according to the number of vehicles entering the target intersection from the lanes included in each traffic flow direction in the entrance road segment. m ; According to the total number of vehicles C entering the import road segment x The number of remaining vehicles in the import road section is determined by the number of vehicles in each traffic flow direction entering the target intersection of the import road section L is the total number of traffic flows from the entrance road segment into the target intersection; The number of remaining vehicles C in the import road section y Determine the number of stranded vehicles in any traffic direction m by proportional allocation 9. The traffic signal adaptive control method according to claim 1, characterized in that: The traffic signal adaptive control method further includes: The congestion degree of the congested flow direction is displayed through an inductive display screen arranged at a specified distance upstream of the stop line of the upstream intersection connected to the entrance road segment where the congested flow direction is located.

10. The traffic signal adaptive control method according to claim 1, characterized in that: The traffic signal adaptive control method further includes: When it is detected that the congestion level of the congested flow direction in the previous control cycle is lower than the first-level congestion level, the additional extended effective green light duration added to the corresponding phase of the congested flow direction is reduced.

Images