Intersection signal dynamic coordination control method and system based on real-time trajectory data

Abstract

The invention discloses an intersection signal dynamic coordination control method and system based on real-time trajectory data, which belongs to the technical field of urban traffic management and signal control and comprises three aspects of traffic state estimation, traffic state deduction and coordination control scheme optimization. Firstly, a traffic state is represented by road network cellular division and real-time trajectory data calibration cellular states. Secondly, decomposing the traffic flow into convection and diffusion motions by referring to molecular one-dimensional convection-diffusion motion, establishing a dynamic motorcade discrete model, and simulating traffic state evolution; and finally, designing a two-stage framework optimization coordination control scheme, in the first stage, calculating the public period duration and the green time ratio, and in the second stage, optimizing the phase difference to minimize the delay. According to the invention, traffic changes can be dynamically responded, signal timing is optimized, and the traffic efficiency of the intersection is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of urban traffic management and signal control, and more particularly to a method and system for dynamic coordination control of intersection signals based on real-time trajectory data. Background Art

[0002] With the acceleration of global urbanization and the improvement of residents' quality of life, urban transportation demand continues to rise, bringing unprecedented pressure to the existing urban transportation system. Especially during peak hours in the morning and evening, traffic congestion occurs frequently and commuting time is significantly extended, which not only affects the daily quality of citizens, but also brings a considerable burden to urban economic development. According to the "2021 Urban Mobility Scorecard", the losses caused by traffic congestion in the United States in 2020 reached 4.3 billion hours and 1.7 billion gallons of fuel, further highlighting the urgency of alleviating traffic congestion and improving commuting efficiency worldwide.

[0003] As a key node in the urban road network, the traffic efficiency of intersections is directly related to the smoothness of the entire road network. During periods of high traffic volume, reasonable traffic control measures are essential to ensure traffic safety and improve traffic efficiency. Among them, signal control, as one of the core means of urban traffic management, has a significant effect in alleviating intersection congestion, reducing the number of stops, and controlling queue overflow. However, traditional signal control methods are often based on fixed time scheduling schemes, which are difficult to adapt to real-time changing traffic needs, resulting in low traffic efficiency and increased congestion.

[0004] Therefore, how to provide a method and system for dynamic coordination control of intersection signals based on real-time trajectory data to achieve more efficient and intelligent traffic signal control, reduce traffic congestion, and improve the operating efficiency of urban transportation systems is an urgent problem that technical personnel in this field need to solve. Summary of the invention

[0005] In view of this, the present invention provides a method and system for dynamic coordinated control of intersection signals based on real-time trajectory data, which accurately estimates the traffic status through cell division and real-time trajectory data, simulates the convection and diffusion movement of traffic flow, and optimizes signal control schemes to reduce traffic congestion and delays.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] In one aspect, the present invention provides a method for dynamic coordinated control of intersection signals based on real-time trajectory data, comprising the following steps:

[0008] Divide the target road network into several types of cells, obtain real-time trajectory data, and estimate the cell states of the cells based on the real-time trajectory data;

[0009] Establishing a dynamic fleet discrete model, using the dynamic fleet discrete model to predict the traffic state evolution process under different signal timing schemes, obtaining prediction results, and updating the cell state according to the prediction results;

[0010] Calculate the public cycle duration and green-to-signal ratio based on the updated cell state;

[0011] According to the common cycle duration, the green-to-signal ratio and the prediction results, a phase difference optimization model is constructed with the goal of minimizing the total delay of the connected vehicles;

[0012] An optimized signal control scheme is output based on the phase difference optimization model.

[0013] Preferably, estimating the cell state of the cell based on the real-time trajectory data includes:

[0014] The sampled inventory of the cell is estimated based on the real-time trajectory data:

[0015]

[0016] Among them, Ω i (t) represents the set of connected vehicle serial numbers in cell i at time t, |Ω i (t)| represents the set Ω i The number of elements in (t), represents the set of all cells;

[0017] Estimate the sampling capacity of cells based on the maximum value of the sampled stock of each type of cells:

[0018]

[0019] Among them, a represents the entire road network area, represents the endpoint cell set;

[0020] Calculate the sampled inflow and sampled outflow of the cell respectively:

[0021]

[0022] in, represents the sampling inflow of cell i, Ω i( t) represents the set of vehicle numbers in cell i at the current moment, Ω i (t-1) represents the set of vehicle numbers in cell i at the previous moment, represents the sampled outflow of cell i, Ω i(t+1) represents the set of vehicle numbers in cell i at the next moment;

[0023] Estimate the cell capacity based on the maximum value of the sampled outflow of each type of cell:

[0024]

[0025] Preferably, a dynamic fleet discrete model is established, including:

[0026] Decompose the movement of the convective vehicle fleet between adjacent intersections into convection movement and diffusion movement;

[0027] The flow rate driven by the convective motion is expressed as:

[0028] q c (x, t) = v(t)k(x, t);

[0029] Among them, v(t) represents the average speed of the road section at time t; k(x,t) represents the density of the x position at time t;

[0030] According to Fick's first law, the flow rate driven by the diffusion motion is expressed as:

[0031]

[0032] Where D is the fleet diffusion coefficient;

[0033] The convective motion and the diffusion motion are superimposed to obtain a dynamic fleet discrete model:

[0034]

[0035] Preferably, the dynamic fleet discrete model is used to predict the traffic state evolution process under different signal timing schemes to obtain prediction results, including:

[0036] Based on the dynamic fleet discrete model, a conversion relationship between the distribution of traffic status in time and in space is constructed, and the conversion relationship is:

[0037]

[0038] Based on the conversion relationship, the estimated result of cell outflow is obtained:

[0039]

[0040] Preferably, calculating the common cycle duration and the green-to-signal ratio according to the updated cell state includes:

[0041] The flow rate of the lane group in the new cycle from path j is:

[0042]

[0043] in, is the latest time when the vehicle that has reached lane group m on the new periodic path j at time t0 enters the road, is the earliest time when the vehicle with the new cycle path j reaches lane group m enters the road, t0 is the time when the old cycle and the new cycle alternate, represents the set of source cells that path j passes through;

[0044] The flow rate of lane group m is calculated based on the flow rate of lane group from path j in the new cycle:

[0045]

[0046] Based on the flow rate of lane group m, calculate the critical flow rate ratio vc of the connected vehicles at each stage s for:

[0047]

[0048] Among them, CAP m Indicates the capacity of the lane group;

[0049] Calculate the cycle time based on the key flow rate ratio of the connected vehicle:

[0050]

[0051] Among them, Y n represents the sum of the critical flow rate ratios of intersection n; L is the total signal loss time;

[0052] The common cycle duration is the maximum value of all intersection cycle durations:

[0053]

[0054] After determining the common cycle length, the green signal ratio g of each stage at intersection n s Meeting the minimum green time It is determined according to the critical flow rate ratio of each stage under the constraints;

[0055]

[0056] Preferably, the phase difference optimization model aims to minimize the total delay of the new cycle:

[0057]

[0058] in, is the new cycle time set, They represent regular cells, shunt cells, inflow cells, and signal control cells respectively.

[0059] When the cycle length C n If the minimum green light time requirement is not met, the cycle time C n To make corrections:

[0060]

[0061] in, is the minimum green light duration in signal phase s.

[0062] On the other hand, the present invention provides a dynamic coordination control system for intersection signals based on real-time trajectory data, which is used to implement the above-mentioned dynamic coordination control method for intersection signals based on real-time trajectory data. The system comprises:

[0063] A state calculation module, used for dividing the target road network into a plurality of cell types, obtaining real-time trajectory data, and estimating the cell state of the cell based on the real-time trajectory data;

[0064] A prediction module is used to establish a dynamic fleet discrete model, use the dynamic fleet discrete model to predict the traffic state evolution process under different signal timing schemes, obtain prediction results, and update the cell state according to the prediction results;

[0065] A calculation module is used to calculate the common cycle duration and green-to-signal ratio according to the updated cell state;

[0066] An optimization module, for constructing a phase difference optimization model with the goal of minimizing the total delay of the connected vehicles according to the common cycle duration, the green-to-signal ratio and the prediction result;

[0067] An output module is used to output an optimized signal control solution based on the phase difference optimization model.

[0068] Through the above technical solutions, it can be known that compared with the prior art, the present invention discloses a method and system for dynamic coordination control of intersection signals based on real-time trajectory data, which adopts road network cell division technology, combined with real-time trajectory data, accurately calibrates and characterizes traffic status, and realizes refined estimation of traffic flow and congestion; secondly, drawing on the molecular convection-diffusion motion theory, the traffic flow is decomposed into convection and diffusion, and a dynamic fleet discrete model based on trajectory data is constructed. The cellular transmission model framework is improved to simulate the evolution of traffic status to predict the trend of traffic flow changes; finally, a two-stage framework is designed to optimize the traffic signal control scheme. The first stage calculates the common cycle duration and green signal ratio based on the predicted networked vehicle flow, and the second stage optimizes the phase difference with the goal of minimizing delays. Overall, the present invention aims to reduce traffic congestion and improve road traffic efficiency through refined traffic state estimation, deduction and signal control optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0070] Figure 1 The data flow diagram of the present invention is:

[0071] Figure 2 It is the main interface of the present invention;

[0072] Figure 3 This is the evaluation result diagram;

[0073] Figure 4 It is the algorithm operation process of the present invention;

[0074] Figure 5 It is the cell partitioning method;

[0075] Figure 6 It is a system framework diagram of the present invention. DETAILED DESCRIPTION

[0076] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0077] The embodiment of the present invention discloses a method for dynamic coordination control of intersection signals based on real-time trajectory data. Figure 1-2 As shown, the following steps are included:

[0078] S1: Divide the target road network into several types of cells, obtain real-time trajectory data, and estimate the cell state of the cells based on the real-time trajectory data;

[0079] The cellular transport model assumes that all vehicles travel at free flow speed. In the non-congested case, all vehicles in the upstream cell enter the downstream cell within one time step. In the case of congestion, the outflow of the cell within one time step is limited by the road capacity and the capacity of the downstream adjacent cells. Definition is the outflow of cell i in time step t, then It can be calculated by the following formula:

[0080]

[0081] Among them, n i (t) represents the number of vehicles in cell i at the beginning of time step t (hereinafter referred to as cell stock); Q i (t) represents the maximum number of vehicles that can flow out of cell i in time step t (hereinafter referred to as cell capacity); N i (t) represents the maximum number of vehicles that cell i can accommodate at time step t (hereinafter referred to as the maximum cell capacity). i+1 (t)-n i+1 (t) represents the residual capacity of the downstream cell at the beginning of time step t (hereinafter referred to as the downstream residual capacity). It should be noted that intuitively, the vehicles in cell i+1 in time step t will move forward, allowing more vehicles to enter cell i+1 from cell i. However, in the manual driving environment, the vehicles in cell i will not immediately rush into cell i+1 as soon as the surplus capacity is added downstream. The reason for this phenomenon is that the basic flow-density diagram is right-skewed, and the evacuation wave speed is generally not greater than the free flow speed.

[0082] Except for the road exit section, each road section is divided into three types of areas according to whether it is within the intersection range and the function of the intersection lane: basic section (blue area), left-turn section at the intersection (orange area) and straight section at the intersection (green area). The basic section does not divide the lane function, and all lanes are straight lanes. The left-turn section at the intersection only contains the left-turn lane, and the straight section at the intersection only contains the straight lane. According to the idea of ​​the cellular transmission model, this section divides the road into several interconnected cells, and divides the cells into six types according to their characteristics: conventional cells (first type of cells), diversion cells (second type of cells), confluence cells (third type of cells), signal-controlled cells (fourth type of cells), source cells (fifth type of cells) and terminal cells (sixth type of cells). The positions of various types of cells in the road network are shown as follows. Figure 5As shown. The basic section connects the left-turn section and the right-turn section of the intersection that are parallel in space. Therefore, the last cell of the basic section is defined as the diversion cell. When calculating its outflow, it is necessary to calculate the left-turn outflow and the straight-through outflow respectively according to the turning ratio. The first cell of the basic section on the internal section of the road receives the traffic that flows in through left turns and straight-through at the intersection. Therefore, these cells are defined as the inflow cells. When calculating their inflow, it is necessary to consider multiple inflow directions. The last downstream cells of the left-turn section and the straight-through section of the intersection are defined as signal-controlled cells. The capacity of such cells is constrained to 0 during the red light period due to signal control. Its downstream cells need to match the corresponding inflow cells on the exit road according to the lane function. The first cell on the basic section of the road entrance section is defined as the source cell. Its cell stock and maximum capacity are both set to positive infinity, and the cell capacity is set to traffic demand. The traffic flow on the road exit section has already left the signal control range. In most studies, the traffic state evolution process and delay of the exit section are not considered in the model. Therefore, the road exit section as a whole is represented by a cell, called the terminal cell. The maximum capacity of the terminal cell is set to positive infinity, which can carry all vehicles leaving the section.

[0083] S2: Establish a dynamic fleet discrete model, use the dynamic fleet discrete model to predict the evolution of traffic status under different signal timing schemes, obtain the prediction results, and update the cell state according to the prediction results.

[0084] On the road section between intersections, the process of the convoy moving forward while being dispersed due to the heterogeneity of vehicles is highly similar to the one-dimensional convection-diffusion movement of gas molecules. Gas convection refers to the relative flow of various parts within a gas, which can be compared to the process of the overall traffic flow moving from the upstream intersection to the downstream intersection. The gas diffusion phenomenon refers to the phenomenon of gas molecules moving from high-concentration areas to low-concentration areas until they are evenly distributed, which can be compared to the process of the change in the head-to-head distance within a convoy. One-dimensional means that vehicles can only move in the direction of the road. The movement of traffic flow between adjacent intersections can be decomposed into two processes: (1) convection movement at an average speed driven by the intention of "driving to the downstream intersection", and (2) diffusion movement driven by the uneven distribution of vehicle density on the road section.

[0085] S3: Calculate the common cycle duration and green-to-signal ratio based on the updated cell state; construct a phase difference optimization model with the goal of minimizing the total delay of connected vehicles based on the common cycle duration, green-to-signal ratio and prediction results;

[0086] Figure 4The system operation flow chart is shown in Figure 2. The system divides the optimized signal timing parameters into two stages, which are respectively calculating the common cycle duration, allocating the green-to-signal ratio, and optimizing the phase difference. The signal timing plan is updated once a cycle, and at the end of each cycle, the signal timing plan for the next cycle will be generated based on the trajectory data.

[0087] The first stage is the common cycle duration calculation and green-to-signal ratio allocation model. The system first estimates the sampled turning flow and sampled saturated flow rate of the intersection from the real-time trajectory data, and calculates the critical flow rate ratio of the connected vehicles in each signal stage based on the estimated results. Based on the Webster minimum delay signal cycle theory, the system uses the critical flow rate ratio of the connected vehicles as input to calculate the cycle duration of each intersection, uses the shortest green light time of each signal stage to constrain the shortest cycle duration, and takes the maximum value of the cycle duration of each intersection as the common cycle duration. Under the determined cycle duration, the model allocates the green light time of each stage according to the critical flow rate ratio of the connected vehicles, and ensures that the green light time meets the shortest green light time constraint.

[0088] The second stage is the intersection phase difference optimization model. The phase difference optimization is based on the cycle time and green signal ratio determined in the first stage. The system first uses the dynamic fleet discrete model to predict the traffic state evolution process and delay indicators under different phase differences, and then establishes a phase difference optimization model with the goal of minimizing delays.

[0089] The two-stage optimization framework of the dynamic control scheme optimizes the common cycle duration, green-to-signal ratio and phase difference in stages. Compared with the joint optimization, the model has fewer decision variables and is suitable for dynamic signal control scenarios.

[0090] S4: Output optimized signal control scheme based on phase difference optimization model. Figure 3 To adopt the control method of this application

[0091] Specifically, the cell state of the cell is estimated based on the real-time trajectory data, including:

[0092] The real-time trajectory data contains the serial number and location information of the connected vehicles. According to the location of the cell, the vehicle serial number within the spatial range covered by the cell can be filtered out, and Ω i (t) represents the set of connected vehicle numbers in cell i at time t, and the sampled stock of the cell is estimated based on the real-time trajectory data:

[0093]

[0094] Among them, Ω i (t) represents the set of connected vehicle serial numbers in cell i at time t, |Ω i (t)| represents the set Ω i The number of elements in (t), represents the set of all cells;

[0095] The cell capacity can be calculated by multiplying the cell length by the blocking density. The cell sampling capacity is theoretically equal to the product of the cell capacity and the permeability. Considering that the lengths of all cells except the end cell are the same, when the permeability is unknown, the cell sampling capacity is estimated based on the maximum value of the sampling stock of each type of cell:

[0096]

[0097] Among them, a represents the entire road network area, represents the set of key cells;

[0098] Calculate the sampled inflow and sampled outflow of the cell respectively:

[0099]

[0100] in, represents the sampling inflow of cell i, Ω i (t) represents the set of vehicle numbers in cell i at the current moment, Ω i (t-1) represents the set of vehicle numbers in cell i at the previous moment, represents the sampled outflow of cell i, Ω i (t+1) represents the set of vehicle numbers in cell i at the next moment; |Ω i (t)-Ω i (t-1)| represents the set of vehicle numbers that are in cell i at the current moment but not in cell i at the previous moment. The size of the set represents the number of newly added connected vehicles entering the cell at the current moment. |Ω i (t-Ω i (t+1)| represents the set of vehicle numbers that are in cell i at the current moment but not in cell i at the next moment. The size of the set represents the number of connected vehicles that have left the cell at the current moment.

[0101] Cell capacity is a constraint on the maximum outflow of cells, which can be replaced by road capacity. Theoretically, the cell sampling capacity is equal to the product of the cell capacity and the permeability. When the permeability is unknown, the cell capacity is estimated based on the maximum value of the sampled outflow of each type of cell:

[0102]

[0103] Furthermore, on the road section between intersections, the process in which the convoy moves forward while being dispersed due to the heterogeneity of vehicles is highly similar to the one-dimensional convection-diffusion movement of gas molecules. Gas convection refers to the relative flow of various parts within the gas, which can be compared to the process of the overall traffic flow moving from the upstream intersection to the downstream intersection. The gas diffusion phenomenon refers to the phenomenon that gas molecules transfer from high-concentration areas to low-concentration areas until they are evenly distributed, which can be compared to the process of changes in the head-to-head distance within the convoy. One-dimensional means that vehicles can only move in the direction of the road. The movement of traffic flow between adjacent intersections can be decomposed into two processes: (1) convection movement forward at an average speed driven by the intention of "driving to the downstream intersection", and (2) diffusion movement driven by the uneven distribution of vehicle density on the road section. Therefore, a dynamic convective model is established, including:

[0104] Decompose the movement of the convective vehicle fleet between adjacent intersections into convection movement and diffusion movement;

[0105] The flow rate driven by convection motion is expressed as:

[0106] q c (x, t) = v(t)k(x, t);

[0107] Among them, v(t) represents the average speed of the road section at time t; k(x,t) represents the density of the x position at time t;

[0108] For diffusion motion, the uneven distribution of gas molecules leads to a concentration gradient, which further drives the molecular diffusion. Fick's first law establishes the relationship between diffusion flux and concentration gradient:

[0109]

[0110] Among them, J is the diffusion flux, which represents the amount of material flowing through a unit area per unit time; D is the diffusion coefficient, which is related to the properties of the gas itself; represents concentration; x represents position. Fick's first law expresses that the diffusion flux is proportional to the concentration gradient, and the ratio D reflects the diffusion capacity of the gas itself. The negative sign in the formula indicates that the diffusion direction is opposite to the concentration gradient, that is, if the concentration gradient at a certain position is positive (the concentration at the downstream position is greater than that at the upstream position), the molecules in this area will diffuse upstream; conversely, if the concentration gradient at a certain position is negative (the concentration at the downstream position is less than that at the upstream position), the molecules in this area will diffuse downstream. Analogous to Fick's first law, the flow rate driven by diffusion motion is expressed as:

[0111]

[0112] Among them, D is the diffusion coefficient of the fleet, which is a property of the traffic flow. The negative sign in the formula indicates that the diffusion direction is opposite to the direction of the density gradient, that is, if the density gradient at a certain position is positive (the density at the downstream position is greater than that at the upstream position, and the downstream is congested), the driver will slow down to avoid joining the congestion; conversely, if the density gradient at a certain position is negative (the density at the downstream position is less than that at the upstream position, and the upstream is congested), the driver has just experienced congestion and will speed up to leave the congested area. There is a certain difference between the traffic flow diffusion coefficient and the gas diffusion coefficient. Since the thermal motion of gas molecules is disordered, the diffusion of gas molecules always diffuses from high concentration to low concentration, and the gas diffusion coefficient is positive. However, the driver's perception of density is relatively vague, and the driver may have a certain tolerance for joining the congested area, so the diffusion coefficient can be positive or negative.

[0113] By superimposing the convection motion and the diffusion motion, we can get the dynamic fleet discrete model:

[0114]

[0115] The existing fleet discrete model usually uses the flow of the upstream section to predict the flow of the downstream section, that is, the model input is the flow-time curve of the upstream section. However, the information on the time dimension of the flow-time curve is not directly contained in the trajectory data. The trajectory data directly reflects the density distribution in the spatial dimension. The dynamic fleet discrete model established in this paper hopes to use the traffic state in the spatial dimension at the current moment to deduce the traffic state in the spatial dimension at the future moment. The realization of the deduction process requires a conversion relationship between the distribution of traffic state in time and the distribution in space, which can be expressed by the flow conservation equation. The conservation equation means that when the flow rate decreases with distance, the density increases with time. The partial derivative of the flow with respect to the position can be converted into the partial derivative of the density with respect to time. According to this relationship, the density of each position at the future moment can be deduced from the flow and density of each position at the current moment obtained from the trajectory data. The cellular transport model is a simulation method of traffic state evolution based on the conservation equation.

[0116] Therefore, the dynamic fleet discrete model is used to predict the traffic state evolution process under different signal timing schemes, and the prediction results are obtained, including:

[0117] Based on the dynamic fleet discrete model, the conversion relationship between the distribution of traffic status in time and in space is constructed. The conversion relationship is:

[0118]

[0119] Based on the conversion relationship, the estimated result of cell outflow is obtained:

[0120]

[0121] Furthermore, the public cycle duration and green-to-signal ratio are calculated based on the updated cell state, including:

[0122] Define t0 as the time when the old cycle and the new cycle alternate, and C' as the length of the old cycle. At t0, some vehicles that are expected to arrive at intersection n in the new cycle are already within the road range under study, while the other part has not yet entered the road. Definition is the earliest time when the vehicle in the new cycle path j that arrives at lane group m enters the road, is the latest time that the vehicle that has reached lane group m on the new periodic path j at time t0 enters the road. When the connected vehicles are evenly distributed in the traffic flow, theoretically, the sampled turning flow and sampled saturation flow rate are the product of the real turning flow and the real saturation flow rate and the penetration rate. When the penetration rate is unknown, the sampled turning flow and sampled saturation flow rate can be directly estimated through real-time trajectory data. For path j, the flow rate q of the lane group of the new cycle from path j j→m Calculated as:

[0123]

[0124] in, is the latest time when the vehicle that has reached lane group m on the new periodic path j at time t0 enters the road, is the earliest time when the vehicle with the new cycle path j reaches lane group m enters the road, t0 is the time when the old cycle and the new cycle alternate, Represents the set of source cells that path j passes through, and there is only one cell in the set;

[0125] The flow rate for lane group m is calculated as:

[0126]

[0127] The following method can be used to estimate based on trajectory data: select the left-turn section and the straight section of the intersection as the judgment area, obtain the time when all vehicles enter the road in the area according to the trajectory data, and take the entry time of the earliest vehicle in the area as the judgment area. Under the basic assumption of unsaturated traffic, vehicles arriving in each cycle can be evacuated in the same cycle. Therefore, if The vehicles entering the road at any time can reach the intersection, All vehicles entering the road within the next C seconds can reach the intersection within the new cycle. Vehicles entering the road within the next C seconds may also be unable to reach the intersection in the new cycle due to the influence of upstream signal control. However, since the signal control scheme is unknown when predicting the lane group flow, the above estimation method conservatively assumes that the signal control scheme is efficient enough. At the same time, under the basic assumption that traffic demand changes slowly, the old cycle length C' is used to replace C. Estimated

[0128] Based on the predicted q m , the key flow rate ratio of connected vehicles in each stage is vc s The calculation is as follows:

[0129]

[0130] Among them, CAP m Indicates the lane group capacity. The calculation cycle duration is as follows based on the key flow rate ratio of the connected vehicle:

[0131]

[0132] Among them, Y n represents the sum of the critical flow rate ratios of intersection n; L is the total signal loss time. When a cycle contains four stages, L = 20. In order to ensure that the green light time of each signal stage is not too short, when C n When the minimum green time requirement is not met, corrections need to be made as follows:

[0133]

[0134] in, is the minimum green light duration of signal phase s. The common cycle duration is set to the maximum value of all intersection cycle durations:

[0135]

[0136] After determining the common cycle length, the green signal ratio g of each stage at intersection n s Meeting the minimum green time It is determined according to the critical flow rate ratio of each stage under the constraints;

[0137]

[0138] Preferably, the phase difference optimization model aims to minimize the total delay of the new cycle:

[0139]

[0140] in, is the new cycle time set, They represent regular cells respectively; diversion cell, the last cell of the basic section of the intersection; confluence cell, the first cell of the exit road of the intersection (excluding the road exit section); signal control cell, the first cell before the stop line of the intersection.

[0141] On the other hand, the present invention provides a dynamic coordination control system for intersection signals based on real-time trajectory data, which is used to implement the above-mentioned dynamic coordination control method for intersection signals based on real-time trajectory data, such as Figure 6 As shown, the system includes:

[0142] A state calculation module is used to divide the target road network into several types of cells, obtain real-time trajectory data, and estimate the cell state of the cells based on the real-time trajectory data;

[0143] The prediction module is used to establish a dynamic fleet discrete model, use the dynamic fleet discrete model to predict the traffic state evolution process under different signal timing schemes, obtain the prediction results, and update the cell state according to the prediction results;

[0144] A calculation module is used to calculate the common cycle duration and green-to-signal ratio according to the updated cell state;

[0145] The optimization module is used to build a phase difference optimization model with the goal of minimizing the total delay of connected vehicles based on the common cycle duration, green-to-signal ratio and prediction results;

[0146] The output module is used to output an optimized signal control solution based on the phase difference optimization model.

[0147] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0148] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for dynamic coordination control of intersection signals based on real-time trajectory data, characterized in that: The following steps are involved: Divide the target road network into several types of cells, obtain real-time trajectory data, and estimate the cell states of the cells based on the real-time trajectory data; Establishing a dynamic fleet discrete model, using the dynamic fleet discrete model to predict the traffic state evolution process under different signal timing schemes, obtaining prediction results, and updating the cell state according to the prediction results; Calculate the public cycle duration and green-to-signal ratio based on the updated cell state; According to the common cycle duration, the green-to-signal ratio and the prediction results, a phase difference optimization model is constructed with the goal of minimizing the total delay of the connected vehicles; An optimized signal control scheme is output based on the phase difference optimization model.

2. The method for dynamic coordination control of intersection signals based on real-time trajectory data according to claim 1, characterized in that: The step of estimating a cell state of the cell based on the real-time trajectory data includes: The sampled inventory of the cell is estimated based on the real-time trajectory data: Among them, Ω i (t) represents the set of connected vehicle serial numbers in cell i at time t, |Ω i (t)| represents the set Ω i The number of elements in (t), represents the set of all cells; Estimate the cell sampling capacity based on the maximum value of the sampled stock of each type of cell: Among them, a represents the entire road network area, represents the endpoint cell set; Calculate the sampled inflow and sampled outflow of the cell respectively: in, represents the sampling inflow of cell i, Ω i (t) represents the set of vehicle numbers in cell i at the current moment, Ω i (t-1) represents the set of vehicle numbers in cell i at the previous moment, represents the sampled outflow of cell i, Ω i (t+1) represents the set of vehicle numbers in cell i at the next moment; Estimate the cell capacity based on the maximum value of the sampled outflow of each type of cell:

3. The method for dynamic coordinated control of intersection signals based on real-time trajectory data according to claim 1, characterized in that: Establish a dynamic fleet discrete model, including: Decompose the movement of the convective vehicle fleet between adjacent intersections into convection movement and diffusion movement; The flow rate driven by the convective motion is expressed as: q c (x,t)=v(t)k(x,t); Among them, v(t) represents the average speed of the road section at time t; k(x,t) represents the density of the x position at time t; According to Fick's first law, the flow rate driven by the diffusion motion is expressed as: Where D is the fleet diffusion coefficient; The convective motion and the diffusion motion are superimposed to obtain a dynamic fleet discrete model:

4. The method for dynamic coordination control of intersection signals based on real-time trajectory data according to claim 1, characterized in that: The dynamic fleet discrete model is used to predict the traffic state evolution process under different signal timing schemes, and the prediction results are obtained, including: Based on the dynamic fleet discrete model, a conversion relationship between the distribution of traffic status in time and in space is constructed, and the conversion relationship is: Based on the conversion relationship, the estimated result of cell outflow is obtained:

5. The method for dynamic coordination control of intersection signals based on real-time trajectory data according to claim 4 is characterized in that: The public cycle duration and green-to-signal ratio are calculated based on the updated cell state, including: The flow rate of the lane group in the new cycle from path j is: in, is the latest time when the vehicle that has reached lane group m on the new periodic path j at time t0 enters the road, is the earliest time when the vehicle with the new cycle path j reaches lane group m enters the road, t0 is the time when the old cycle and the new cycle alternate, represents the set of source cells that path j passes through; The flow rate of lane group m is calculated based on the flow rate of lane group from path j in the new cycle: Based on the flow rate of lane group m, calculate the critical flow rate ratio vc of the connected vehicles at each stage s for: Among them, CAP m Indicates the capacity of the lane group; Calculate the cycle time based on the key flow rate ratio of the connected vehicle: Among them, Y n represents the sum of the critical flow rate ratios of intersection n; L is the total signal loss time; The common cycle duration is the maximum value of all intersection cycle durations: After determining the common cycle length, the green signal ratio g of each stage at intersection n s Meeting the minimum green time It is determined according to the critical flow rate ratio of each stage under the constraints; 6. The method for dynamic coordinated control of intersection signals based on real-time trajectory data according to claim 5, characterized in that: The phase difference optimization model aims to minimize the total delay of the new cycle: in, is the new cycle time set, They represent regular cells, shunt cells, inflow cells, and signal control cells respectively.

7. The method for dynamic coordinated control of intersection signals based on real-time trajectory data according to claim 5, characterized in that: When the cycle length C n If the minimum green light time requirement is not met, the cycle time C n To make corrections: in, is the minimum green light duration in signal phase s.

8. A dynamic coordination control system for intersection signals based on real-time trajectory data, characterized in that: A method for realizing a dynamic coordinated control method of intersection signals based on real-time trajectory data as described in any one of claims 1 to 7, the system comprising: A state calculation module, used for dividing the target road network into a plurality of cell types, obtaining real-time trajectory data, and estimating the cell state of the cell based on the real-time trajectory data; A prediction module is used to establish a dynamic fleet discrete model, use the dynamic fleet discrete model to predict the traffic state evolution process under different signal timing schemes, obtain prediction results, and update the cell state according to the prediction results; A calculation module is used to calculate the common cycle duration and green-to-signal ratio according to the updated cell state; An optimization module, for constructing a phase difference optimization model with the goal of minimizing the total delay of the connected vehicles according to the common cycle duration, the green-to-signal ratio and the prediction result; An output module is used to output an optimized signal control solution based on the phase difference optimization model.

Images