A bus exit control method based on master-slave game in intelligent network environment

Abstract

The present invention belongs to the field of road vehicle control methods, and relates to a bus exit control method based on master-slave game in an intelligent network environment, the method comprising the following steps: (1) setting a control area CZ with a bay-type stop as the center; (2) the bus completes boarding and disembarking, triggering a control decision, and sharing the real-time position, speed, acceleration and other information of each vehicle in the bus B and the social vehicle group FV; (3) obtaining the bus driving trajectory when the revenue of the bus B and the total revenue of the social vehicle group FV are maximized through an upper-level planning model; obtaining the social vehicle driving trajectory when the social vehicle group FV chooses to maximize individual revenue based on the decision on the bus B through a lower-level planning model; (4) solving the double-level planning model through NSGA-II to obtain the optimal bus and social vehicle driving trajectory; (5) executing the vehicle trajectory. The method is conducive to minimizing the negative impact of bus exit.

Description

Technical Field

[0001] The present invention belongs to the field of road vehicle control, and in particular, relates to a bus exit control method based on master-slave game in an intelligent network environment. Background Art

[0002] In recent years, self-driving vehicles are equipped with advanced sensing equipment, which can realize information sharing between vehicles and between vehicles and roads, and show great potential advantages in improving driving safety and traffic efficiency. The behavioral decision-making planning of self-driving vehicles has become a research hotspot in recent years. Self-driving vehicles do not exist in isolation in the decision-making process, but all traffic participants, including self-driving vehicles, predict the possible reactions of other traffic participants to their decisions while planning their own decisions. Game theory pays more attention to the interactivity of traffic participants. Its application covers various scenarios from simple competition to complex cooperation of vehicles, and can more truly reflect the decision-making behavior of drivers. In the context of a bay-type bus stop, the bus's exit behavior is a typical vehicle decision-making behavior. Due to its complex and high-frequency lane change requirements, it will produce complex interactions with social vehicles running normally on the lane. The existing bus lane change decision planning in the bay-type stop scenario is only applicable to the interaction between the bus and the nearest single social vehicle, and does not take into account the game between the bus exit and multiple social vehicles, and does not consider the overall impact of the bus exit on the stop area. Therefore, how to effectively control bus exits based on the use of vehicle network communication technology and autonomous driving technology while considering the overall impact of the stop area is an issue that technical personnel in this field urgently need to solve. Summary of the invention

[0003] In view of the above technical problems and defects, the purpose of the present invention is to provide a bus exit control method based on master-slave game in an intelligent network environment. The method involves the coordinated application of vehicle network communication technology and automatic driving technology. On the basis of determining the control area, a master-slave game two-level programming model is constructed, and the NSGA-II algorithm is used to solve it to obtain the optimal bus and social vehicle driving trajectory, thereby minimizing the negative impact of bus exit.

[0004] To achieve the above object, the present invention adopts the following technical solutions:

[0005] A bus exit control method based on master-slave game in an intelligent network environment, the method comprising the following steps:

[0006] Step S1: Taking the bay-type stop as a reference, the upstream and downstream ranges L meters away from the stop are set as the control zone CZ, the vehicles behind the bus, i.e. the upstream vehicles, are defined as the social vehicle group FV, and the vehicles overlapping with the bus or located in front of the bus, i.e. the downstream vehicles, are defined as the social vehicle group LV, and the lanes and social vehicles in the control zone are numbered sequentially from near to far;

[0007] Step S2: The bus completes boarding and disembarking, generates a desire to leave the station, and triggers a control decision. At this time, the real-time position, speed, and acceleration information of each bus and social vehicle group FV are shared through the communication system;

[0008] Step S3: Obtaining the driving trajectories of the public transport vehicle and the social vehicle through a double-layer programming model; wherein the double-layer programming model includes an upper-layer programming model and a lower-layer programming model; the upper-layer programming model is used to calculate and obtain the driving trajectory of the public transport vehicle when the total revenue of the public transport vehicle B and the social vehicle group FV is maximized;

[0009] The objective function of the upper-level planning model is as follows (1):

[0010]

[0011] in, is the optimal strategy for bus B, , , They are the strategy, strategy set and benefit of bus B respectively; is the strategy set of the social vehicle group FV under the decision of bus B; is the optimal strategy set of the social vehicle group FV under the decision of bus B; is the total revenue of bus B and social vehicle group FV;

[0012] The total revenue of bus B and social vehicle group FV The expression is as follows (2):

[0013]

[0014] in, For social vehicles FV ij of income, i represents the number of lanes, j Represents the vehicle number, FV ij Representative lane i No. j social vehicles;

[0015] The revenue of bus B The calculation method is as follows (3):

[0016]

[0017] in, , , Respectively represent the safety cost, comfort cost and passage cost of bus B; , , Respectively represent the weight coefficients of safety cost, comfort cost and traffic cost;

[0018] The profit calculation formula of the social vehicle group FV is as follows (11):

[0019]

[0020] In the formula, , , are the safety cost, comfort cost and traffic cost of the social vehicle group FV, , , Respectively represent the weight coefficients of safety cost, comfort cost and traffic cost;

[0021] The lower-level planning model is used to calculate the social vehicle driving trajectory of the social vehicle group FV based on the decision of the public transport vehicle B to maximize the individual benefits. The expression is as follows (35):

[0022]

[0023] in, , , They are social vehicles FV ij Strategies, strategy sets, and game payoffs, For social vehicles FV ij The optimal strategy, Except for social vehicles FV ij The strategy set of other social vehicles other than Representative vehicle FV ij The benefits of the decision made after comprehensively considering the decisions of bus B and other social vehicles besides itself;

[0024] Step S4: Solve the bi-level programming model by using the NSGA-II algorithm to obtain the optimal driving trajectory of the bus and social vehicles, use the quintic polynomial model to plan the horizontal and longitudinal trajectories of the bus and social vehicles, and only plan the longitudinal trajectory for non-lane-changing vehicles;

[0025] Step S5: Execute the vehicle trajectory according to the trajectory output in step S4.

[0026] As a preferred embodiment of the present invention, the safety of bus B is considered to be affected by the speed difference and relative distance with the vehicle behind in lane 1, and the safety cost The calculation formulas are as follows (4) to (7):

[0027]

[0028]

[0029]

[0030]

[0031] in, , They are public transport vehicle B and private vehicle FV 11 The weight coefficients of relative speed and relative distance; , They are bus B and vehicle FV 11 The longitudinal relative speed and relative distance; t1 is the time taken by bus B to complete the lane change and exit the station; A tiny number set to avoid a zero denominator; To adjust the parameters; , They are public transport vehicle B and social vehicle FV 11 The longitudinal velocity of , They are public transport vehicle B and social vehicle FV 11 The longitudinal position of is the body length of bus B.

[0032] As a preferred embodiment of the present invention, the integral of the absolute value of the acceleration of bus B over time represents the comfort of passengers on the bus B during the process of changing lanes and leaving the station. The calculation formula is as follows (8):

[0033]

[0034] in, , are the weight coefficients of the longitudinal and lateral comfort benefits of bus B respectively; , are the longitudinal and lateral accelerations of bus B respectively; This is the instruction for bus B to change lanes and exit the station. When the bus changes lanes and exits the station, , when the bus does not change lanes, ;n Represents the number of passengers on bus B.

[0035] As a preferred embodiment of the present invention, the travel cost of bus B is The calculation formulas are as follows (9) to (10):

[0036]

[0037]

[0038] in, is the expected speed of bus B; is the maximum speed of bus B; is the longitudinal speed of the front vehicle in lane 1.

[0039] As a preferred embodiment of the present invention, the safety cost of the social vehicle group FV is The calculation formula is as follows (12):

[0040] ;

[0041] Social Vehicle FV ij The security cost The calculation formula is as follows (13):

[0042] ;

[0043] Among them, when the social vehicle FV 1j When changing lanes, , when the social vehicle FV 1j When not changing lanes, ; , They are social vehicles FV ij vertical and horizontal security costs;

[0044] Social Vehicle FV 11 The lateral safety of the vehicle is determined by the social vehicle FV 11 With social vehicles FV on lane 2 21 Determined by relative speed and relative position;

[0045] Social Vehicle FV 11 The longitudinal safety of the social vehicle FV 11 It is determined by the relative speed and relative position with the vehicle in front of it in the same lane and whether bus B changes lanes to exit the station;

[0046] when i≠1 ∩ j≠1， Social Vehicle FV ijThe longitudinal safety cost is related to the relative speed and relative position of the vehicle in front of the lane, and the lateral safety cost is related to the relative speed and position of the vehicle behind in the adjacent lane.

[0047] As a preferred embodiment of the present invention, the comfort cost of the social vehicle group FV is The calculation formula is as follows (30):

[0048]

[0049] Among them, social vehicles FV 1j and social vehicle FV 2j The security cost The calculation formula is as follows (31):

[0050]

[0051] in, , They are social vehicles FV ij Longitudinal and lateral comfort cost weighting coefficients; , They are social vehicles FV ij longitudinal and lateral acceleration; For social vehicles FV 1j Lane change command, 1 when changing lanes, 0 when not changing lanes.

[0052] As a preferred embodiment of the present invention, the traffic cost of the social vehicle group FV is The calculation formula is as follows (32):

[0053]

[0054] Among them, social vehicles FV ij The cost of travel The calculation formulas are as follows (33) to (34):

[0055]

[0056]

[0057] in, For social vehicles FV ij Longitudinal speed; For social vehicles FV ij The expected driving speed; For social vehicles FV ij Maximum speed; , They are the front car in the lane and the front car in the target lane respectively.

[0058] As a preferred embodiment of the present invention, step S4 includes the following steps:

[0059] Step S41: Initialize the population;

[0060] The initial population is randomly generated, and each individual is composed of the decision variables of the bus and the social vehicle. Each individual is represented as a vector, which contains the lane change command and trajectory planning parameters of the bus, as well as the lane change command and trajectory planning parameters of each social vehicle participating in the game.

[0061] Step S42: fitness evaluation;

[0062] Evaluate the fitness of each individual and map it into the profit function of the game subject;

[0063] Step S43: non-dominated sorting;

[0064] The population is sorted non-dominatedly, and individuals are divided into different levels. Individuals within a level do not dominate each other, while individuals in different levels have a dominance relationship.

[0065] Step S44: Calculate the congestion degree;

[0066] Calculate the crowding degree of each individual to maintain the diversity of the population;

[0067] Step S45: Select an operation;

[0068] The tournament selection method is used to select parent individuals according to their rank and crowding degree to generate the parent population of the next generation.

[0069] Step S46: crossover and mutation;

[0070] Perform crossover and mutation operations on the selected parent individuals to generate offspring individuals;

[0071] Step S47: environment selection;

[0072] The parent and child populations are merged, non-dominated sorting and crowding calculations are performed, and the first half of the merged population is selected as the new generation population;

[0073] Step S48: iteration;

[0074] Repeat the fitness evaluation, non-dominated sorting, crowding calculation, crossover and mutation, and environment selection steps until the predetermined number of iterations is reached or the convergence condition is met, and output the optimal state.

[0075] As a further preferred embodiment of the present invention, the social vehicle FV 11 The calculation formula of the lateral safety cost is as follows (14) to (17):

[0076]

[0077]

[0078]

[0079]

[0080] in, , They are social vehicles FV 11 With social vehicle FV 21 The weight coefficients of relative speed and relative distance; , They are social vehicles FV 11 With social vehicle FV 21 Longitudinal relative speed and relative distance; , They are social vehicles FV 21 、FV 11 The longitudinal velocity of , They are social vehicles FV 21 、FV 11 The longitudinal position of For social vehicles FV 11 The length of the vehicle body;

[0081] When bus B changes lanes to leave the station, , social vehicle FV 11 Vertical security costs Determined by public transport vehicle B, social vehicle FV 11 The longitudinal safety cost of is the same as the lateral safety cost of bus B, that is, ;

[0082] When bus B does not change lanes when leaving the station, , social vehicle FV 11 Vertical security costs The calculation formulas are as follows (18) to (21):

[0083]

[0084]

[0085]

[0086]

[0087] in, , They are social vehicles FV11 LV with social vehicles 11 The weight coefficients of relative speed and relative distance; , They are social vehicles FV 11 LV with social vehicles 11 Longitudinal relative speed and relative distance; , They are social vehicles LV 11 、FV 11 The longitudinal velocity of , They are social vehicles LV 11 、FV 11 The longitudinal position of For social vehicles LV 11 Length of the vehicle body.

[0088] As a further preference of the present invention, when i≠1 ∩ j≠1， Social Vehicle FV ij Vertical security costs The calculation formulas are as follows (22) to (25):

[0089]

[0090]

[0091]

[0092]

[0093] in, , They are social vehicles FV ij and Social Vehicle FV ij-1 The weight coefficients of relative speed and relative distance; , They are social vehicles FV ij With social vehicle FV ij-1 The longitudinal relative speed and relative distance, They are social vehicles FV ij With social vehicle FV ij-1 The longitudinal velocity, They are social vehicles FV ij With social vehicle FV ij-1 The longitudinal position of For social vehicles FV ij-1 The length of the vehicle body;

[0094] Social Vehicle FVij Horizontal security costs As shown in equations (26) to (29):

[0095]

[0096]

[0097]

[0098]

[0099] in, , They are social vehicles FV 1j With social vehicle FV 2j The weight coefficients of relative speed and relative distance; , They are social vehicles FV 1j With social vehicle FV 2j Longitudinal relative speed and relative distance; , They are social vehicles FV 2j 、FV 1j The longitudinal velocity of , They are social vehicles FV 2j 、FV 1j The longitudinal position of For social vehicles FV 2j Length of the vehicle body.

[0100] Compared with the prior art, the advantages and beneficial effects of the present invention are:

[0101] 1. The bus exit control method provided by the present invention involves the coordinated application of vehicle networking communication technology and automatic driving technology. The method can effectively realize the automatic driving bus exit decision in the bay-type stop scenario.

[0102] 2. The present invention constructs a master-slave game two-level planning model and realizes the optimal bus exit trajectory solution through vehicle-to-vehicle interaction. The upper-level planning model simultaneously considers the overall benefits of autonomous buses and multiple social vehicles under the influence of exit. Such bus exit decision results are conducive to minimizing the negative impact of exit.

[0103] 3. The lower-level planning model of the master-slave game two-level planning model in the present invention adopts a dynamic game approach, which fully considers the autonomous decision-making of each affected social vehicle and its individual benefits. This design is more in line with actual decision-making scenarios.

[0104] 4. The control method provided by the present invention takes into account multiple objective factors such as safety, efficiency, and comfort, making the decision-making process more scientific and reasonable. BRIEF DESCRIPTION OF THE DRAWINGS

[0105] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present invention.

[0106] Figure 1 It is a flow chart of the bus exit control method provided by the present invention;

[0107] Figure 2 It is a plan view of the bus exit control area provided by the present invention;

[0108] Figure 3 It is a schematic diagram of the planned trajectory provided by the present invention; wherein, (a) is a spatial distribution diagram of the vehicle trajectory; (b) is a diagram of the longitudinal velocity change of the vehicle; (c) is a diagram of the acceleration change of the bus;

[0109] Figure 4 It is a flow chart of the NSGA-II algorithm provided by the present invention.

[0110] Figure 5 are the parameters in the bi-level programming model of the present invention. DETAILED DESCRIPTION

[0111] The technical solution of the present application will be described below in conjunction with the drawings in the present invention. Obviously, the specific implementation described is one implementation of the present invention, not all implementations. Based on the specific implementation of the present invention, all other implementations obtained by ordinary technicians in the field without creative work are within the scope of protection of the present invention.

[0112] See also Figures 1 to 4 , an embodiment of the present invention provides a bus exit control method based on master-slave game in an intelligent network environment, the method comprising the following steps:

[0113] Step S1: Taking the bay-type stop as a reference, the upstream and downstream ranges L meters away from the stop are set as the control zone CZ, the vehicles behind the bus, i.e. the upstream vehicles, are defined as the social vehicle group FV, and the vehicles overlapping with the bus or located in front of the bus, i.e. the downstream vehicles, are defined as the social vehicle group LV, and the lanes and social vehicles in the control zone are numbered sequentially from near to far;

[0114] Specifically, in this embodiment, the length of the control zone CZ is set to 60 meters, and the controlled social vehicles are numbered. The social vehicles in the social vehicle group FV are numbered FV ij , the social vehicle number in the social vehicle group LV is LV ij ， in i represents the number of lanes, j represents the vehicle number. The numbers of the social vehicles specifically involved in this embodiment are detailed in Figure 2 ; For information about specific roads and stops, see Table 1.

[0115] Table 1 Scenario design parameters

[0116]

[0117] Step S2: After the bus completes boarding and disembarking, it generates the intention to exit the station, triggering a control decision. At this time, the real-time position, speed, acceleration and other information of the bus B and each vehicle in the social vehicle group FV are shared through the communication system (V2X technology); the social vehicle status information when the bus triggers the control decision is shown in Table 2.

[0118] Table 2 Vehicle initial status information

[0119]

[0120] Step S3: Obtaining the driving trajectories of public transportation vehicles and social vehicles through a double-layer planning model; wherein the double-layer planning model includes an upper-layer planning model and a lower-layer planning model;

[0121] The upper-level planning model is used to calculate and obtain the public transport vehicle driving trajectory when the total benefit of the public transport vehicle B and the social vehicle group FV is maximized;

[0122] The objective function of the upper-level planning model is as follows (1):

[0123]

[0124] in, is the optimal strategy for bus B, , , They are the strategy, strategy set and benefit of bus B respectively; is the strategy set of the social vehicle group FV under the decision of bus B; is the optimal strategy set of the social vehicle group FV under the decision of bus B; is the total revenue of bus B and social vehicle group FV;

[0125] The total revenue of bus B and social vehicle group FV The expression is as follows (2):

[0126]

[0127] in, For social vehicles FV ij of income, i represents the number of lanes, j Represents the vehicle number, FV ij Representative lane i No. j social vehicles;

[0128] The revenue of bus B The calculation method of is as follows (3):

[0129]

[0130] in, , , represent the safety cost, comfort cost and passage cost of bus B. , , Respectively represent the weight coefficients of safety cost, comfort cost and traffic cost;

[0131] The safety of bus B is affected by the speed difference and relative distance between it and the vehicle behind it in lane 1. The safety cost The calculation formulas are as follows (4) to (7):

[0132]

[0133]

[0134]

[0135]

[0136] in, , They are public transport vehicle B and private vehicle FV 11 The weight coefficients of relative speed and relative distance; , They are bus B and vehicle FV 11 The longitudinal relative speed and relative distance; t1 is the time taken by bus B to complete the lane change and exit the station; A tiny number set to avoid the denominator being zero (a number that approaches 0 infinitely, but is not equal to 0); To adjust the parameters; , They are public transport vehicle B and social vehicle FV 11 The longitudinal velocity of , They are public transport vehicle B and social vehicle FV 11 The longitudinal position of is the body length of bus B;

[0137] The integral of the absolute value of the acceleration of bus B over time represents the comfort of the passengers on board when bus B changes lanes and leaves the station. The calculation formula is as follows (8):

[0138]

[0139] in, , are the weight coefficients of the longitudinal and lateral comfort benefits of bus B respectively; , are the longitudinal and lateral accelerations of bus B respectively; This is the instruction for bus B to change lanes and exit the station. When the bus changes lanes and exits the station, , when the bus does not change lanes, ; n Represents the number of passengers on bus B;

[0140] The travel cost of bus B The calculation formulas are as follows (9) to (10):

[0141]

[0142]

[0143] in, is the expected speed of bus B; is the maximum speed of bus B; is the longitudinal speed of the front vehicle in lane 1;

[0144] The profit calculation formula of the social vehicle group FV is as follows (11):

[0145]

[0146] In the formula, , , are the safety cost, comfort cost and traffic cost of the social vehicle group FV, , , Respectively represent the weight coefficients of safety cost, comfort cost and traffic cost;

[0147] The safety cost calculation formula of the social vehicle group FV is as follows (12):

[0148]

[0149] Social Vehicle FV ij The security cost The calculation formula is as follows (13):

[0150]

[0151] Among them, when the social vehicle FV 1j When changing lanes, , when the social vehicle FV 1j When not changing lanes, ; , They are social vehicles FV ij vertical and horizontal security costs;

[0152] Social Vehicle FV 11 The lateral safety of the vehicle is determined by the social vehicle FV 11 With social vehicles FV on lane 2 21 The social vehicle FV is determined by the relative speed and relative position. 11 The calculation formula of the lateral safety cost is as follows (14) to (17):

[0153]

[0154]

[0155]

[0156]

[0157] in, , They are social vehicles FV 11 With social vehicle FV 21 The weight coefficients of relative speed and relative distance; , They are social vehicles FV 11 With social vehicle FV 21 Longitudinal relative speed and relative distance; , They are social vehicles FV 21 、FV 11 The longitudinal velocity of , They are social vehicles FV 21 、FV 11The longitudinal position of For social vehicles FV 11 The length of the vehicle body;

[0158] Social Vehicle FV 11 The longitudinal safety of the social vehicle FV 11 It is determined by the relative speed and relative position with the vehicle in front of it in the same lane and whether bus B changes lanes to exit the station;

[0159] When bus B changes lanes to leave the station, , social vehicle FV 11 Vertical security costs Determined by public transport vehicle B, social vehicle FV 11 The longitudinal safety cost of is the same as the lateral safety cost of bus B, that is, ;

[0160] When bus B does not change lanes when leaving the station, , social vehicle FV 11 Vertical security costs The calculation formulas are as follows (18) to (21):

[0161]

[0162]

[0163]

[0164]

[0165] in, , They are social vehicles FV 11 LV with social vehicles 11 The weight coefficients of relative speed and relative distance; , They are social vehicles FV 11 LV with social vehicles 11 Longitudinal relative speed and relative distance; , They are social vehicles LV 11 、FV 11 The longitudinal velocity of , They are social vehicles LV 11 、FV 11 The longitudinal position of For social vehicles LV 11 The length of the vehicle body;

[0166] Social Vehicle FV ij (i≠1 ∩ j≠1 The longitudinal safety cost of the vehicle in the lane is related to the relative speed and relative position of the vehicle in front of it, and the lateral safety cost is related to the relative speed and position of the vehicle behind it in the adjacent lane. The calculation formulas are as follows (22) to (25):

[0167]

[0168]

[0169]

[0170]

[0171] in, , They are social vehicles FV ij and Social Vehicle FV ij-1 The weight coefficients of relative speed and relative distance; , They are social vehicles FV ij With social vehicle FV ij-1 The longitudinal relative speed and relative distance, They are social vehicles FV ij With social vehicle FV ij-1 The longitudinal velocity, They are social vehicles FV ij With social vehicle FV ij-1 The longitudinal position of For social vehicles FV ij-1 The length of the vehicle body;

[0172] Social Vehicle FV ij Horizontal security costs As shown in equations (26) to (29):

[0173]

[0174]

[0175]

[0176]

[0177] in, , They are social vehicles FV 1j With social vehicle FV 2j The weight coefficients of relative speed and relative distance; , They are social vehicles FV 1j With social vehicle FV 2j Longitudinal relative speed and relative distance; , They are social vehicles FV 2j 、FV 1j The longitudinal velocity of , They are social vehicles FV 2j 、FV 1j The longitudinal position of For social vehicles FV 2j The length of the vehicle body;

[0178] The comfort cost of social vehicle group FV The calculation formula is as follows (30):

[0179]

[0180] Among them, social vehicles FV 1j and social vehicle FV 2j The security cost The calculation formula is as follows (31):

[0181]

[0182] in, , They are social vehicles FV ij Longitudinal and lateral comfort cost weighting coefficients; , They are social vehicles FV ij longitudinal and lateral acceleration; For social vehicles FV 1j Lane change command, 1 when changing lanes, 0 when not changing lanes;

[0183] The travel cost of the social vehicle group FV The calculation formula is as follows (32):

[0184]

[0185] Among them, social vehicles FV ij The cost of travel The calculation formulas are as follows (33) to (34):

[0186]

[0187]

[0188] in, For social vehicles FV ij Longitudinal speed; For social vehicles FV ij The expected driving speed; For social vehicles FV ij Maximum speed; , They are the front car in the own lane and the front car in the target lane respectively;

[0189] The lower-level planning model is used to calculate the social vehicle driving trajectory of the social vehicle group FV based on the decision of the public transport vehicle B to maximize the individual benefits. The expression is as follows (35):

[0190]

[0191] in, , , They are social vehicles FV ij Strategies, strategy sets, and game payoffs, For social vehicles FV ij The optimal strategy, Except for social vehicles FV ij The strategy set of other social vehicles other than Representative vehicle FV ij The benefits of the decision made after comprehensively considering the decisions of bus B and other social vehicles besides itself;

[0192] The game rules between vehicles are explained by taking two vehicles as an example. The optional strategies of each vehicle can be acceleration, deceleration, and lane change. Assuming that the optional strategies of vehicle a are a1 and a2, and the optional strategies of vehicle b are b1 and b2, the game benefits of vehicle a under the strategy combination are Ua (a1, b1), Ua (a1, b2), Ua (a2, b1), Ua (a2, b2), and the game benefits of vehicle b are Ub (a1, b1), Ub (a1, b2), Ub (a2, b1), Ub (a2, b2). Vehicles a and b make the optimal strategy that maximizes their own benefits through game.

[0193] In this embodiment, the relevant parameters in the upper-level planning model and the lower-level planning model are set as follows: Figure 5 shown.

[0194] Step S4: Solve the bi-level programming model by using the NSGA-II algorithm to obtain the optimal driving trajectory of the bus and social vehicles, use the quintic polynomial model to plan the horizontal and longitudinal trajectories of the bus and social vehicles, and only plan the longitudinal trajectory for non-lane-changing vehicles;

[0195] Specifically, step S4 includes the following steps:

[0196] Step S41: Initialize the population;

[0197] The initial population is randomly generated, and each individual consists of the decision variables of the bus and the social vehicle. Each individual is represented as a vector, which contains the lane change instructions and trajectory planning parameters of the bus, as well as the lane change instructions and trajectory planning parameters of each social vehicle participating in the game. The decision variables constitute the decision space of the game subject (bus and social vehicle).

[0198] Step S42: fitness evaluation;

[0199] The fitness of each individual is evaluated and mapped into the profit function of the game subject. Specifically, the comprehensive cost function of public buses and social vehicles (including safety cost, comfort cost and travel time cost) is used as the profit function. The profit of social vehicles is the objective function of the lower-level game, while the profit of public bus B is combined with the profits of all social vehicles to form the objective function of the upper-level game.

[0200] Step S43: non-dominated sorting;

[0201] The population is sorted non-dominatedly and individuals are divided into different levels (Front). In this process, the strategy combination of each game subject and their mutual influence are considered. Individuals within a level do not dominate each other, while there is a dominance relationship between individuals at different levels, which reflects the game behavior in multi-objective optimization.

[0202] Step S44: Calculate the congestion degree;

[0203] The crowding degree of each individual is calculated to maintain the diversity of the population. The crowding degree is calculated based on the distribution of individuals in the objective function space, ensuring the diversity of strategy combinations among game players.

[0204] Step S45: Select an operation;

[0205] The tournament selection method is used to select parent individuals based on their rank and crowding to generate the parent population for the next generation. This step simulates the strategy selection process in the game, and the dominant strategy combination is more likely to be passed on to the next generation.

[0206] Step S46: crossover and mutation;

[0207] The selected parent individuals are subjected to crossover and mutation operations to generate offspring individuals. Crossover and mutation simulate the strategy evolution in the game, allowing the population to gradually optimize its strategy combination during the game.

[0208] Step S47: environment selection;

[0209] The parent and child populations are merged, non-dominated sorting and crowding calculations are performed, and the first half of the merged population is selected as the new generation population, which reflects the adaptability of the game subject in a dynamic environment.

[0210] Step S48: iteration;

[0211] Repeat the steps of fitness evaluation, non-dominated sorting, crowding calculation, selection, crossover and mutation until the predetermined number of iterations is reached or the convergence condition is met. In this process, the strategy combination of the population gradually evolves and eventually reaches the Nash equilibrium or Pareto optimal state.

[0212] In this embodiment, the NSGA-II algorithm parameter settings are shown in Table 3.

[0213] Table 3 Algorithm parameter settings

[0214] Parameter name Population size Crossover probability Mutation probability Maximum evolutionary generations Parameter Value 100 0.35 0.01 300

[0215] Furthermore, in this embodiment, the trajectory formula of the quintic polynomial model is as follows:

[0216]

[0217]

[0218] In the formula, , is the longitudinal and lateral position of the vehicle at time t; , are the longitudinal and transverse coefficients of the quintic polynomial trajectory, respectively.

[0219] The first-order derivative and second-order derivative of the vehicle's lateral and longitudinal trajectory functions are calculated to obtain the lateral and longitudinal velocities and accelerations of the lane-changing vehicle, as shown in the following formula:

[0220]

[0221]

[0222]

[0223]

[0224] in, express x(t) right t The first derivative of ; represent x(t) right t The second derivative of represent y(t) right t The first derivative of ; represent y(t) right t The second-order derivative of the lane-changing vehicles in this scenario (including the bus B and the social vehicle FV ij ) Its trajectory should comply with the following two conditions: (1) The curvature of the trajectory is 0 at the beginning and end of the lane change; (2) The vehicle body is parallel to the lane at the beginning and end of the lane change, that is, the lateral velocity is 0, the lateral and longitudinal accelerations are 0, and the width of the vehicle lateral across the entire lane is y d .

[0225] Assuming that the states of the vehicle at the initial and final lane change times are known, we can obtain the fifth-order polynomial path boundary conditions and calculate the vehicle trajectory coefficient, which is:

[0226]

[0227] in, The time when the vehicle starts changing lanes; , Indicates the longitudinal and lateral position coordinate points of the lane-changing vehicle's starting point; , Indicates the longitudinal speed and lateral speed of the lane-changing vehicle at the starting point; , Indicates the longitudinal acceleration and lateral acceleration of the lane-changing vehicle starting point; similarly, The time when the vehicle lane change ends; , Indicates the longitudinal and lateral position coordinate points of the lane-changing vehicle's end point; , Indicates the longitudinal speed and lateral speed of the lane-changing vehicle at the end point; , Indicates the longitudinal acceleration and lateral acceleration of the lane-changing vehicle end point.

[0228] Step S5: executing the vehicle trajectory according to the trajectory output in step S4;

[0229] In this embodiment, the horizontal and vertical trajectories of public transport vehicles and private vehicles are shown in FIG. Figure 3 shown.

[0230] The present invention also provides an electronic device, comprising: one or more processors and a memory; wherein the memory is used to store one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned bus exit control method based on master-slave game in the intelligent network environment.

[0231] The present invention also provides a computer-readable medium having a computer program stored thereon, and when the computer program is executed by a processor, the bus exit control method based on master-slave game in the above-mentioned intelligent network environment is implemented.

[0232] Those skilled in the art will appreciate that all or part of the functions of the various methods / modules in the above embodiments may be implemented by hardware or by computer programs. When all or part of the functions in the above embodiments are implemented by computer programs, the program may be stored in a computer-readable storage medium, which may include: a read-only memory, a random access memory, a disk, an optical disk, a hard disk, etc. The program is executed by a computer to implement the above functions. For example, the program is stored in the memory of the device, and when the program in the memory is executed by the processor, all or part of the above functions can be implemented.

[0233] In addition, when all or part of the functions in the above-mentioned embodiments are implemented by means of a computer program, the program can also be stored in a storage medium such as a server, another computer, a disk, an optical disk, a flash drive or a mobile hard disk, and saved to the memory of a local device by downloading or copying, or the system of the local device is updated. When the program in the memory is executed by the processor, all or part of the functions in the above-mentioned embodiments can be implemented.

[0234] The above specific examples are used to illustrate the present invention, which is only used to help understand the present invention and is not intended to limit the present invention. For those skilled in the art of the present invention, according to the idea of ​​the present invention, some simple deductions, deformations or substitutions can be made. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A bus exit control method based on master-slave game in an intelligent network environment, characterized in that: The method comprises the following steps: Step S1: Taking the bay-type stop as a reference, the upstream and downstream ranges L meters away from the stop are set as the control zone CZ, the vehicles behind the bus, i.e. the upstream vehicles, are defined as the social vehicle group FV, and the vehicles overlapping with the bus or located in front of the bus, i.e. the downstream vehicles, are defined as the social vehicle group LV, and the lanes and social vehicles in the control zone are numbered sequentially from near to far; Step S2: The bus completes boarding and disembarking, generates a desire to leave the station, and triggers a control decision. At this time, the real-time position, speed, and acceleration information of each bus and social vehicle group FV are shared through the communication system; Step S3: Obtaining the driving trajectories of the public transport vehicle and the social vehicle through a double-layer programming model; wherein the double-layer programming model includes an upper-layer programming model and a lower-layer programming model; the upper-layer programming model is used to calculate and obtain the driving trajectory of the public transport vehicle when the total revenue of the public transport vehicle B and the social vehicle group FV is maximized; The objective function of the upper-level planning model is as follows (1): in, is the optimal strategy for bus B, s b , S b , U b are the strategy, strategy set and benefit of bus B respectively; s fv is the strategy set of the social vehicle group FV under the decision of bus B; is the optimal strategy set of the social vehicle group FV under the decision of bus B; is the total revenue of bus B and social vehicle group FV; The total revenue of bus B and social vehicle group FV The expression is as follows (2): in, For social vehicles FV ij The revenue of FV is, i represents the lane number, j represents the vehicle number, ij represents the jth social vehicle in lane i; The revenue of bus B is U b The calculation method of is as follows (3): in, They represent the safety cost, comfort cost and passage cost of bus B, α1, α2 and α3 represent the weight coefficients of safety cost, comfort cost and passage cost respectively; The profit calculation formula of the social vehicle group FV is as follows (11): In the formula, are the safety cost, comfort cost and passage cost of the social vehicle group FV, β1, β2 and β3 represent the weight coefficients of safety cost, comfort cost and passage cost respectively; The lower-level planning model is used to calculate the social vehicle group FV based on the decision of the public transport vehicle B to select the social vehicle driving trajectory under the condition of maximizing individual benefits, which is expressed as follows (35): in, They are social vehicles FV ij Strategies, strategy sets, and game payoffs, For social vehicles FV ij The optimal strategy, Except for social vehicles FV ij The strategy set of other social vehicles other than Representative vehicle FV ij The benefits of the decision made after comprehensively considering the decisions of bus B and other social vehicles besides itself; Step S4: Solve the bi-level programming model by using the NSGA-II algorithm to obtain the optimal driving trajectory of the bus and social vehicles, use the quintic polynomial model to plan the horizontal and longitudinal trajectories of the bus and social vehicles, and only plan the longitudinal trajectory for non-lane-changing vehicles; Step S5: Execute the vehicle trajectory according to the trajectory output in step S4.

2. According to the method of controlling bus exit based on master-slave game in intelligent network environment in claim 1, it is characterized in that: The safety of bus B is affected by the speed difference and relative distance between it and the vehicle behind it in lane 1. The safety cost The calculation formulas are as follows (4) to (7): in, They are public transport vehicle B and private vehicle FV 11 The weight coefficients of relative speed and relative distance; They are bus B and vehicle FV 11 The longitudinal relative speed and relative distance; t1 is the time taken by bus B to complete the lane change and exit the station; ε is a small number set to avoid the denominator being zero; To adjust the parameters; They are public transport vehicle B and social vehicle FV 11 The longitudinal velocity of b (t) They are public transport vehicle B and social vehicle FV 11 The longitudinal position of b is the body length of bus B.

3. According to the method for controlling bus exit based on master-slave game in an intelligent network environment in claim 1, it is characterized in that: The integral of the absolute value of the acceleration of bus B over time represents the comfort of the passengers on board when bus B changes lanes and leaves the station. The calculation formula is as follows (8): in, are the weight coefficients of the longitudinal and lateral comfort benefits of bus B respectively; are the longitudinal and lateral accelerations of bus B respectively; λ is the lane-changing instruction for bus B to leave the station. When the bus changes lanes to leave the station, λ=1; when the bus does not change lanes, λ=0; n represents the number of passengers on bus B.

4. According to the method for controlling bus exit based on master-slave game in an intelligent network environment in claim 1, it is characterized in that: The travel cost of bus B The calculation formulas are as follows (9) to (10): in, is the expected speed of bus B; is the maximum speed of bus B; is the longitudinal speed of the front vehicle in lane 1.

5. According to the method of controlling bus exit based on master-slave game in intelligent network environment in claim 1, it is characterized in that: Safety cost of social vehicle group FV The calculation formula is as follows (12): Social Vehicle FV ij The security cost The calculation formula is as follows (13): Among them, when the social vehicle FV 1j When changing lanes, When social vehicle FV 1j When not changing lanes, They are social vehicles FV ij vertical and horizontal security costs; Social Vehicle FV 11 The lateral safety of the vehicle is determined by the social vehicle FV 11 With social vehicles FV on lane 2 21 Determined by relative speed and relative position; Social Vehicle FV 11 The longitudinal safety of the social vehicle FV 11 It is determined by the relative speed and relative position with the vehicle in front of it in the same lane and whether bus B changes lanes to exit the station; When i≠1∩j≠1, the social vehicle FV ij The longitudinal safety cost is related to the relative speed and relative position of the vehicle in front of the lane, and the lateral safety cost is related to the relative speed and position of the vehicle behind in the adjacent lane.

6. The bus exit control method based on master-slave game in an intelligent network environment according to claim 1 is characterized in that: The comfort cost of social vehicle group FV The calculation formula is as follows (30): Among them, social vehicles FV 1j and social vehicle FV 2j The security cost The calculation formula is as follows (31): in, They are social vehicles FV ij Longitudinal and lateral comfort cost weighting coefficients; They are social vehicles FV ij Longitudinal and lateral acceleration; For social vehicles FV 1j Lane change command, 1 when changing lanes, 0 when not changing lanes.

7. The bus exit control method based on master-slave game in an intelligent network environment according to claim 1 is characterized in that: The travel cost of the social vehicle group FV The calculation formula is as follows (32): Among them, social vehicles FV ij The cost of travel The calculation formulas are as follows (33) to (34): in, For social vehicles FV ij Longitudinal speed; For social vehicles FV ij Expected driving speed; For social vehicles FV ij Maximum speed; They are the front car in the lane and the front car in the target lane respectively.

8. The bus exit control method based on master-slave game in an intelligent network environment according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: Initialize the population; The initial population is randomly generated, and each individual is composed of the decision variables of the bus and the social vehicle. Each individual is represented as a vector, which contains the lane change command and trajectory planning parameters of the bus, as well as the lane change command and trajectory planning parameters of each social vehicle participating in the game. Step S42: fitness evaluation; Evaluate the fitness of each individual and map it into the profit function of the game subject; Step S43: non-dominated sorting; The population is sorted non-dominatedly, and individuals are divided into different levels. Individuals within a level do not dominate each other, while individuals in different levels have a dominance relationship. Sub-step S44: congestion calculation; Calculate the crowding degree of each individual to maintain the diversity of the population; Step S45: Select an operation; The tournament selection method is used to select parent individuals according to their rank and crowding degree to generate the parent population of the next generation. Step S46: crossover and mutation; Perform crossover and mutation operations on the selected parent individuals to generate offspring individuals; Step S47: environment selection; The parent and child populations are merged, non-dominated sorting and crowding calculations are performed, and the first half of the merged population is selected as the new generation population; Step S48: iteration; Repeat the fitness evaluation, non-dominated sorting, crowding calculation, crossover and mutation, and environment selection steps until the predetermined number of iterations is reached or the convergence condition is met, and output the optimal state.

9. The bus exit control method based on master-slave game in an intelligent network environment according to claim 5 is characterized in that: Social Vehicle FV 11 The calculation formula of the horizontal safety cost is as follows (14) to (17): in, They are social vehicles FV 11 With social vehicle FV 21 The weight coefficients of relative speed and relative distance; They are social vehicles FV 11 With social vehicle FV 21 Longitudinal relative speed and relative distance; They are social vehicles FV 21 、FV 11 The longitudinal velocity of They are social vehicles FV 21 、FV 11 The longitudinal position of For social vehicles FV 11 The length of the vehicle body; When bus B changes lanes to leave the station, that is, λ=1, the social vehicle FV 11 Vertical security costs Determined by public transport vehicle B, social vehicle FV 11 The longitudinal safety cost of is the same as the lateral safety cost of bus B, that is, When bus B does not change lanes when leaving the station, that is, λ=0, the social vehicle FV 11 Vertical security costs The calculation formulas are as follows (18) to (21): in, They are social vehicles FV 11 LV with social vehicles 11 The weight coefficients of relative speed and relative distance; They are social vehicles FV 11 LV with social vehicles 11 Longitudinal relative speed and relative distance; They are social vehicles LV 11 、FV 11 The longitudinal velocity of They are social vehicles LV 11 、FV 11 The longitudinal position of For social vehicles LV 11 Length of the vehicle body.

10. The bus exit control method based on master-slave game in an intelligent network environment according to claim 5, characterized in that: When i≠1∩j≠1, the social vehicle FV ij Vertical security costs The calculation formulas are as follows (22) to (25): in, They are social vehicles FV ij With social vehicle FV ij-1 The weight coefficients of relative speed and relative distance; are the longitudinal relative speed and relative distance between the social vehicle FVij and the social vehicle FVij-1, are the longitudinal velocities of the social vehicle FVij and the social vehicle FVij-1, are the longitudinal positions of the social vehicle FVij and the social vehicle FVij-1 respectively; is the body length of the social vehicle FVij-1; Social Vehicle FV ij Horizontal security costs As shown in equations (26) to (29): in, They are social vehicles FV 1j With social vehicle FV 2j The weight coefficients of relative speed and relative distance; They are social vehicles FV 1j With social vehicle FV 2j Longitudinal relative speed and relative distance; They are social vehicles FV 2j 、FV 1j The longitudinal velocity of They are social vehicles FV 2j 、FV 1j The longitudinal position of For social vehicles FV 2j Length of the vehicle body.

Images