Coordinated control method of signal timing and vehicle speed at single intersection based on CPS

Abstract

This invention belongs to the field of intelligent transportation and relates to a single-intersection signal timing and vehicle speed coordinated control method based on CPS. The method includes: constructing a traffic control scenario for a two-way six-lane crossroads in a connected environment; modeling the scenario elements and determining the constraints for optimized control; designing an objective function for coordinated control of intersection signal timing and vehicle speed to reduce vehicle travel time and fuel consumption at the intersection; and designing a hierarchical control method based on model predictive control based on the objective function of coordinated control of intersection signal timing and vehicle speed in step S3 to optimize traffic lights and vehicle speed. This invention fully utilizes the advantages of coordinated control of intersection signal timing and vehicle speed, providing a novel intelligent intersection control technology for urban traffic management. It has significant advantages in improving intersection traffic efficiency, alleviating traffic congestion, reducing vehicle travel delays, reducing fuel consumption, and reducing pollutant emissions.

Description

Technical Field

[0001] This invention belongs to the field of cloud control technology for connected vehicles, specifically relating to a single-intersection signal timing and vehicle speed coordinated control method based on CPS. Background Technology

[0002] Intersections are key nodes in urban transportation systems. From the perspective of traffic flow theory, intersections, as typical intermittent flow facilities, exert mandatory blocking control on traffic flow. Traffic lights control the operational order of intersections by isolating traffic flows from different directions in time. As a key node and basic unit of traffic control in urban transportation systems, intersections bear the functions of traffic flow convergence and dispersion. A rational and efficient intersection control system is of great significance to the efficiency, safety, and sustainable development of urban transportation systems.

[0003] The novel traffic cooperative control system based on connected and autonomous vehicles (CAVs) represents the forefront of academic research and an inevitable trend in the field of intelligent transportation. In recent years, with the development of information and communication technologies such as 5G, Vehicle-to-Everything (V2X), and LTE, emerging connected and autonomous vehicles (CAVs) have provided unique solutions to these problems, demonstrating enormous potential in improving traffic efficiency and safety. Equipped with necessary communication and automation technologies, CAVs can interact with surrounding vehicles and infrastructure, achieving vehicle-to-vehicle and vehicle-to-infrastructure collaboration. Compared to traditional human-driven vehicles (HDVs), CAVs possess advantages such as a wider perception range and stronger control capabilities. They can integrate wide-area dynamic traffic information and rely on onboard and cloud control platforms to obtain broader and more rational decision-making and planning results, significantly improving road traffic efficiency and vehicle fuel economy.

[0004] Based on intelligent transportation systems, researchers have proposed technologies such as green wave traffic, sensor signal control, and adaptive traffic signal control for urban intersection management and control. However, due to insufficient computing power, limited perception range and accuracy, and weak vehicle control capabilities, these methods still suffer from short-sightedness and insufficient efficiency, making it difficult to achieve a sustained and significant improvement in traffic efficiency and failing to meet the traffic management and control requirements of the current complex and ever-changing traffic environment.

[0005] Extensive research has been conducted both domestically and internationally on the optimal control of signalized intersections, achieving significant progress and improvements in intersection efficiency and vehicle delay compared to traditional control schemes. However, current research on intersection signal control largely focuses on optimizing traffic signal timing and vehicle speed control separately, lacking effective utilization of vehicle-to-everything (V2X) information and with few studies on the coordinated optimization of intersection signal timing and vehicle speed. In reality, however, the two are highly interdependent. Especially with the rapid development of communication technologies such as 5G, V2X, and DSRC, real-time information transmission between intelligent connected vehicles and traffic signals has become possible. Some researchers have already demonstrated that coupled control of vehicles and traffic signals will further enhance road capacity, providing new opportunities for urban signalized intersection management. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to provide a single-intersection signal timing and vehicle speed coordinated control method based on CPS.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A single-intersection signal timing and vehicle speed coordinated control method based on CPS includes the following steps:

[0009] S1. Construct a traffic control scenario for a two-way six-lane intersection in a connected environment;

[0010] The intersection traffic light controller is essentially an edge cloud node, possessing communication capabilities and the ability to perform computing tasks; all vehicles in the scenario are CAVs, which share their own driving status in real time through DSRC and 5G communication technologies, and can obtain the status of traffic lights and the decision results of the edge cloud; vehicles have already completed lane changing and overtaking before entering the control area;

[0011] S2. Model the scene elements and determine the constraints for optimization control;

[0012] For each scene element, a vehicle kinematics model, a vehicle dynamics model, a vehicle energy consumption model, a traffic light model, and a traffic light-vehicle relationship model are modeled, and safety constraints are determined.

[0013] S3. Design an objective function for coordinated control of intersection signal timing and vehicle speed to reduce vehicle travel time and fuel consumption at intersections;

[0014] S4. Based on the objective function of the intersection signal timing and vehicle speed coordinated control in step S3, design a hierarchical control method based on model predictive control to optimize traffic lights and vehicle speed.

[0015] Furthermore, in step S1, the communication range between the edge cloud and CAV is 200m.

[0016] Furthermore, in step S2, the vehicle kinematic model is used to reflect the state changes of the vehicle under the control strategy. The longitudinal motion behavior of the vehicle should satisfy the corresponding kinematic constraints, which are given in discrete form below:

[0017] v i,j (k f +1)=v i,j (k f )+a i,j (k f )ΔT f

[0018]

[0019] 0≤ν i,j (k f )≤v max -a min ≤a i,j (k f )≤a max

[0020] Where s i,j (k f ), v i,j (k f ), a i,j (k f ) represents the number of vehicles i in lane j at the node level k f Position, velocity, and acceleration at any given moment; v max The maximum speed limit for this section of road; a min and a max These are the absolute values ​​of the vehicle's maximum braking deceleration and maximum acceleration, respectively.

[0021] Furthermore, in step S2, the vehicle dynamics model is used to reflect the quantitative relationship between vehicle power and the kinematic model. The vehicle dynamics model is based on the effective traction force F of the engine. e Braking force F b and the driving force F on the road d To calculate the vehicle's acceleration:

[0022] ma = F e -F b +F d

[0023] Among them, the road driving force F d Combining aerodynamic drag, rolling resistance, and road gradient drag, the expression is:

[0024]

[0025] Furthermore, regarding the engine's effective traction force F e have:

[0026]

[0027] After combining the results, we can conclude that:

[0028]

[0029] In the formula, P is the vehicle's output power; F b It is the braking force when the vehicle decelerates; m is the mass of the vehicle; C D ρ is the air resistance coefficient; A is the vehicle's frontal area; a ρ is air density; f is the rolling resistance coefficient of the tire; g is gravitational acceleration; β is the road slope angle; v is the vehicle speed.

[0030] Furthermore, in step S2, the vehicle energy consumption model is used to evaluate the vehicle's eco-driving function, and its expression is as follows:

[0031]

[0032] In the formula, β0, β1, and β2 are the fuel consumption model coefficients determined by the vehicle type; P(k f () represents the instantaneous power of the vehicle.

[0033] Furthermore, in step S2, the traffic light model is based on signal constraints established using a standard double-ring barrier traffic signal structure. It has four linear motions and four left-turn motions, comprising upper and lower rings. Each ring is divided into two groups, and traffic lights in different rings within each group can coexist. s The phase state at time t is expressed in vector form as follows:

[0034] Φ(k s )=[θ1(k s ) θ2(k s ) … θ p (k s ) … θ P (k s )] T

[0035] Where P represents all phases; θ p (k s ) is a binary variable, θ p (k s ) = 1 indicates that in k sPhase p is active at any given time; to prevent traffic flow conflicts, only one phase can be active at any given time, therefore there are constraints between phases:

[0036]

[0037] In addition, define matrix A J×P The signal phase Φ(k) s The right-of-way mapped to the lane, i.e.:

[0038]

[0039] In the formula, For k s At any given time, the right-of-way status of all lanes J is as follows: j (k s ) = 1 indicates that in k s Lane j has the right-of-way, meaning vehicles can enter the intersection area; otherwise, they must stop and wait at the stop line.

[0040] Furthermore, in step S2, the traffic light-vehicle relationship model introduces a binary variable δ. i,j (k f (), used to indicate whether a vehicle has passed the stop line at a signalized intersection:

[0041] When the vehicle has not passed the stop line, that is... i,j (k f )≤s j At that time, δ i,j (k f ) = 1;

[0042] When the vehicle passes the stop line, that is, s i,j (k f )>s j At that time, δ i,j (k f )=0;δ i,j (k f The calculation formula for ) is as follows:

[0043] -Mδ i,j (k f )≤s i,j (k f )-s j ≤M(1-δ i,j (k f ))

[0044] When the traffic light is red, and taking the yellow light into account, vehicles cannot cross the stop line at a signalized intersection. The constraint in this case is expressed as:

[0045] s i,j (kf )≤s j +M(1-δ i,j (k f )+r j (k f )r j (k f +ε))

[0046] In the formula, ε is the preset yellow light step size, and the corresponding yellow light time is ΔT. f ·ε; when the remaining green light time is less than ΔT f When ε is reached, it can be considered a yellow light, i.e., r j (k f If (+ε) = 0, vehicles that have not passed the stop line at the intersection must not proceed.

[0047] There are two cases in which constraints can be ignored:

[0048] When δ i,j (k f ) = 1, r j (k f ) = 1 and r j (k f When +ε)=1, meaning vehicle i has not crossed the stop line, and the phase corresponding to lane j is green with a remaining time greater than the yellow light time, the vehicle can normally cross the stop line at the signalized intersection. At this time, constraint s i,j (k f )≤s j Can be relaxed;

[0049] When δ i,j (k f When ) = 0, meaning vehicle i has already passed the stop line at the signalized intersection, constraint s is then... i,j (k f )≤s j They were relaxed.

[0050] Furthermore, in step S2, the safety constraint refers to the requirement that when a vehicle is following another vehicle, adjacent vehicles in the same lane should maintain a safe distance to avoid collisions.

[0051] The expression for the safety constraint is:

[0052] s i,j (k f )-s i-1,j (k f )≥τ h v i,j (k f )+d0

[0053] Where s i,j (k f), s i-1,j (k f ) represent the positions of the i-th and i-1-th vehicles in lane j, respectively; τ h d0 represents the minimum headway; d0 represents the safe distance when the vehicle is stationary.

[0054] Furthermore, in step S3, the objective function for coordinated control of intersection signal timing and vehicle speed takes minimizing the total time for vehicles to pass through the intersection within the control area as the primary optimization objective and minimizing vehicle fuel consumption as the secondary objective. The calculation and decision-making are performed at the edge node of the subsystem-level cyber-physical system to obtain the optimal phase sequence and duration.

[0055] The expression for calculating the objective function of coordinated control of intersection signal timing and vehicle speed is as follows:

[0056]

[0057] In the formula, J tt and J fc α1 and α2 represent the vehicle travel time and fuel consumption at the intersection, respectively; α1 and α2 are the weights of α1 and α2 in the system objective function J, where α1 > α2 > 0; The moment when vehicle i in lane j crosses the stop line at the intersection; β1 represents the moment when vehicle i enters the control zone in lane j; β2 represents the weights of the primary and secondary objectives.

[0058] Furthermore, the specific steps of step S4 are as follows:

[0059] Combining model predictive control and hierarchical optimization control, a hierarchical control method based on model predictive control is proposed, which performs rolling optimization in the upper and lower layers to optimize traffic lights and vehicle speeds.

[0060] The upper-level intersection traffic light control problem can be represented in the form of rolling optimization as follows:

[0061]

[0062] The constraints are:

[0063]

[0064] The vehicle speed optimization problem at the lower level can be represented in the form of rolling optimization as follows:

[0065]

[0066] The constraints are:

[0067]

[0068] In the formula, h represents the time step of the first h slow scales.

[0069] Beneficial effects:

[0070] 1. This invention designs a collaborative control method for signal timing and vehicle speed at a single intersection in a fully CAV environment based on the multi-scale characteristics of cyber-physical systems. It studies the collaborative control strategy for signal timing and vehicle speed at urban intersections based on cyber-physical systems at the system level. This collaborative control strategy aims to reduce vehicle travel time and fuel consumption. The collaborative control is explicitly modeled as a mixed-integer nonlinear programming (MINLP) problem involving two scales: node-level and subsystem-level, and the coupling relationship between the two scales is clarified. By modeling the traffic scenario, including vehicle models, traffic light-related models, and the relationship between traffic lights and vehicles, the correct interaction between vehicles and intersection traffic lights is ensured. The collaborative control objective function is designed with minimizing the total time for vehicles to pass through the intersection as the subsystem-level objective and minimizing vehicle fuel consumption as the node-level objective.

[0071] 2. A coordinated control model for signal timing and vehicle speed at a single intersection is constructed. Addressing the complexity and difficulty in solving this multi-scale problem due to inter-scale coupling, a hierarchical model predictive control (MMDC) solution method is designed, leveraging the varying real-time requirements at different scales. This method performs rolling optimization at both the upper and lower levels to optimize traffic lights and vehicle speeds. The upper-level optimization problem is solved using a roadside edge cloud, while the lower-level optimization problem is solved by an onboard computing platform. This method effectively reduces problem complexity, facilitates the acquisition of high-quality solutions, and minimizes computational resource consumption.

[0072] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0073] Figure 1 This is a flowchart of a single-intersection signal timing and vehicle speed coordinated control method based on CPS according to the present invention;

[0074] Figure 2 This is a schematic diagram illustrating the principle of a single-intersection signal timing and vehicle speed coordinated control method based on CPS according to the present invention.

[0075] Figure 3 This is a schematic diagram of a typical two-way six-lane intersection in a connected environment.

[0076] Figure 4 Flowchart of traffic optimization model predictive control;

[0077] Figure 5 The average travel time of vehicles under asymmetric traffic flow intensity;

[0078] Figure 6 Comparison of average fuel consumption under asymmetric traffic flow. Detailed Implementation

[0079] To make the technical solutions, advantages, and objectives of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the protection scope of this application.

[0080] like Figure 1 , Figure 2 and Figure 4 As shown, this invention provides a CPS-based method for coordinated control of signal timing and vehicle speed at a single intersection, comprising the following steps:

[0081] S1. Construct a traffic control scenario for a two-way six-lane intersection under a connected environment. A schematic diagram of this scenario is shown below. Figure 3 As shown;

[0082] The intersection traffic light controller is essentially an edge cloud node, possessing communication capabilities and the ability to perform computational tasks. It has autonomous perception, decision-making, control, and coordination capabilities, and can share the vehicle's own driving status in real time through communication technologies such as DSRC and 5G. It can also obtain the status of traffic lights and the decision results of the edge cloud. The reliable communication range between the edge cloud and the CAV is 200m, within which all communication is close to ideal, ignoring data packet loss and communication delays. The vehicle has already completed lane changing and overtaking before entering the control area.

[0083] S2. Model the scene elements and determine the constraints for optimization control;

[0084] For each scene element, a vehicle kinematics model, a vehicle dynamics model, a vehicle energy consumption model, a traffic light model, and a traffic light-vehicle relationship model are modeled, and safety constraints are determined.

[0085] The vehicle kinematic model is used to reflect the state changes of the vehicle under the control strategy. The longitudinal motion behavior of the vehicle should satisfy the corresponding kinematic constraints, which are given in discrete form below:

[0086] v i,j (k f+1)=v i,j (k f )+a i,j (k f )ΔT f

[0087]

[0088] 0≤ν i,j (k f )≤v max -a min ≤a i,j (k f )≤a max

[0089] Where s i,j (k f ), v i,j (k f ), a i,j (k f ) represents the number of vehicles i in lane j at the node level k f Position, velocity, and acceleration at any given moment; v max The maximum speed limit for this section of road; a min and a max These are the absolute values ​​of the vehicle's maximum braking deceleration and maximum acceleration, respectively.

[0090] The vehicle dynamics model is used to reflect the quantitative relationship between vehicle power and kinematic model. This invention uses a longitudinal dynamics model as the vehicle dynamics model, which is based on the effective traction force F of the engine. e Braking force F b and the driving force F on the road d To calculate the vehicle's acceleration:

[0091] ma = F e -F b +F d

[0092] Among them, the road driving force F d Combining aerodynamic drag, rolling resistance, and road gradient drag, the expression is:

[0093]

[0094] Furthermore, regarding the engine's effective traction force F e have:

[0095]

[0096] Combining the above formulas, we can obtain:

[0097]

[0098] In the formula, P is the vehicle's output power; F b It is the braking force when the vehicle decelerates; m is the mass of the vehicle; C D ρ is the air resistance coefficient; A is the vehicle's frontal area; a ρ is air density; f is the rolling resistance coefficient of the tire; g is gravitational acceleration; β is the road slope angle; v is the vehicle speed.

[0099] The vehicle energy consumption model is used to evaluate the eco-driving function of a vehicle, and its expression is as follows:

[0100]

[0101] In the formula, β0, β1, and β2 are the fuel consumption model coefficients determined by the vehicle type; P(k f The instantaneous power of the vehicle is represented by P; this is achieved by controlling the engine power P and the braking force F. b The vehicle acceleration was adjusted; by combining kinematic and energy consumption models, the relationship between vehicle engine power, vehicle acceleration, vehicle speed, and vehicle travel time was established.

[0102] When following another vehicle in the same lane, a safe distance should be maintained between adjacent vehicles to avoid collisions. Vehicles should adhere to safety constraints. The expression for the safety constraints is:

[0103] s i,j (k f )-s i-1,j (k f )≥τ h v i,j (k f )+d0

[0104] Where s i,j (k f ), s i-1,j (k f ) represent the positions of the i-th and i-1-th vehicles in lane j, respectively; τ h d0 represents the minimum headway; d0 represents the safe distance when the vehicle is stationary.

[0105] The traffic light model is based on the signal constraints established by the standard double-ring barrier traffic signal structure. It has four linear motions and four left-turn motions, comprising upper and lower rings. Each ring is divided into two groups, and traffic lights in different rings within each group can coexist. s The phase state at time t is expressed in vector form as follows:

[0106] Φ(k s )=[θ1(k s ) θ2(ks )…θ p (k s )…θ P (k s )] T

[0107] In the formula, P represents all phases, and in this embodiment, P = 8; θ p (k s ) is a binary variable, θ p (k s ) = 1 indicates that in k s Phase p is active at any given time; to prevent traffic flow conflicts, only one phase can be active at any given time, therefore there are constraints between phases:

[0108]

[0109] In addition, define matrix A J×P The signal phase Φ(k) s The right-of-way mapped to the lane, i.e.:

[0110]

[0111] In the formula, For k s At any given time, the right-of-way status of all lanes J is as follows: j (k s ) = 1 indicates that in k s Lane j has the right-of-way, meaning vehicles can enter the intersection area; otherwise, they must stop and wait at the stop line.

[0112] The traffic light and vehicle relationship model introduces a binary variable δ. i,j (k f (), used to indicate whether a vehicle has passed the stop line at a signalized intersection:

[0113] When the vehicle has not passed the stop line, that is... i,j (k f )≤s j At that time, δ i,j (k f ) = 1;

[0114] When the vehicle passes the stop line, that is, s i,j (k f )>s j At that time, δ i,j (k f )=0;δ i,j (k f The calculation formula for ) is as follows:

[0115] -Mδ i,j (kf )≤s i,j (k f )-s j ≤M(1-δ i,j (k f ))

[0116] When the traffic light is red, and taking the yellow light into account, vehicles cannot cross the stop line at a signalized intersection. The constraint in this case is expressed as:

[0117] s i,j (k f )≤s j +M(1-δ i,j (k f )+r j (k f )r j (k f +ε))

[0118] In the formula, ε is the preset yellow light step size, and the corresponding yellow light time is ΔT. f ·ε; when the remaining green light time is less than ΔT f When ε is reached, it can be considered a yellow light, i.e., r j (k f If (+ε) = 0, vehicles that have not passed the stop line at the intersection must not proceed.

[0119] There are two cases in which constraints can be ignored:

[0120] When δ i,j (k f ) = 1, r j (k f ) = 1 and r j (k f When +ε)=1, meaning vehicle i has not crossed the stop line, and the phase corresponding to lane j is green with a remaining time greater than the yellow light time, the vehicle can normally cross the stop line at the signalized intersection. At this time, constraint s i,j (k f )≤s j Can be relaxed;

[0121] When δ i,j (k f When ) = 0, meaning vehicle i has already passed the stop line at the signalized intersection, constraint s is then... i,j (k f )≤s j They were relaxed.

[0122] S3. Design an objective function for the coordinated control of intersection signal timing and vehicle speed, with the system objective of reducing vehicle travel time and fuel consumption at the intersection. The coordinated control problem of traffic lights and vehicle speed at a single signalized intersection is explicitly modeled as a mixed integer nonlinear programming (MINLP) problem with two scales: node level and subsystem level, and the coupling relationship between the two scales is clarified.

[0123] The coordinated control of traffic light timing and vehicle speed at a single intersection is explicitly modeled as follows:

[0124] min J=α1J tt +α2J fc

[0125] Among them, J tt and J fc Let α1 and α2 represent the vehicle travel time and fuel consumption at the intersection, respectively, and α1 and α2 be their weights in the system objective function J. Since the subsystem-level scale of the intersection is larger than the node-level scale of the vehicle, macro-level regulation often brings greater benefits. Therefore, the weight of the former is larger, i.e., α1 > α2 > 0.

[0126] Based on the previously established vehicle kinematics model, dynamics model, safety constraints, and traffic signal phase model, the primary optimization objective is to minimize the total time for vehicles to pass through the intersection within the control area, with minimizing vehicle fuel consumption as a secondary objective. Calculations and decisions are performed at the edge nodes of the subsystem-level cyber-physical system to obtain the optimal (or suboptimal, depending on computational efficiency) phase sequence and duration under the current system state. The mathematical expression for the optimization objective is:

[0127]

[0128] in, Let i be the moment when vehicle i in lane j crosses the stop line at the intersection. The moment the vehicle enters the controlled area, β1 and β2 are the weights of the primary and secondary objectives, respectively;

[0129] The optimal timing result Φ(k) of the traffic lights is obtained by the subsystem-level cyber-physical system calculation. s This information is transmitted to the vehicle node-level cyber-physical system via a communication link to complete information interaction between different scales.

[0130] Combining vehicle kinematics, dynamics, safety constraints, and a model of the relationship between traffic lights and the vehicle, this study optimizes fuel economy in real time on a vehicle-side computing platform. The engine output power is adjusted to optimize vehicle speed and improve fuel economy. The optimization objective of minimizing vehicle fuel consumption is established as follows:

[0131]

[0132] Among them, f i,j (k f The vehicle fuel consumption model is established. In summary, the coordinated control problem of traffic lights and vehicle speed at a single signalized intersection can be described as follows:

[0133]

[0134] S4. Based on the objective function of the intersection signal timing and vehicle speed coordinated control in step S3, and combining model predictive control and hierarchical optimization control, a hierarchical control method based on model predictive control is designed. The method is carried out in the upper and lower layers in a rolling optimization manner to optimize the traffic lights and vehicle speed.

[0135] Given that the coupling effects between different scales in this multi-scale, multi-objective optimization problem lead to model complexity and solution difficulties, and that different real-time requirements exist, this invention designs a hierarchical control method based on model predictive control, such as... Figure 2 As shown;

[0136] This method employs a rolling optimization approach at both the upper and lower layers to optimize traffic lights and vehicle speeds. In the hierarchical structure, the control variable for the upper-level subsystem-level intelligent vehicle cyber-physical system is the traffic light phase sequence θ. * (k s )=[θ * (k s |k s ),θ * (k s +1|k s ),…,θ * (k s +T sc -1|k s [and the position sequence of vehicles within the control area] The control input is the status information of the vehicles in the control area, i.e., the vehicle position s. i,j (k s ) and vehicle speed v i,j (k s ), where j∈J, i∈N j The signal crossroads (edge ​​cloud) at each subsystem level at time k s Vehicle status information is acquired and combined with the current traffic light phase. Model predictive control is then implemented in a rolling time domain, with a predicted time domain step size T. s Control time-domain step size T sc (T sc ≤T s Solve the signal optimization problem at the signalized intersection, obtain and implement k sThe first ΔT after time 1 s The time-step signal light control sequence simultaneously generates a sequence of key vehicle positions within the current control area, enabling vehicles to collaboratively achieve the system's control objectives. Similarly, in the next k... s At time +1, repeat the above optimization steps to continuously obtain and apply new traffic light sequences;

[0137] The upper-level intersection traffic light control problem can be represented in the form of rolling optimization as follows:

[0138]

[0139] The constraints are:

[0140]

[0141] In the hierarchical structure, the control variable of the lower-level node-level intelligent vehicle cyber-physical system is the vehicle's power sequence. Used to control the vehicle's acceleration and speed; the control input is the vehicle position sequence obtained from the upper-level intelligent vehicle cyber-physical system. Where j∈J, i∈N j That is, the node level follows the position guidance of the subsystem scale, but only the top N obtained from the subsystem-level solution are applied. h Speed ​​planning is performed using vehicle position sequences; therefore, the prediction time domain for node-scale model predictive control is N. h ΔT s During operation, the onboard computing platform operates at time intervals ΔT. f For periodicity, at time k in each node... f Obtain the latest vehicle position sequence and signal phase, and combine them with the vehicle's own state to predict the time-domain step T. f =N h ΔT s / ΔT f Control time-domain steps T fc (T fc ≤T f Solve the vehicle speed optimization problem to reduce fuel consumption, but only implement the first ΔT. f The speed control sequence is determined by the time step, followed by a similar rolling optimization process.

[0142] The vehicle speed optimization problem at the lower level can be represented in the form of rolling optimization as follows:

[0143]

[0144] The constraints are:

[0145]

[0146] In the formula, h represents the time step of the first h slow scales.

[0147] The speed planning for vehicles on the lower level needs to follow the position sequence obtained from the optimized traffic lights on the upper level, meaning that vehicles need to arrive at specific locations at specific times. This ensures the achievement of the upper-level control objectives. Based on the known conditions, after determining the target distance and required time, the vehicle can obtain the speed that minimizes fuel consumption according to the cost function and constraints.

[0148] Some specific parameters used in this embodiment are shown in Table 1 below.

[0149] Table 1

[0150]

[0151]

[0152] Finally, the proposed intersection signal timing and vehicle speed coordinated control method was tested and verified. The simulation process and results are as follows: Figures 5-6 As shown, the evaluation results demonstrate the effectiveness and feasibility of the method of the present invention.

[0153] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the protection scope of the present invention.

Claims

1. A single-intersection signal timing and vehicle speed coordinated control method based on CPS, characterized in that, Includes the following steps: S1. Construct a traffic control scenario for a two-way six-lane intersection in a connected environment; The intersection traffic light controller is essentially an edge cloud node, possessing communication capabilities and the ability to perform computing tasks; all vehicles in the scenario are CAVs, which share their own driving status in real time through DSRC and 5G communication technologies, and can obtain the status of traffic lights and the decision results of the edge cloud; vehicles have already completed lane changing and overtaking before entering the control area; S2. Model the scene elements and determine the constraints for optimization control; For each scene element, a vehicle kinematics model, a vehicle dynamics model, a vehicle energy consumption model, a traffic light model, and a traffic light-vehicle relationship model are modeled, and safety constraints are determined. S3. Design an objective function for coordinated control of intersection signal timing and vehicle speed to reduce vehicle travel time and fuel consumption at intersections; S4. Based on the objective function of the coordinated control of intersection signal timing and vehicle speed in step S3, a hierarchical control method based on model predictive control is designed. This method is carried out in the upper and lower layers in a rolling optimization manner to optimize traffic lights and vehicle speed. In the hierarchical structure, the control variable of the upper-level subsystem-level intelligent vehicle cyber-physical system is the traffic light phase sequence θ. * (k s )=[θ * (k s |k s ),θ * (k s +1|k s ),…,θ * (k s +T sc -1|k s [and the position sequence of vehicles within the control area] The control input is the status information of the vehicles in the control area, i.e., the vehicle position s. i,j (k s ) and vehicle speed v i,j (k s ), where j∈J, i∈N j j is the lane number, J is the set of lane numbers; i is the vehicle number, N j Let j be the number of vehicles in lane j; the edge cloud at time k in each subsystem. s Vehicle status information is acquired and combined with the current traffic light phase. Model predictive control is then implemented in a rolling time domain, with a predicted time domain step size T. s Control time-domain step size T sc Solve the signal optimization problem at a signalized intersection, T sc ≤T s , obtain and implement k s The first ΔT after time 1 s The time-step signal light control sequence simultaneously generates the key position sequence of vehicles within the current control area, enabling vehicles to collaboratively achieve the system control objective; similarly, in the next k... s At time +1, repeat the above optimization steps to continuously obtain and apply new traffic light sequences; In the hierarchical structure, the control variable of the lower-level node-level intelligent vehicle cyber-physical system is the vehicle's power sequence. Used to control the vehicle's acceleration and speed; the control input is the vehicle position sequence obtained from the upper-level intelligent vehicle cyber-physical system. Where j∈J, i∈N j That is, the node level follows the position guidance of the subsystem scale, but only the top N obtained from the subsystem-level solution are applied. h Speed ​​planning is performed using vehicle position sequences; therefore, the prediction time domain for node-scale model predictive control is N. h ΔT s During operation, the onboard computing platform operates at time intervals ΔT. f For periodicity, at time k in each node... f Obtain the latest vehicle position sequence and signal phase, and combine them with the vehicle's own state to predict the time-domain step T. f =N h ΔT s / ΔT f Control time-domain steps T fc Solve the vehicle speed optimization problem, T fc ≤T f To reduce fuel consumption, but only the first ΔT is implemented. f The speed control sequence is determined by the time step, followed by a similar rolling optimization process.

2. The method for coordinated control of signal timing and vehicle speed at a single intersection based on CPS according to claim 1, characterized in that: In step S1, the communication range between the edge cloud and CAV is 200m.

3. The method for coordinated control of signal timing and vehicle speed at a single intersection based on CPS according to claim 1, characterized in that: In step S2, the vehicle kinematic model is used to reflect the state changes of the vehicle under the control strategy. The longitudinal motion behavior of the vehicle should satisfy the corresponding kinematic constraints, which are given in discrete form below: v i,j (k f +1)=v i,j (k f )+a i,j (k f )ΔT f 0≤ν i,j (k f )≤v max -a min ≤a i,j (k f )≤a max Where s i,j (k f ), v i,j (k f ), a i,j (k f ) represents the number of vehicles i in lane j at the node level k f Position, velocity, and acceleration at any given moment; v max The maximum speed limit for this section of road; a min and a max These are the absolute values ​​of the vehicle's maximum braking deceleration and maximum acceleration, respectively.

4. The method for coordinated control of signal timing and vehicle speed at a single intersection based on CPS according to claim 3, characterized in that: In step S2, the vehicle dynamics model is used to reflect the quantitative relationship between vehicle power and kinematics. The vehicle dynamics model is based on the effective traction force F of the engine. e Braking force F b and the driving force F on the road d To calculate the vehicle's acceleration: in=F e -F b +F d Among them, the road driving force F d Combining aerodynamic drag, rolling resistance, and road gradient drag, the expression is: Furthermore, regarding the engine's effective traction force F e have: After combining the results, we can conclude that: In the formula, P is the vehicle's output power; F b It is the braking force when the vehicle decelerates; m is the mass of the vehicle; C D ρ is the air resistance coefficient; A is the vehicle's frontal area; a ρ is air density; f is the rolling resistance coefficient of the tire; g is gravitational acceleration; β is the road slope angle; v is the vehicle speed.

5. The method for coordinated control of signal timing and vehicle speed at a single intersection based on CPS according to claim 4, characterized in that: In step S2, the vehicle energy consumption model is used to evaluate the vehicle's eco-driving function, and its expression is as follows: In the formula, β0, β1, and β2 are the fuel consumption model coefficients determined by the vehicle type; P(k f () represents the instantaneous power of the vehicle.

6. The method for coordinated control of signal timing and vehicle speed at a single intersection based on CPS according to claim 5, characterized in that: In step S2, the traffic light model is based on signal constraints established using a standard double-ring barrier traffic signal structure. It has four linear motions and four left-turn motions, comprising upper and lower rings. Each ring is divided into two groups, and traffic lights in different rings within each group can coexist. s The phase state at time t is expressed in vector form as follows: Φ(k s )=[θ1(k s )θ2(k s )…θ p (k s )…θ P (k s )] T Where P represents all phases; θ p (k s ) is a binary variable, θ p (k s ) = 1 indicates that in k s Phase p is active at any given time; to prevent traffic flow conflicts, only one phase can be active at any given time, therefore there are constraints between phases: In addition, define matrix A J×P The signal phase Φ(k) s The right-of-way mapped to the lane, i.e.: In the formula, For k s At any given time, the right-of-way status of all lanes J is as follows: j (k s ) = 1 indicates that in k s Lane j has the right-of-way, meaning vehicles can enter the intersection area; otherwise, they must stop and wait at the stop line.

7. The method for coordinated control of signal timing and vehicle speed at a single intersection based on CPS according to claim 6, characterized in that: In step S2, the traffic light and vehicle relationship model introduces a binary variable δ. i,j (k f (), used to indicate whether a vehicle has passed the stop line at a signalized intersection; When the vehicle has not passed the stop line, that is... i,j (k f )≤s j At that time, δ i,j (k f ) = 1; When the vehicle passes the stop line, that is, s i,j (k f )>s j At that time, δ i,j (k f )=0;δ i,j (k f The calculation formula for ) is as follows: -Mδ i,j (k f )≤s i,j (k f )-s j ≤M(1-δ i,j (k f )) When the traffic light is red, and taking the yellow light into account, vehicles cannot cross the stop line at a signalized intersection. The constraint in this case is expressed as: s i,j (k f )≤s j +M(1-d i,j (k f )+r j (k f )r j (k f +e)) In the formula, ε is the preset yellow light step size, and the corresponding yellow light time is ΔT. f ·ε; when the remaining green light time is less than ΔT f When ε is reached, it can be considered a yellow light, i.e., r j (k f If (+ε) = 0, vehicles that have not passed the stop line at the intersection must not proceed. There are two cases in which constraints can be ignored: When δ i,j (k f ) = 1, r j (k f ) = 1 and r j (k f When +ε)=1, meaning vehicle i has not crossed the stop line, and the phase corresponding to lane j is green with a remaining time greater than the yellow light time, the vehicle can normally cross the stop line at the signalized intersection. At this time, constraint s i,j (k f )≤s j Can be relaxed; When δ i,j (k f When ) = 0, meaning vehicle i has already passed the stop line at the signalized intersection, constraint s is then... i,j (k f )≤s j They were relaxed.

8. The method for coordinated control of signal timing and vehicle speed at a single intersection based on CPS according to claim 7, characterized in that: In step S2, the safety constraint refers to the requirement that when a vehicle is following another vehicle, adjacent vehicles in the same lane should maintain a safe distance to avoid collisions. The expression for the safety constraint is: s i,j (k f )-s i-1,j (k f )≥τ h v i,j (k f )+d0 Where s i,j (k f ), s i-1,j (k f ) represent the positions of the i-th and i-1-th vehicles in lane j, respectively; τ h d0 represents the minimum headway; d0 represents the safe distance when the vehicle is stationary.

9. A CPS-based method for coordinated control of signal timing and vehicle speed at a single intersection, as described in claim 8, is characterized in that... In step S3, the objective function for coordinated control of intersection signal timing and vehicle speed takes minimizing the total time for vehicles to pass through the intersection within the control area as the primary optimization objective and minimizing vehicle fuel consumption as the secondary objective. The calculation and decision-making are performed at the edge node of the subsystem-level cyber-physical system to obtain the optimal phase sequence and duration. The expression for calculating the objective function of coordinated control of intersection signal timing and vehicle speed is as follows: In the formula, J tt and J fc α1 and α2 represent the vehicle travel time and fuel consumption at the intersection, respectively; α1 and α2 are the weights of the two in the system objective function J, α1 > α2 > 0; β1 and β2 are the weights of the primary objective and the secondary objective, respectively.

10. A CPS-based method for coordinated control of signal timing and vehicle speed at a single intersection, as described in claim 9, is characterized in that... The specific steps of step S4 are as follows: Combining model predictive control and hierarchical optimization control, a hierarchical control method based on model predictive control is proposed, which performs rolling optimization in the upper and lower layers to optimize traffic lights and vehicle speeds. The upper-level intersection traffic light control problem can be represented in the form of rolling optimization as follows: The constraints are: The vehicle speed optimization problem at the lower level can be represented in the form of rolling optimization as follows: The constraints are: In the formula, h represents the time step of the first h slow scales.

Images