Coordinated optimization method and system for dynamic variable lanes at entrance and exit of intersection in mixed driving environment of automatic driving and manual driving

Abstract

The invention discloses an intersection entrance and exit dynamic variable lane coordination optimization method and system in an automatic driving and manual driving mixed driving environment, and the method comprises the steps: setting a left-turn entrance lane at the left side of an exit lane, and enabling left-turn and opposite straight vehicles to pass at the same time; and setting entrance and exit dynamic variable lanes, enabling a lane of an entrance lane of a left-turn motorcade to be used as an exit lane of an opposite straight motorcade at the same time, and obtaining the functions of each lane at the intersection, the stop line position of a conflict lane and the number of vehicles in each lane by solving a nonlinear programming model with the minimum intersection passing time as a target. The left-turn entrance lane is arranged on the left side of the exit lane, so that left-turn and opposite straight-going motorcades can be simultaneously released in the same phase, the number of conflicts and phases is reduced, and the waiting time of vehicles is shortened. Meanwhile, an entrance and exit dynamic variable lane is arranged, so that a left-turn motorcade can enter the intersection through the lane, an opposite straight lane can leave the intersection through the lane, and efficient utilization of lane resources is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent transportation and automatic driving, in particular to a method for coordinating and optimizing dynamic variable lanes at the entrance and exit of an intersection in an automatic driving and manual driving mixed environment. BACKGROUND

[0002] The traffic time at the intersection is greatly affected by the lane function allocation. Unreasonable lane allocation often causes more vehicle delays and traffic conflicts. Traditional traffic management methods rely on fixed lane allocation and signal control, and do not consider the differences between automatic driving vehicles in an automatic driving and manual driving mixed environment, so that the advantages of automatic driving cannot be effectively utilized, and even more traffic problems may occur.

[0003] In the prior art, in order to reduce the conflict and delay between straight vehicles and left-turn vehicles, straight phases and left-turn phases are usually set to release different traffic flows in turn. Although this approach effectively avoids the conflict between straight vehicles and left-turn vehicles, it also leads to an increase in signal cycle and frequent phase switching, thereby increasing the waiting time of vehicles and reducing the traffic efficiency of the intersection. In order to solve this problem, some studies have proposed a method of transferring left-turn entrance lanes to the left side of the exit lane to optimize traffic organization without increasing the waiting time. This method reorganizes the road section so that left-turn vehicles can be released at the same time as opposite straight vehicles, thereby reducing signal phase conflicts, shortening signal cycles and improving intersection traffic efficiency. However, this fixed lane allocation-based solution lacks flexibility, and when the straight and left-turn traffic demands are unbalanced, it is easy to cause excessive saturation of some lanes or waste of lane resources.

[0004] In addition, in a traffic environment where automatic driving vehicles and manual driving vehicles are mixed, the operating characteristics of the two types of vehicles are significantly different. CN119495201A discloses a method for coordinating and controlling intersection lane division and signal timing, which determines a recommended lane function division scheme and signal timing scheme according to the traffic demand of each direction of the intersection, and interacts information through multiple sets of lane function adjustment signs and lane-level signal lights. However, this approach adjusts the signal light duration when changing the function of each lane, and requires different traffic flows to adjust the vehicle order upstream through trajectory algorithms, which increases the computational load of the algorithm. The setting of automatic driving dedicated phases increases the signal phases, thereby prolonging the signal cycle and increasing the waiting time of vehicles at the intersection. Reusing lanes, left-turn borrowed lanes are only allowed to be used by automatic driving vehicles and need to be used with automatic driving dedicated lanes, so that only automatic driving vehicle fleets are optimized, and manual driving vehicle fleets or mixed fleets of automatic driving and manual driving vehicles are not optimized, which leads to a lack of significant optimization effect and waste of road resources at low automatic driving penetration rates. SUMMARY

[0005] The application aims to provide an intersection import / export dynamic variable lane coordination optimization method and system in an automatic driving and manual driving mixed driving environment, which can reduce the number of phases, reduce the waiting time of vehicle fleet, and improve the utilization rate of lane.

[0006] The application discloses an intersection import / export dynamic variable lane coordination optimization method in an automatic driving and manual driving mixed driving environment, which sets a left turn import lane on the left side of an export lane, and releases left turn vehicles and opposite straight vehicles at the same time; sets an import / export dynamic variable lane, so that the lane of the left turn vehicle fleet import lane is used as the export lane of the opposite straight vehicle fleet at the same time, and a dynamic lane distribution and passing scheme is obtained by solving a nonlinear programming model.

[0007] The nonlinear programming model takes minimizing intersection passing time as an objective, and the dynamic lane distribution and passing scheme includes the function of each lane of the intersection, the stop line position of the conflict lane, and the number of vehicles in each lane.

[0008] Further, the calculation method of the intersection passing time is as follows:

[0009] The vehicle fleet passing time of each lane in two opposite directions is calculated respectively, and the maximum value of the vehicle fleet passing time is the intersection passing time.

[0010] Further, the nonlinear programming model is subjected to constraint conditions, including:

[0011] Any lane in the same direction cannot be used as the import lane of straight lane and left turn lane at the same time;

[0012] The opposite lane of the straight lane cannot be the straight lane;

[0013] At least one straight lane and one left turn lane are allocated in any direction;

[0014] The import lane for left turn is not more than the number of corresponding export lanes in the adjacent direction;

[0015] The total number of vehicles allocated to each lane in any direction meets the vehicle quantity demand in the direction;

[0016] The straight vehicle fleet enters the import / export dynamic variable lane after the opposite left turn vehicle fleet leaves the conflict point;

[0017] The left turn lane is allocated first and then the straight lane in the same direction;

[0018] The lane without allocated lane function cannot be allocated with vehicles.

[0019] Further, the genetic algorithm is used to solve the nonlinear programming model.

[0020] The functions of each lane of the intersection and the number of vehicles in each lane are encoded by integers, and the position of the stop line of the conflict lane is encoded by a floating point number.

[0021] Further, in the genetic algorithm, the reciprocal of the intersection passing time is used as the fitness evaluation index.

[0022] Further, in the crossover and mutation operations of the genetic algorithm,

[0023] Single-point crossover and random reset mutation are performed on integer variables encoded by integers;

[0024] Simulated binary crossover and polynomial mutation are performed on continuous variables encoded by floating point numbers.

[0025] The intersection entrance and exit dynamic variable lane coordination optimization system in the automatic driving and manual driving mixed driving environment sets the left turn entrance lane on the left side of the exit lane, and releases the left turn and the opposite straight vehicles at the same time; the dynamic variable lane of the entrance and exit is set, so that the lane of the left turn vehicle team entrance is used as the exit lane of the opposite straight vehicle team at the same time.

[0026] The system includes a model establishment and solving unit, and a nonlinear programming model with the objective of minimizing the intersection passing time, and a dynamic lane allocation and passing scheme obtained by solving the nonlinear programming model, wherein the dynamic lane allocation and passing scheme includes the functions of each lane of the intersection, the position of the stop line of the conflict lane, and the number of vehicles in each lane.

[0027] Further, the calculation method of the intersection passing time is:

[0028] The vehicle team passing time of each lane in two opposite directions is calculated respectively, and the maximum value of the vehicle team passing time is the intersection passing time.

[0029] The nonlinear programming model is subjected to constraint conditions, including:

[0030] Any lane in the same direction cannot be used as the entrance lane of the straight lane and the left turn lane at the same time.

[0031] The opposite lane of the straight lane cannot be the straight lane.

[0032] At least one straight lane and one left turn lane are allocated in any direction.

[0033] The number of entrance lanes for left turns is not more than the number of corresponding exit lanes in the adjacent direction.

[0034] The total number of vehicles allocated to each lane in any direction meets the vehicle quantity demand of the direction.

[0035] Straight traffic platoon enters the import and export dynamic variable lane after the opposite left-turn traffic platoon leaves the conflict point;

[0036] The left-turn lane is allocated first and then the straight lane in the same direction;

[0037] The lane without lane allocation function cannot allocate vehicles.

[0038] The computer readable storage medium of the application stores a computer program, and the computer program is executed by a processor to realize the intersection import and export dynamic variable lane coordination optimization method in the automatic driving and manual driving mixed driving environment.

[0039] The computer program product of the application comprises a computer program, and the computer program is executed by a processor to realize the intersection import and export dynamic variable lane coordination optimization method in the automatic driving and manual driving mixed driving environment.

[0040] Advantages: Compared with the prior art, the application has the following advantages:

[0041] (1) Intersection layout and lane function collaborative innovation. The left-turn import lane is arranged on the left side of the export lane, so that the left-turn and opposite straight traffic platoon can be released at the same phase, reducing the conflict and phase number and reducing the vehicle waiting time. On this basis, the import and export dynamic variable lane is set, so that the left-turn traffic platoon can enter the intersection by using the lane, and the opposite straight lane can leave the intersection by using the lane, realizing efficient use of lane resources.

[0042] (2) Comprehensive coordination mechanism of multi-parameter joint optimization. The application establishes an optimization model with the minimum travel time as the target, and simultaneously optimizes the lane function, vehicle allocation and stop line position. After determining the lane function, the stop line position of the opposite straight lane is dynamically adjusted, and the number of vehicles in each lane is reasonably allocated, so that the left-turn and opposite straight traffic platoon can be safely and efficiently released at the same phase, improving the overall operation efficiency of the intersection.

[0043] (3) Considering the applicability of mixed traffic platoon under different automatic driving penetration rates. The application fully considers the actual characteristics of manual driving and automatic driving mixed traffic platoon under different penetration rates in the optimization process. By introducing the automatic driving penetration rate parameter, the mixed traffic platoon with different automatic driving penetration rates can efficiently pass through the intersection. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 The intersection import and export dynamic variable lane coordination optimization method flowchart of the application.

[0045] Figure 2 The optimization scheme schematic diagram of the embodiment of the application. DETAILED DESCRIPTION

[0046] The application provides an intersection import and export dynamic variable lane coordination optimization method and system in an automatic driving and manual driving mixed environment, which can recombine lanes at an intersection according to characteristics of automatic driving vehicles, shift a left-turn import lane to the left side of an export lane, and enable left-turn and opposite straight driving automatic driving and manual driving mixed vehicle fleets to be released in the same phase; on this basis, an import and export dynamic variable lane is arranged, which can be used as a straight export lane for opposite straight driving vehicle fleets after the left-turn vehicle fleet leaves, and the lane function, the stop line position of a conflict lane and the number of vehicles allocated on each lane are dynamically adjusted according to the number of vehicles entering the intersection to wait for traffic and the automatic driving penetration rate, so that when the left-turn vehicle fleet on the import and export dynamic variable lane leaves the conflict area, the straight driving vehicle fleet can immediately use the import and export dynamic variable lane as an export lane for the straight driving vehicle fleet, realize seamless traffic of vehicle fleets in the mixed environment, and reduce the traffic time.

[0047] The technical solutions of the application will be further described below with reference to the drawings.

[0048] As shown in Figure 1 , the intersection import and export dynamic variable lane coordination optimization method in the automatic driving and manual driving mixed environment comprises the following steps.

[0049] S1, acquiring vehicle fleet information. A roadside unit is used to acquire the number of left-turn and straight driving vehicles entering an intersection in p and r directions , and an automatic driving penetration rate , . The p and r directions are two opposite directions of the intersection, for example, the east-west direction or the south-north direction.

[0050] S2, constructing a mixed integer nonlinear programming model based on the intersection optimization scheme. The traffic time is taken as an optimization target, and the objective function is as follows:

[0051] ; (1)

[0052] ; (2)

[0053] In the formula, is the traffic time of a queued vehicle fleet passing through the intersection, is the traffic time of a vehicle fleet on a p-direction lane, is the traffic time of a vehicle fleet on an r-direction lane.

[0054] ​​The constraints are imposed on the mixed integer nonlinear programming model by equations (3)-(41). Among them, constraints (3) and (4) ensure that the same direction lane cannot be used as both straight lane and left-turn lane entrance; constraint (5) ensures that the opposite lane of straight lane cannot be straight lane; constraints (6)-(9) ensure that at least one straight lane and one left-turn lane are allocated for each direction; constraints (10) and (11) ensure that the number of left-turn entrance lanes cannot be more than the number of exit lanes of adjacent lanes, for example, the number of west left-turn entrance lanes cannot be more than the number of north exit lanes; constraints (12)-(15) ensure that the allocated vehicles meet the vehicle demand of each direction; constraints (16) and (17) ensure that the straight vehicle platoon enters the dynamic variable lane after the opposite left-turn vehicle platoon leaves the conflict point, wherein the time of vehicle platoon arriving at and leaving the conflict point is calculated according to equations (18)-(33); constraints (34) and (35) ensure that the left-turn lane is allocated before the straight lane to avoid the conflict between straight and left-turn vehicle platoons of the same direction; constraints (36)-(39) ensure that the lane without corresponding lane function cannot be allocated with corresponding vehicles; constraint (40) ensures that the number of allocated vehicles is a positive integer; and constraint (41) ensures that the distance of stop line is within a reasonable range.

[0055] (3)

[0056] (4)

[0057] (5)

[0058] (6)

[0059] (7)

[0060] (8)

[0061] (9)

[0062] (10)

[0063] (11)

[0064] (12)

[0065] (13)

[0066] (14)

[0067] ；（15）

[0068] ；（16）

[0069] ；（17）

[0070] ；（18）

[0071] ；（19）

[0072] ；（20）

[0073] ；（21）

[0074] ；（22）

[0075] ；（23）

[0076] ；（24）

[0077] ；（25）

[0078] ；（26）

[0079] ；（27）

[0080] ；（28）

[0081] ；（29）

[0082] ；（30）

[0083] ；（31）

[0084] ；（32）

[0085] ；（33）

[0086] ；（34）

[0087] ；（35）

[0088] ；（36）

[0089] ；（37）

[0090] ; (38)

[0091] ; (39)

[0092] ; (40)

[0093] ; (41)

[0094] wherein: is a binary variable, 1 indicates that the p-direction lane is assigned to left-turn as a left-turn approach, 0 indicates not to be assigned; is a binary variable, 1 indicates that the p-direction lane i is assigned to straight as a straight approach, 0 indicates not to be assigned, the lanes are numbered from left to right as 1, 2, …, i, …, N. is a binary variable, 1 indicates that the r-direction lane is assigned to left-turn as a left-turn approach, 0 indicates not to be assigned; is a binary variable, 1 indicates that the r-direction lane is assigned to straight as a straight approach, 0 indicates not to be assigned, the lanes are numbered from left to right as 1, 2, …, j, …, N. G is the number of exit approaches in the vertical direction. and are the number of queued vehicles of the p-direction lane and the r-direction lane , respectively. is the time for the first vehicle of the p-direction i-th lane to arrive at the conflict point with the r-direction j-th lane; is the time for the first vehicle of the r-direction j-th lane to arrive at the conflict point with the p-direction i-th lane; is the time for all vehicles of the p-direction i-th lane to leave the conflict point with the j-th lane; is the time for all vehicles of the r-direction j-th lane to leave the conflict point with the i-th lane. is the start time of the k-th vehicle of the p-direction i-th lane; . is the start time of the k-th vehicle of the r-direction j-th lane, . is the time for the k-th vehicle of the p-direction i-th lane to arrive at the conflict point with the r-direction j-th lane after starting, . is the time for the k-th vehicle of the r-direction j-th lane to arrive at the conflict point with the p-direction i-th lane after starting, . is whether the first vehicle of the p-direction i-th lane is an autonomous vehicle, 1 is yes, 0 is no; is the first vehicle in the r-direction j-lane, 1 is yes, and 0 is no. and are the reaction times of the autonomous vehicle and the human-driven vehicle, respectively. is the distance between the kth vehicle in the p-direction i-lane and the conflict point of the r-direction j-lane, . is the distance between the kth vehicle in the r-direction j-lane and the conflict point of the p-direction i-lane, . is the distance between the left-turn platoon entrance of the p-direction i-lane and the conflict point of the conflict of the straight-going platoon of the r-direction j-lane, is the distance between the left-turn platoon entrance of the r-direction j-lane and the conflict point of the conflict of the straight-going platoon of the p-direction i-lane, is the distance between the straight-going platoon entrance of the p-direction i-lane and the conflict point of the conflict of the left-turn platoon of the r-direction j-lane, is the distance between the straight-going platoon entrance of the r-direction j-lane and the conflict point of the conflict of the left-turn platoon of the p-direction i-lane. is the acceleration distance of the kth vehicle in the p-direction i-lane, . is the acceleration distance of the kth vehicle in the r-direction j-lane, . is the acceleration time of the kth vehicle in the p-direction i-lane, . is the acceleration time of the kth vehicle in the r-direction j-lane, . and are the distances from the stop line of the p-direction i-lane and the r-direction j-lane to the intersection, respectively, is the maximum distance of the stop line. is the headway, is the vehicle acceleration, , are the left-turn and straight-going speeds, and M is the penalty coefficient.

[0095] The travel time of the platoon through the intersection is calculated by equations (42)-(47).

[0096] ; (42)

[0097] ; (43)

[0098] ; (44)

[0099] ; (45)

[0100] (46)

[0101] (47)

[0102] In the formula: For the i-th lane in the p direction, the first... The time it takes for the vehicle to reach the exit after it starts moving. . For the j-th lane in the r direction, the first... The time it takes for the vehicle to reach the exit after it starts moving. . Let k be the total distance traveled by the k-th vehicle in the i-th lane in the direction p. . Let be the total distance traveled by the k-th vehicle in the j-th lane in the r-direction. . Let p be the distance inside the intersection of the left-turn lane i in the direction of p. Let be the distance inside the intersection of the left-turn lane j in the direction r. This refers to the distance within the intersection for straight-ahead lanes.

[0103] S3 uses a genetic algorithm to solve the mathematical model in S2, and the decision variables (lane function allocation) are used. , , , Vehicle allocation , and the position of the stop line , Real number encoding is performed, where discrete variables ( , , , , , ) uses integer encoding, continuous variables ( , It uses floating-point encoding.

[0104] Initialize the hyperparameters of the genetic algorithm, using the reciprocal of the objective function, 1 / T, as the fitness evaluation index. Randomly generate n feasible solutions as the initial population. Individuals that do not meet the constraints will have their fitness significantly reduced by a penalty function. A tournament selection method is used to select individuals with higher fitness from the current population to enter the next generation. Simulated binary crossover is used for continuous variables, while single-point crossover is used for integer variables, with a crossover probability of [value missing]. For continuous variables, polynomial mutation is used; for integer variables, random reset mutation is used. The mutation probability is set to 1. After crossover and mutation operations, individuals that do not meet the constraints are discarded. This process continues until the number of iterations reaches a certain threshold. When the iteration is stopped, the optimal solution is output.

[0105] S4, the lane function allocation scheme , , , , vehicle allocation scheme , , and stop line position setting , are applied to the intersection. Wherein when , and , the i-th lane in the p direction is a dynamic variable lane for entering and exiting, when , and , the j-th lane in the r direction is a dynamic variable lane for entering and exiting, and the vehicle is allocated to the lane with the corresponding function according to the vehicle allocation scheme, and is queued behind the stop line, and starts from the stop line at the same time when the green light is on.

[0106] The method described in the application is verified by a specific case.

[0107] In this example, a four-way intersection is constructed using SUMO, and the optimization object is the east-west direction lane of the intersection. There are 6 lanes in the east-west direction, and 24 straight vehicles and 16 left-turn vehicles are placed in the intersection in the east-west direction. The automatic driving penetration rate of each direction is set to 50%, the manual driving following model is selected as the IDM following model, and the automatic driving following model is selected as the CACC following model. The number of left-turn and straight vehicles entering the intersection in the east-west direction is obtained by using the roadside unit , , the automatic driving penetration rate = = 0.5, and the penalty coefficient .

[0108] The super parameter settings in the genetic algorithm are shown in Table 1.

[0109] Table 1 Super parameter settings

[0110]

[0111] The solution is obtained , , , , , , , .

[0112] According to the above solving result, the lane function allocation scheme , , , , vehicle allocation scheme , , and stop line position setting , The application is applied to the intersection. That is, the first, second and third lanes from left to right are left-turn lanes for westbound, and 12, 3 and 1 vehicles are allocated respectively, and the fourth and fifth lanes are straight lanes, and 14 and 10 vehicles are allocated respectively. The first, second and third lanes from left to right are left-turn lanes for eastbound, and 12, 3 and 1 vehicles are allocated respectively, and the fourth and fifth lanes are straight lanes, and 14 and 10 vehicles are allocated respectively. The stop lines of the fourth and fifth lanes for westbound need to be moved back by 1.2 meters and 34.6 meters respectively, and the stop lines of the fourth and fifth lanes for eastbound need to be moved back by 1.2 meters and 34.6 meters respectively.

[0113] Since the second and third lanes for westbound are both left-turn entrance lanes for westbound and exit lanes for the straight vehicle team of the fourth and fifth lanes for eastbound, the second and third lanes from left to right for westbound are entrance and exit dynamic variable lanes. Similarly, the second and third lanes from left to right for eastbound are also entrance and exit dynamic variable lanes. As shown in FIG. 4, vehicles are allocated to the corresponding functional lanes according to the vehicle allocation scheme, and are queued behind the stop line, and start from the stop line at the same time when the green light is on. Figure 2

[0114] In this example, the travel time is used as the output index to analyze the optimization effect of setting the entrance and exit dynamic variable lanes at the intersection. In order to more clearly observe the optimization effect of the present application, the simulation experiment compares the optimization scheme proposed by the present application with the traditional intersection. Two simulation scenes are established respectively, and 24 straight vehicles and 16 left-turn vehicles are placed in the east-west direction at the intersection, and the automatic driving penetration rate is set to 50%. Among them, the traditional intersection is provided with two straight lanes and one left-turn lane in the east-west direction. Finally, the travel time of the traditional intersection is 58.9s, and the travel time of the intersection optimized by the present application is 24.8s, and the travel time of the intersection optimized by the present application is reduced by 57.9% compared with the traditional intersection.

[0115] The intersection entrance and exit dynamic variable lane coordination optimization system in the automatic driving and manual driving mixed driving environment sets the left-turn entrance lane on the left side of the exit lane, and releases the left-turn vehicles and the opposite straight vehicles at the same time; the entrance and exit dynamic variable lane is set, so that the lane of the left-turn vehicle team entrance lane is used as the exit lane of the opposite straight vehicle team at the same time;

[0116] The system comprises a model establishing and solving unit, a nonlinear programming model, which takes the minimization of the intersection travel time as the target, and a dynamic lane allocation and travel scheme obtained by solving the nonlinear programming model, wherein the dynamic lane allocation and travel scheme comprises the functions of each lane at the intersection, the stop line positions of the conflict lanes and the number of vehicles on each lane. ​

[0117] The computer readable storage medium described in the application stores a computer program, and the computer program is executed by a processor to realize the intersection import and export dynamic variable lane coordination optimization method in the automatic driving and manual driving mixed environment.

[0118] The computer program product described in the application comprises a computer program, and the computer program is executed by a processor to realize the intersection import and export dynamic variable lane coordination optimization method in the automatic driving and manual driving mixed environment.

[0119] The computer readable storage medium can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, flash memory or any other medium that can be used to store program codes in the form of instructions or data structures and can be accessed by a computer.

[0120] The processor is used to execute the computer program stored in the memory to realize each step in the method related to the above-mentioned embodiments.

Claims

1. An intersection entrance and exit dynamic variable lane coordination optimization method in an automatic driving and manual driving mixed driving environment, characterized in that, The left-turn entry lane is arranged at the left side of the exit lane, and the left-turn and opposite straight vehicles are released simultaneously; the entry and exit dynamic variable lane is arranged, so that the lane of the left-turn vehicle team entry lane is used as the exit lane of the opposite straight vehicle team simultaneously; a dynamic lane distribution and passing scheme is obtained by solving a nonlinear programming model; The nonlinear programming model takes minimizing the intersection passing time as an objective, and the dynamic lane distribution and passing scheme includes the functions of each lane at the intersection, the stop line positions of the conflict lanes, and the vehicle numbers of each lane.

2. The intersection entrance and exit dynamic variable lane coordination optimization method in the mixed driving environment of automatic driving and manual driving according to claim 1, characterized in that, The intersection passing time is calculated by: Respectively calculating the vehicle team passing time of each lane in two opposite directions, and the maximum value of the vehicle team passing time is the intersection passing time.

3. The intersection entrance and exit dynamic variable lane coordination optimization method in the mixed driving environment of automatic driving and manual driving according to claim 2, characterized in that, The nonlinear programming model is subjected to constraint conditions, including: Any lane in the same direction cannot be used as the entry lane of the straight lane and the left-turn lane simultaneously; The opposite lane of the straight lane cannot be the straight lane; At least one straight lane and one left-turn lane are allocated in any direction; The number of entry lanes for left-turn is not more than the number of corresponding exit lanes in the adjacent direction; The total number of vehicles allocated to each lane in any direction meets the vehicle number demand in the direction; The straight vehicle team enters the entry and exit dynamic variable lane after the opposite left-turn vehicle team leaves the conflict point; The left-turn lane is allocated first and then the straight lane in the same direction; The lane without allocated function cannot be allocated with vehicles.

4. The intersection entrance and exit dynamic variable lane coordination optimization method in the mixed driving environment of automatic driving and manual driving according to claim 2, characterized in that, The nonlinear programming model is solved by using a genetic algorithm; The functions of each lane at the intersection and the vehicle numbers of each lane are encoded by integers, and the stop line positions of the conflict lanes are encoded by floating-point numbers.

5. The intersection entrance and exit dynamic variable lane coordination optimization method in the mixed driving environment of automatic driving and manual driving according to claim 4, characterized in that, In the genetic algorithm, the reciprocal of the intersection passing time is used as the fitness evaluation index.

6. The intersection entrance and exit dynamic variable lane coordination optimization method in the mixed driving environment of automatic driving and manual driving according to claim 4, characterized in that, In the crossover and mutation operations of the genetic algorithm, The integer variables encoded by integers are subjected to single-point crossover and random reset mutation; The continuous variables encoded by floating-point numbers are subjected to simulated binary crossover and polynomial mutation.

7. An intersection entrance and exit dynamic variable lane coordination optimization system in an automatic driving and manual driving mixed driving environment, characterized in that, The left-turn entry lane is arranged at the left side of the exit lane, and the left-turn and opposite straight vehicles are released simultaneously; the entry and exit dynamic variable lane is arranged, so that the lane of the left-turn vehicle team entry lane is used as the exit lane of the opposite straight vehicle team simultaneously; The system includes a model establishment and solving unit, the nonlinear programming model takes minimizing the intersection passing time as an objective, and a dynamic lane distribution and passing scheme is obtained by solving the nonlinear programming model, and the dynamic lane distribution and passing scheme includes the functions of each lane at the intersection, the stop line positions of the conflict lanes, and the vehicle numbers of each lane. 8.The intersection entrance and exit dynamic variable lane coordination optimization system in the mixed driving environment of automatic driving and manual driving according to claim 7, characterized in that, The intersection passing time is calculated by: Respectively calculating the vehicle team passing time of each lane in two opposite directions, and the maximum value of the vehicle team passing time is the intersection passing time; The nonlinear programming model is subjected to constraint conditions, including: Any lane in the same direction cannot be used as the entry lane of the straight lane and the left-turn lane simultaneously; The opposite lane of the straight lane cannot be the straight lane; At least one straight lane and one left-turn lane are allocated in any direction; The number of entry lanes for left-turn is not more than the number of corresponding exit lanes in the adjacent direction; The total number of vehicles allocated to each lane in any direction meets the vehicle number demand in the direction; Straight traffic flow enters the dynamic variable lane after the opposite left-turn traffic flow leaves the conflict point; The left-turn lane is allocated first and then the straight lane in the same direction; The lane without lane allocation function cannot allocate vehicles.

9. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 8. The computer program is executed by the processor to realize the intersection entrance and exit dynamic variable lane coordination optimization method in the mixed driving environment of automatic driving and manual driving according to any one of claims 1-6.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to realize the intersection entrance and exit dynamic variable lane coordination optimization method in the mixed driving environment of automatic driving and manual driving according to any one of claims 1-6.

Images