Man-machine mixed driving intersection signal timing and vehicle trajectory optimization method

Abstract

The invention discloses a man-machine mixed driving intersection signal timing and vehicle trajectory optimization method, and aims to solve the problem of traffic efficiency and energy consumption optimization during mixed driving of intelligent network connection vehicles and manual driving vehicles. Firstly, intersection passing right is dynamically allocated based on vehicle arrival information, and a special lane and a special signal phase are designed to reduce mutual interference of two types of vehicles. And secondly, by constructing a signal timing optimization model adaptive to a man-machine mixed driving environment, the intelligent network connection vehicle cooperatively passes in a special lane phase, and shares signal timing with the manual driving vehicle in a common lane phase, so that the passing delay is minimized. For the intelligent network connection vehicle, the speed trajectory of the intelligent network connection vehicle is optimized by combining an intelligent driving model and an energy balance fuel consumption calculation formula, so that the intersection passing efficiency is improved, and the energy consumption is reduced. According to the method, traffic flow changes under different intelligent network connection vehicle permeability can be effectively coped with, the overall traffic operation efficiency is improved on the basis of guaranteeing safety, and energy consumption is reduced.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of traffic engineering and traffic information and control system, relates to the field of traffic signal and vehicle trajectory control, and more particularly to a method for intersection signal timing and vehicle trajectory optimization in human-machine mixed driving. BACKGROUND

[0002] With the rapid development of networked and automated driving technology, connected automated vehicles (CAV) are gradually becoming an important part of future urban transportation systems. Precise planning of intelligent connected vehicle trajectories is considered a key strategy to solve urban traffic congestion and improve traffic efficiency. Studies have shown that when the following distance of connected automated vehicles is not limited by the type of the vehicle in front, it can significantly improve the capacity of intersections (Qian GM, et al. Intersection space-time resource allocation optimization under intelligent connected mixed driving environment[J]. Journal of Zhejiang University: Engineering Science, 2021, 55(06): 1019-1026). However, in the foreseeable future, urban traffic flow will be in a long-term mixed state of human-driven vehicles and intelligent connected vehicles. This human-machine mixed driving traffic environment brings new challenges to the signal control and traffic organization of intersections. Existing technologies have proposed various optimization methods for signal control of mixed traffic flow at intersections. Most studies focus on implementing a unified control strategy for vehicles of both driving modes, i.e., mixed traffic flow sharing the same right-of-way and green time. However, this strategy ignores the differences in operating characteristics between intelligent connected vehicles and human-driven vehicles, and fails to fully utilize the advantages of intelligent connected vehicles in response speed, path planning, and traffic efficiency. In addition, the shared phase strategy can easily cause interference between different types of vehicles, affecting the overall traffic efficiency and safety of mixed traffic flow. To further optimize the efficiency of mixed traffic flow, some researchers have proposed dedicated signal phase design schemes in recent years. A literature search revealed that Niroumand et al. (2020) proposed a "white phase" signal design, allowing manually driven vehicles to follow autonomous vehicles in an orderly convoy through intersections (Niroumand R, et al. Joint optimization of vehicle-group trajectory and signal timing: Introducing the white phase for mixed-autonomy traffic stream[J]. Transportation Research Part C - Emerging Technologies, 2020, 116.); Rey et al. (2019) designed a "blue phase" for autonomous vehicles only, achieving inter-vehicle passage through a conflict-point-based separation strategy, without considering intersection signal timing (Rey D, Levin M W. Blue phase: Optimal network traffic control for legacy and autonomous vehicles[J]. Transportation Research Part C - Emerging Technologies, 2020, 116.). B-Methodological, 2019, 130: 105-129.); Wu Wei et al. (2023) proposed setting up dedicated lanes for autonomous driving at intersection entrances and designing dedicated phases based on the conflict point method (Wu Wei, et al. Intersection signal control method considering dedicated phases for autonomous driving [J]. China Journal of Highway and Transport, 2023, 36(10): 183-196.), without considering the driving trajectory of autonomous vehicles on the road segment and the coordinated optimization of signal timing. However, these studies theoretically verified the potential of dedicated phase design in improving the capacity of mixed traffic flow. In addition, the division of intersection lane functions is an important basis for signal timing optimization. Literature shows that setting an automatic driving exclusive lane in the middle lane of the intersection not only can better balance traffic flow and reduce the complexity of lane weaving and merging operations, but also can reduce the contact opportunities between automatic driving vehicles and vulnerable road users such as pedestrians and bicycles, thereby further improving traffic safety (Dai R, et al. Coupling Control of Traffic Signal and Entry Lane at Isolated Intersections Under the Mixed-Autonomy Traffic Environment[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24(10): 10628-10642). Although existing research has provided important references for signal timing optimization for mixed traffic flow, existing methods still have the following shortcomings: Insufficient time and space separation: existing signal control strategies mainly focus on optimization in the time dimension, and lack of separation design in the spatial dimension for human-driven vehicles and automatic driving vehicles, failing to significantly reduce mutual interference between the two types of vehicles. Lack of coordination between trajectory optimization and signal timing: existing researches mainly focus on single dimension of signal timing or vehicle trajectory optimization, without unified modeling and collaborative optimization of the two, making it difficult to maximize the overall traffic efficiency of the intersection. Based on the above background, the present application proposes a method for signal timing and vehicle trajectory optimization at an intersection with mixed human and machine driving, which separates and collaboratively optimizes mixed traffic flow in time and space dimensions by dividing exclusive lanes, setting automatic driving vehicle exclusive signal phases, and combining vehicle trajectory optimization models, aiming to improve intersection traffic efficiency, reduce delay and improve traffic safety, and provide a scientific and effective solution for future urban traffic management. SUMMARY

[0003] Technical problem: After achieving time and space separation of human-driven vehicles and intelligent connected vehicles by setting exclusive signal phases and exclusive lanes, the primary technical problem is how to dynamically allocate the right of way at the intersection based on vehicle arrival information, thereby maximizing the traffic efficiency of the intersection. After determining the right of way of different types of vehicles, the second technical problem is how to make intelligent connected vehicles and human-driven vehicles adjust their driving trajectories under the guidance of the signal control scheme to achieve weighted optimization of energy consumption and traffic efficiency. Technical solution: To solve the above technical problems, a method for signal timing and vehicle trajectory optimization at an intersection with mixed human and machine driving according to the present application comprises the following steps:

[0004] Step 1: Collect relevant information about the intersection, including the number of lanes, existing signal timing scheme, traffic flow and vehicle distribution; divide the intersection into entrance lanes according to lane function, forming two types: dedicated lanes for intelligent connected vehicles and ordinary lanes. The dedicated lanes are only for use by intelligent connected vehicles, while the ordinary lanes are shared by both manually driven vehicles and intelligent connected vehicles. Step 2: Based on the traffic flow characteristics and vehicle behavior features of the intersection, construct a signal timing optimization model adapted to the human-machine mixed driving environment; during the green light phase of the dedicated lane phase, intelligent connected vehicles pass through the intersection in a cooperative manner; during the signal timing phase of the ordinary lane phase, intelligent connected vehicles and manually driven vehicles can pass simultaneously, and the signal timing model aims to minimize delay. Step 3: Based on the optimal signal timing scheme determined in Step 2, optimize the speed trajectory of intelligent connected vehicles to improve throughput and reduce energy consumption; for manually driven vehicles, use an intelligent driving model to calculate speed, acceleration, and displacement parameters in real time; in the speed trajectory optimization calculation of intelligent connected vehicles, combine the driving trajectory information of the preceding vehicle as the input condition for the speed optimization of the following vehicle to ensure safe distance between vehicles and driving safety. Step 4: Based on the real-time traffic flow dynamic data, iteratively update the signal timing and trajectory optimization scheme.

[0005] Step 2 involves constructing a signal timing optimization model, including the following steps: Step 21: Green light time allocation and time discretization; The right-of-way at the intersection is allocated to manually driven vehicles through signal control. The green light time is allocated according to the rules of left turn in the east-west direction, straight in the east-west direction, left turn in the north-south direction, and straight in the north-south direction. The time loss factors such as the full red time are not considered. The time is discretized by using a step size of Δt. Based on the theoretical arrival time t of the vehicle at the intersection i The step length for a vehicle to reach the intersection is calculated using formula (1): k i =t i / Δt (1) In the formula: k i Indicates the time step k at which vehicle i arrives at the intersection; t i This represents the theoretical arrival time of vehicle i at the intersection; Δt represents the duration of the green light corresponding to each distance from the step; Step 22: Calculation of green light phase matrix and vehicle entry time; In the matrix, vehicle i arrives at the intersection in k=4 steps and needs to pass through phase p. i Enter the intersection when the light is green; If phase p is green at step k, then vehicle i corresponds to phase p. i The green light distribution is shown in the third row of the matrix. As shown, the light is green in steps k = 1, 2, 6, and 7, and red in the remaining steps; vehicles stop and wait after reaching the intersection in step 4, and enter the intersection when they encounter the first phase of the green light. ρ i,6 =1, indicating that the vehicle enters the intersection at k=6; θ i,6 =1 indicates that the vehicle leaves the intersection in steps k=6; Vehicle i enters the intersection step z i Calculated using the following formula (2): In the formula: z i This indicates the step in which vehicle i enters the intersection; ρ i,k For a binary variable, ρ i,k =1 indicates that vehicle i enters the intersection in k steps; otherwise... When a vehicle arrives at the intersection, it may need to stop and wait; it may not be able to enter directly. For example, if a vehicle arrives at the intersection in step 4, it may only be allowed to enter in step 6, i.e., k. i =4, ρ i,6 =1; After the vehicle enters the intersection in step 6, it leaves the intersection in every subsequent step, that is... To ensure the right-of-way for vehicles, the constraints are as shown in formulas (3)-(4): In the formula: p i This indicates that vehicle i enters the intersection from phase p; K represents the maximum step size; I represents the set of vehicles, i∈I; For binary variables, This indicates that phase p is green in step k; otherwise... In signal timing, only one phase is green at any given time, while all other phases are red, as shown in constraint (5): In the formula: P represents the set of signal phases, p∈P; To ensure that all vehicles have the right of way, there must be at least one green light in each phase throughout all steps, as shown in constraint (6): Vehicle i leaves the intersection at step z i Step θ i,k represents whether vehicle i leaves the intersection, then z i after θ i,k all = 1, as shown in the 5th row of the matrix; constraints of formulae (7)-(8) are satisfied: In the formula: θ i,k is a binary variable, θ i,k = 1 indicates that vehicle i leaves the intersection at step k, otherwise M represents a large positive integer, taking the value 9999; The green time of each phase needs to satisfy the constraint of maximum green time g max and minimum green time g min , which is converted to successive k in (9) is not less than g min / Δt, and is not greater than g max / Δt, as shown in formula (9): In the formula: g max and g min represent the maximum green time and the minimum green time, respectively; is a binary variable, indicating that the phase p in which vehicle i is located is green at step k, otherwise x represents the number of steps included in the maximum green and minimum green time, the number of green steps; j and a represent the start and end of the step, respectively; Step 23: The delay d i,k of vehicle i at step k is calculated by formulae (10)-(12): Step 24: The objective is to minimize the delay of all vehicles at the intersection, as shown in formula (13): min∑ i∈I d i (13) In the formula: d i represents the delay of vehicle i.

[0006] The vehicle speed trajectory optimization method on the link in step 3 includes the following steps: Step 31: Calculation of the speed trajectory of the manually driven vehicle; The speed trajectory of the manually driven vehicle is calculated using an intelligent driving model, and the acceleration a of the manually driven vehicle at time t is calculated using formula (14). i (t), s in formula (14) * (v i (t),Δv i,i-1 (t) represents the desired following distance, calculated using formula (15): In the formula: I l,HV This represents the set of manually driven vehicles on lane l; a M This is the vehicle's maximum deceleration. v i (t) represents the speed of vehicle i at time t; v i,d Let i be the desired speed of vehicle i; δ is the acceleration exponent; s i,i-1 (t) represents the distance between vehicle i and vehicle i-1; s * (v i (t),Δv i,i-1 (t) represents the expected following distance between vehicle i and vehicle i-1; Δv i,i-1 (t) represents the speed difference between the front and rear vehicles; a d Decelerate for comfort; T represents the safe headway; The speed of the manually driven vehicle is updated by formula (16), where Δt is the time interval; Step 32: Speed ​​optimization for intelligent connected vehicles; divided into two cases: vehicle entering the intersection and non-stop; when stopping is required, the timing of deceleration initiation, magnitude of deceleration, duration of deceleration, and time to reach the intersection; establishing constraints for the decision variables: Vehicle i in t i,1 To t i,2 Time period, t i,3 To t i,4 The time intervals are respectively decelerated by a i,1 and a i,2 The vehicle is driven at reduced speed, and the speed formula is shown in equations (17)-(18). In the formula: I l,CAV This represents the set of intelligent connected vehicles on lane l; c i,1 denotes the speed after the first deceleration; c i,2 denotes the speed after the second deceleration; a i,1 denotes the acceleration in the first deceleration phase; a i,2 denotes the acceleration in the second deceleration phase; The range of values for the speed and deceleration is shown in equations (19) - (20): wherein a i,1 ,a i,2 denotes the magnitude of the deceleration of vehicle i in the two deceleration phases, without the negative sign; v i,3 denotes the speed at which the vehicle reaches the stop line at the intersection after starting from a standstill on the road section; a m is the maximum deceleration of the vehicle, without the negative sign; v m , V m denote the minimum and maximum speeds at which the vehicle can travel on the road section, respectively, and the vehicle cannot stop in the middle of the road section; The relationships at different times are shown in equation (21): t i,0 ≤ t i,1 ≤ t i,2 ≤ t i,3 ≤ t i,4 ≤ t i,5 ≤ t i,6 (21) wherein t i,0 , t i,1 , t i,2 , t i,3 , t i,4 , t i,5 , t i,6 denote the key time points of vehicle i in the speed control region: t i,0 : time of entering the speed control region; t i,1 : start time of deceleration a i,1 t i,2 : end time of deceleration a i,1 t i,3 : start time of deceleration a i,2 t i,4 : end time of deceleration a i,2 ​​​​ t i,5 : the vehicle is ready to leave the stop state; t i,6 : the time of arrival at the intersection; the longest travel time t i,M and the shortest travel time t i,m of the vehicle on the road section are calculated by equations (22)-(23): where L is the length of the speed control area; t i,M and t i,m represent the longest travel time and the shortest travel time of vehicle i, respectively; The time of arrival of vehicle i at the intersection is constrained by the time of arrival of the preceding vehicle t i-1,6 and t i,M , the constraint condition is shown in equation (24), and T is the safe headway; The speed v i (t) and displacement s i (t) of the intelligent connected vehicle i at time t are calculated by equations (25)-(26): where v i,1 and v i,2 represent the vehicle speed after the end of different deceleration stages, respectively; Step 33: When calculating the speed trajectory of vehicle i, the speed and displacement of the preceding vehicle i-1 are known conditions as inputs; when the vehicle speed of the preceding vehicle i-1 at time t is less than that of vehicle i, i.e., v i-1 (t) < v i (t), there is a possibility of rear-end collision; therefore, vehicle i and i-1 are required to maintain a safe headway at time t, and a safety constraint T i,i-1,t is constructed, which is calculated by equation (27): where T i,i-1,t represents the safe time interval between the preceding and following vehicles; The vehicle in the control area is not allowed to overtake, and the displacement of the preceding vehicle is greater than that of the following vehicle at the same time, as shown in equation (28): where s0 is the minimum headway to ensure the safety distance between vehicles; Step 34: The fuel consumption of the vehicle is calculated based on the energy balance formula, where the total power demand P i,total(t) is calculated by formula (29): P i,total (t) = P i,rolling (t) + P i,aero (t) + P i,grade (t) + P i,accel (t) (29) where: P i,total (t) represents the total power demand of vehicle i at time t, unit: watt; Rolling resistance power P i,rolling (t) in formula (29) is calculated by formula (30): P i,rolling (t) = f r ·m·g·v i (t) (30) where: P i,rolling (t) represents the power of rolling resistance; f r represents the rolling resistance coefficient, which is considered constant on a section of road; m represents the vehicle mass, unit: kilograms; g represents the acceleration of gravity, taking the value of 9.81 m / s 2 ; v i (t) represents the speed of vehicle i, unit m / s; Air resistance power P i,aero (t) in formula (29) is calculated by formula (31): P i,aero (t) = 0.5·p·C d ·A·v i (t) 3 (31) where: p represents air resistance, unit: kg / m 3 ; C d represents the wind resistance coefficient; A represents the vehicle frontal area, unit m 2 ; Slope resistance power P i,grade (t) in formula (29) is calculated by formula (32): P i,grade (t) = m·g·sin(θ)·v i (t) (32) where: P i,grade (t) represents the power of slope resistance; θ represents the road slope angle, unit: radian; Acceleration power P i,accel (t) in formula (29) is calculated by formula (33): Pi,accel (t) = m a i (t) v (33) wherein a i (t) represents the acceleration of vehicle i, with unit m / s 2 ; vehicle i fuel consumption rate FC i (t) is converted from power demand to fuel demand, as shown in equation (34): wherein η represents engine efficiency; LHV represents fuel low heat value, with unit J / kg; vehicle time period [t i,0 ,t i,6 ] from entering the control area to the stop line i Total fuel consumption FC i is calculated by integrating the instantaneous fuel consumption rate, as shown in equation (35): Step 35: Optimize the driving trajectory of the intelligent connected vehicle on the road section with the fuel consumption and efficiency weighted minimum as the target, and the objective function is shown in equation (36): min (ω FC i +(1-w) t i,6 ) (36) wherein ω represents the weight between fuel consumption and efficiency.

[0007] Advantages: Compared with the prior art, the present application has the following advantages:

[0008] The present application aims at the mixed driving scene of intelligent connected vehicles and manually driven vehicles, and proposes to reduce the mutual interference between the two types of vehicles by designing special lanes and special signal phases, so as to fully exert the technical advantages of intelligent connected vehicles. Compared with the existing method which generally executes unified control strategy for the two types of vehicles, the present application can effectively solve the problem that the existing scheme fails to fully utilize the technical advantages of intelligent connected vehicles. In addition, the present application combines the intelligent driving model and the fuel consumption calculation formula based on energy balance to establish a road section trajectory optimization model of intelligent connected vehicles, so as to achieve the coordinated optimization of trajectory optimization and signal timing, and further realize the maximization of overall traffic efficiency at the intersection under different intelligent connected vehicle penetration rates. BRIEF DESCRIPTION OF DRAWINGS

[0009] Figure 1 is the flowchart of the method of the present application;

[0010] Figure 2 is the intersection schematic diagram of the method of the present application.

[0011] Figure 3Trajectory optimization schematic diagram for road section of the method of the present application.

[0012] Figure 4 Trajectory diagram of left-turn vehicles in the south entrance direction on the road section under different intelligent connected vehicle penetration rates in the example.

[0013] Figure 5 Trajectory diagram of intelligent connected vehicles on the road section under the condition of a penetration rate of 30% in the example. DETAILED DESCRIPTION

[0014] The technical solutions of the present application are described in detail below in combination with the drawings and examples:

[0015] Figure 4 Figs. (a)-(d) respectively represent the vehicle trajectory diagrams under the conditions of intelligent connected vehicle penetration rates of 0%, 40%, 60%, and 100%. The dashed lines in the diagrams represent the trajectories of intelligent connected vehicles, and the solid lines represent the trajectories of manually driven vehicles. The effects of the speed optimization model are analyzed through different automatic driving vehicle penetration rates. In Fig. (a), all vehicles are manually driven, and their trajectories are calculated through the intelligent driving model, in which the solid lines represent the trajectory paths. In Fig. (b), vehicles 11, 41, and 58 are intelligent connected vehicles, and their speeds are optimized. However, vehicle 11 needs to stop and wait regardless of the speed adjustment because its arrival time is limited by the green light time, so its speed is optimized through the stopping model. Vehicles 41 and 58 can enter the intersection without stopping by reducing their speeds. In Figs. (c) and (d), vehicles 34 and 21 respectively achieve no-stopping through the intersection by speed optimization. Comparing Fig. (a) with Fig. (d), in Fig. (a), vehicle 72 must slow down because the vehicle in front of it needs to stop before starting to pass through the intersection. However, after applying speed optimization, the speed of vehicle 72 is no longer affected by the start-stop of the vehicle in front.

[0016] Figure 5 Trajectory diagram of intelligent connected vehicles on the road section under the condition of a penetration rate of 30% on the dedicated lane, wherein Figure 5 It can be seen that the trajectories of intelligent connected vehicles do not overlap, indicating that the speed optimization model can ensure the safety of vehicles on the road section. Secondly, vehicles adjust their speeds to enter the intersection as soon as possible at the beginning of the dedicated phase green light, and the times at which different vehicles arrive at the stop line are similar, such as vehicles 44, 59, and 67, which arrive at the intersection between 44s and 47s.

[0017] A bidirectional four-lane flat cross intersection is used for the example. The traffic volume of each direction follows a Poisson distribution, and the traffic volume is 720 pcu / h / l. In the signal timing parameters, according to the “Urban Road Intersection Planning Specification”, the minimum green light duration g min is set to 20s, and the maximum green light duration gmax Set to 45s. Delta t = 1s, the maximum deceleration of the vehicle in the intelligent driving model a M = 3 m / s 2 Acceleration index delta = 4, comfortable deceleration a d = 2.8 m / s 2 Minimum vehicle headway s0 = 1.6s; vehicle expected speed 60km / h. The solving model uses a desktop computer equipped with Win11-64-bit system, the processor is Intel(R) Core(TM) i9-14900K 3.20GHz, and the RAM is 32.0GB memory.

[0018] By optimizing the trajectory of the intelligent connected vehicle on the road section and calculating the trajectory of the manually driven vehicle, the trajectory data of each vehicle per second is obtained, and part of the vehicle data under different penetration rates is selected for display, as shown in the following table: Table 1 Partial vehicle trajectory data table under penetration rate 0% Table 2 Partial vehicle trajectory data table under penetration rate 40%

[0019] The present application is not limited to this single example; any changes, modifications, substitutions, combinations, simplifications made by deviating from the spirit and principles of the present application are equivalent replacements, and are included in the protection scope of the present application.

Claims

1. A method for intersection signal timing and vehicle trajectory optimization for human-machine mixed driving, characterized in that, The method comprises the following steps: Step 1: Collecting intersection-related information, including the number of lanes, the existing signal timing scheme, traffic flow and vehicle distribution; dividing the intersection into import lanes according to the lane function to form two types of intelligent network connection special lanes and ordinary lanes, wherein the special lanes are only for intelligent network connection vehicles, and the ordinary lanes are for manual driving vehicles and intelligent network connection vehicles; Step 2: Based on the traffic flow characteristics and vehicle behavior characteristics of the intersection, a signal timing optimization model suitable for the human-machine mixed driving environment is constructed; in the green light stage of the special lane phase, the intelligent network connection vehicles pass through the intersection in a mutually cooperative manner; in the signal timing stage of the ordinary lane phase, the intelligent network connection vehicles and the manual driving vehicles can pass through at the same time, and the signal timing model takes the minimum delay as the target; Step 3: According to the optimal signal timing scheme determined in step 2, the speed trajectory of the intelligent network connection vehicle is optimized to improve the passing efficiency and reduce the energy consumption; for the manual driving vehicle, an intelligent driving model is used to calculate the speed, acceleration and displacement travel parameters in real time; in the speed trajectory optimization calculation of the intelligent network connection vehicle, the trajectory information of the preceding vehicle is combined as the input condition of the speed optimization of the following vehicle to ensure the safe distance and driving safety between vehicles; Step 4: Based on the real-time collected dynamic traffic flow data, the signal timing and trajectory optimization scheme are iteratively updated.

2. The method of claim 1, wherein the method is characterized by: The signal timing optimization model constructed in step 2 comprises the following steps: Step 21: Green light time allocation and time discretization; the intersection passing right of the manual driving vehicle is allocated through signal control, the green light time is allocated according to the rules of east-west left turn, east-west straight, south-north left turn and south-north straight, the time loss factor such as all-red time is not considered, and the time is discretized by using Δt step length; According to the vehicle theory arrival time t at the intersection i The step length of the vehicle arriving at the intersection is calculated by formula (1): k i = t i / Δt (1) In the formula, Δt represents the green light duration corresponding to each discrete step length; k i denotes the time step at which vehicle i arrives at the intersection; t i denotes the time at which vehicle i is supposed to reach the intersection; In the formula, Δt represents the green light duration corresponding to each discrete step length; Step 22: Green phase matrix and vehicle entering time calculation; in the matrix, vehicle i arrives at the intersection at k = 4 step and needs to pass through phase p i to enter the intersection through the green light of phase p ; represents the green light of phase p at the kth step, then the green light distribution of vehicle i corresponding to phase p i is shown in the 3rd row of the matrix , which is green at k = 1, 2, 6, 7 steps and red at the rest of the steps; after vehicle i arrives at the intersection at the 4th step, it stops and waits, and enters the intersection when it encounters the first green phase, p i,6 = 1, indicating that the vehicle enters the intersection at k = 6; p i,6 = 1 indicates that the vehicle leaves the intersection at k = 6 step; Entry into intersection step z of vehicle i i This is calculated by the following equation (2). In order to ensure the vehicle passing right, the constraint conditions are shown in formulas (3)-(4): z i represents the step in which the vehicle i enters the intersection; ρ i,k is a binary variable, ρ i,k = 1 if vehicle i enters the intersection at step k, otherwise ρ i,k = 0, k ∈ [1,..., K]; The vehicle can need to stop and wait when it reaches the intersection and can not be able to enter directly, for example, the vehicle reaches the intersection at step 4 and is allowed to enter at step 6, i.e. k i = 4, p i,6 = 1; after the vehicle enters the intersection at step 6, the vehicle leaves the intersection at each subsequent step, i.e. θ i,k = 1, In the formula, K represents the maximum step length; I represents a vehicle set, i∈I; p i denotes that vehicle i enters the intersection from phase p; In the signal timing, only one phase is green at the same time, and other phases are red, as shown in constraint (5): In the formula, P represents a set of signal phases, p∈P; is a binary variable, denotes that the kth step phase p is green, otherwise p∈P; In order to ensure that all vehicles have the passing right, at least one green light of each phase in all steps, as shown in constraint (6): In the formula, M represents a large positive integer, and the value is 9999; In the formula, j and a represent the start and end of the step length respectively; Vehicle i enters intersection z i Step θi after vehicle i enters intersection z i,k Indicates whether vehicle i leaves intersection z, then θi i after vehicle i leaves intersection z i,k Both = 1, as shown in the 5th row of the matrix; satisfy the constraints of formulas (7)-(8): Step 24: Taking the minimum delay of all vehicles at the intersection as the target, as shown in formula (13): θ i,k is a binary variable, θ i,k = 1 if vehicle i leaves the intersection at step / , otherwise θ i,k = 0, k e [1,..., K] ; The vehicle speed trajectory optimization method on the road section in step 3 comprises the following steps: The green light time of each phase needs to meet the constraint of maximum green light time g max and minimum green light time g min , which is converted into k in succession in the middle is not less than g min / Δt, and is not greater than g max / Δt, as shown in formula (9): In the formula, δ is an acceleration index; g max and g min respectively represent the maximum green light time and the minimum green light time; is a binary variable, represents the kth step of the phase p in which the vehicle i is located, otherwise p∈P; x represents the number of steps included in the maximum green and minimum green time, the number of steps of the green light duration; T is a safe vehicle head time; Step 23: Vehicle i's delay d at step k i,k calculated from equations (10)-(12). The speed of the manual driving vehicle is updated by formula (16), and Δt is the time interval; min∑ i∈I d i (13) In the formula: d i denotes the delay of the vehicle i.

3. The method of claim 1, wherein the method is characterized by: Step 32: Speed optimization of the intelligent network connection vehicle; divided into two cases of entering the intersection and not stopping; when stopping is needed, the time when deceleration starts, the deceleration size, the deceleration duration and the time of reaching the intersection; Step 31: Manual driving vehicle speed trajectory calculation; the speed trajectory of the manual driving vehicle is calculated by using an intelligent driving model, and the acceleration a of the manual driving vehicle at time t is calculated by formula (14) i s * (v i (t),Δv i,i-1 (t)) is the expected following distance, which is calculated by formula (15): Constraint conditions are established for the decision variables: I t,HV denotes the set of human-driven vehicles on lane / ; a M maximum deceleration of the vehicle; v i (t) is the speed of vehicle i at time t; v i,d is the desired vehicle speed for vehicle i; In the formula, V represents the speed of the vehicle; s i,i-1 (t) is the vehicle distance of vehicle i from vehicle i-1; s * (v i (t),Δv i,i-1 (t)) is the desired following distance of vehicle i to vehicle i-1. Δv i,i-1 (t) represents the front-rear vehicle speed difference; a d comfort deceleration; The value range of the speed and the deceleration is shown in formulas (19)-(20): In the formula, ​ ​ Vehicle i at t i,1 to t i,2 Time period, t i,3 to t i,4 Decelerate at deceleration a i,1 and a i,2 , respectively, with speed formula as shown in equations (17)-(18); ​ I l,CAV denotes a set of intelligent connected vehicles on the lane l; v i,1 V1 represents the speed after the first deceleration; v i,2 V2 represents the speed after the 2nd deceleration; a i,1 a represents the acceleration of the first deceleration phase; a i,2 a represents the acceleration of the 2nd deceleration phase; ​ ​ a i,1 ,a i,2 denotes the magnitude of the deceleration of vehicle i at the two deceleration phases, without the negative sign; v i,3 Vp represents the speed of a vehicle after stopping at a stop line of an intersection; a M maximum deceleration of the vehicle, without the sign; v m , V m respectively represent the minimum and maximum vehicle speed at which the vehicle can travel on the road segment, without stopping in the middle of the road segment; The relationship at different times is shown in equation (21): t i,0 ≤t i,1 ≤t i,2 ≤t i,3 ≤t i,4 ≤t i,5 ≤t i,6 (21) in which: t i,0 , t i,1 , t i,2 , t i,3 , t i,4 , t i,5 , t i,6 denote the key time points of vehicle i in the speed control zone: t i,0 : time of entering the speed control region; t i,1 : deceleration a i,1 start time; t i,2 : deceleration a i,1 end time; t i,3 : deceleration a i,2 start time; t i,4 : deceleration a i,2 end time; t i,5 : the vehicle is ready to leave the parked state; t i,6 : time of arrival at the intersection; the longest travel time t of the vehicle on the road segment i,M and the shortest travel time t i,m calculated by equations (22)-(23) In the formula: L is the length of the speed control area; t i,M and t i,m respectively represent the longest and shortest travel time of vehicle i; The time of arrival of vehicle i at the intersection is given by the time t i-1,6 and t i,M constraints, the constraints are given by equation (24), where T is the safe headway time. a speed v of the intelligent connected vehicle i at a time t i (t) and a displacement s i (t) are calculated by equations (25)-(26): wherein: v i,1 , v i,2 denote the vehicle speed after the end of the respective deceleration phase. Step 33: When calculating the speed trajectory of vehicle i, the speed and displacement of the preceding vehicle i-1 are known conditions as input; when the speed of the preceding vehicle i-1 at time t is less than that of vehicle i, that is, v i-1 (t) < v i (t), there is a possibility of being rear-ended by the following vehicle; therefore, vehicle i and i-1 are required to maintain a safe headway at time t, and a safety constraint T i,i-1,t is calculated by formula (27): In the formula: T i,i-1,t represents the safety time interval of the front and rear vehicles; Vehicles in the control area are not allowed to overtake, and the displacement of the front vehicle is greater than that of the rear vehicle at the same time, as shown in equation (28): In the formula: s0 is the minimum vehicle headway, which ensures the safety distance between vehicles; Step 34: Calculate the fuel consumption of the vehicle based on the energy balance based fuel consumption formula, where the total power demand P i,total (t) is calculated from equation (29) P i,total (t) = P i,rolling (t) + P i,aero (t) + P i,grade (t) + P i,accel (t) (29) where: P i,total (t) denotes the total power demand of vehicle i at time t, in Watts; Rolling resistance power P in formula (29) i,rolling (t) is calculated from formula (30): P i,rolling (t) = f r • m • g • v i (t) (30) where: P i,rolling (t) represents the power of the rolling resistance; f r denotes the rolling resistance coefficient, considered constant over a stretch of road; m represents the mass of the vehicle, in kilograms; g represents the acceleration due to gravity, which has a value of 9.81 m / s 2 ; v i (t) denotes the speed of vehicle i in m / s; The air resistance power P in equation (29) i,aero (t) is calculated from equation (31) P i,aero (t) = 0.5 - p - C d - A - v i (t) 3 (31) wherein: p represents the air resistance, in kg / m 3 ; C d represents the wind resistance coefficient; A represents the vehicle frontal area in m2 2 ; The grade resistance power P in equation (29) i,grade (t) is calculated from equation (32): P i,grade (t) = m - g - sin (0) - v i (t) (32) where: P i,grade (t) represents the power of the slope resistance; θ represents the road slope angle, in radians; The acceleration power P in equation (29) i,accel (t) is calculated from equation (33) P i,accel (t) = m · a i (t) · v (33) wherein: a i (t) represents the vehicle acceleration in m / s 2 ; Vehicle i fuel consumption rate FC i (t) Conversion from power demand to fuel demand as shown in equation (34): In the formula: η represents the engine efficiency; LHV represents the low heat value of fuel, in J / kg; The period of time [t i,0 ,t i,6 ] from the vehicle entering the control area to the stop line i The total fuel consumption FC i is calculated by integrating the instantaneous fuel consumption rate, as shown in equation (35): Step 35: Optimize the driving trajectory of the intelligent connected vehicle on the road section with fuel consumption and efficiency weighted minimum as the target, the objective function is shown in equation (36): min(ω · FC i +(1 - w) · t i,6 ) (36) In the formula: ω represents the weight between fuel consumption and efficiency.

Images