A vehicle dispatching method for unsignalized intersections in an intelligent network environment

Abstract

The present invention belongs to the field of intelligent transportation systems, and specifically relates to a method for dispatching vehicles at signal-free intersections in an intelligent network environment. The present invention constructs a Petri net model based on the spatial resources in the internal area of ​​the intersection and the occupation of various resources in the continuous process of vehicles passing through the intersection; the road test unit collects vehicle information, generates a traffic sequence according to the encoding and decoding rules, and iteratively analyzes the changes and resources in the Petri net to determine whether the initiation of each change in the sequence will cause a deadlock in the system, and repairs the sequence that causes the deadlock; an improved genetic algorithm is used for optimization, combined with the initialization of the population by Circle chaos mapping, chromosome standardization operations, improved adaptive crossover formulas, and operations for mutating similar individuals, to obtain the optimal or better dispatching strategy, thereby improving the traffic efficiency of the intersection and maximizing the utilization of the right-of-way resources at the intersection.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent transportation systems, and in particular relates to a vehicle dispatching method for an unsignalized intersection in an intelligent network environment. Background Art

[0002] With the rapid growth of the number of motor vehicles, intersections, as the hub of the traffic network, have a direct impact on the efficiency of road traffic with their traffic diversion capacity. Although traditional signal light control methods can alleviate traffic pressure, they cannot adapt to changes in traffic flow in real time, resulting in empty green lights and unnecessary parking and waiting. Therefore, studying the scheduling technology of unsignalized intersections and optimizing vehicle traffic through intelligent transportation systems have become important ways to improve the efficiency and safety of intersections. In order to meet the needs of my country's road traffic for solving the problem of vehicle scheduling at intersections, the intersection scheduling method under the intelligent network environment needs to be studied and improved in depth.

[0003] In document 1 (Vehicle Scheduling Method for Unsignalized Intersections Based on Petri Nets and Heuristic Algorithms), researchers adopted a discretized intersection modeling method based on Petri nets, and used a priority-based genetic algorithm to optimize the scheduling of vehicles at unsignalized intersections, thereby obtaining a better scheduling strategy. However, although this method uses a Petri net supervisory controller to avoid the cyclic waiting phenomenon of vehicles, it is unable to dynamically adjust the scheduling sequence online. In addition, its priority-based coding rules may also miss some better solutions. The present invention proposes a vehicle scheduling method for unsignalized intersections, which can dynamically adjust the scheduling strategy of the intersection online according to the real-time traffic conditions and vehicle driving conditions, thereby solving the cyclic waiting phenomenon while maximizing the traffic efficiency of the intersection. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and propose a vehicle dispatching method for an unsignalized intersection in an intelligent network environment, so as to give a vehicle dispatching sequence while avoiding circular waiting at the intersection, so as to improve the traffic efficiency in the intelligent network environment, increase the throughput of the intersection, and thus maximize the utilization of intersection road rights resources.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solutions:

[0006] A method for dispatching vehicles at an unsignalized intersection in an intelligent network environment comprises the following steps:

[0007] S1. Modeling: Based on the spatial resources in the internal area of ​​the intersection and the occupation of various resources in the continuous process of vehicles passing through the intersection, a Petri net model (N, M0) of the unsignalized intersection system is constructed, and the following symbols are defined:

[0008] N: A Petri net N = (P, T, F) consisting of circular nodes, square nodes and directed arcs, representing an intersection traffic system consisting of k right-of-way points and allowing vehicles on n paths to pass through;

[0009] P: Place set, which is the set of all circular nodes in N. Where Pi0=Pis∪Pif, Pis is the buffer zone before the vehicle enters the intersection, and Pif is the buffer zone when the vehicle leaves the intersection; d represents the total number of right-of-way points on path i, j represents the jth right-of-way point passed by the vehicle, Pij represents the jth passage process of the vehicle on path i through the intersection; PR = {Pr}, Pr represents the library set composed of all right-of-way points;

[0010] T: transition set, consisting of all square nodes in N, in d represents the total number of right-of-way points on path i, and the initiation of tij means that the vehicle on the i-th path travels from the j-1th right-of-way point to the jth right-of-way point;

[0011] F: directed arc set, representing the driving conditions of each vehicle in the system and the demand and release of resources when passing each right-of-way point during driving;

[0012] M:P→N is an identifier (N is a non-negative integer set), which indicates the operating status of the system. Each number in the identifier represents the number of vehicles or resources contained in each location. Among them, M0 is the initial identifier, indicating that the current vehicles have not entered the inner area of ​​the intersection, the vehicles are in the buffer zone, and the resources are not occupied.

[0013] A: The incidence matrix, which represents the changing pattern of the number of tokens in each repository under each transition in N. It is a matrix with |T| rows and |P| columns.

[0014] For the unsignalized intersection system, the objective function of system scheduling is to minimize the maximum travel time, that is:

[0015] F=min{max{C l}}

[0016] T max =max{C l}

[0017] Among them, C l is the time when the kth vehicle leaves the intersection, T max The maximum travel time for all vehicles to pass through;

[0018] S2, the intelligent roadside dispatching system obtains vehicle information in the unsignalized intersection area, including the number of vehicles, speed, and lane number;

[0019] S3, encoding and decoding: number the vehicles that are about to arrive at the intersection area and generate a passage sequence according to the passage process of each vehicle passing through each road right point on its path, and decode it into a transition sequence;

[0020] S4, generate chaotic initial population: in accordance with the requirements of S3 encoding and decoding, use Circle chaotic mapping to generate an initial population with chromosome number unit_num;

[0021] S5, standardizing the chromosomes: In order to reduce the algorithm search space, all chromosomes are limited to the coding subspace containing all feasible scheduling solutions, and the chromosomes are standardized to reduce the number of equivalent chromosomes, and then S6 is executed;

[0022] S6, Detection and Repair: By iteratively analyzing the transitions and resources in the Petri net, deadlocks are detected and repaired to ensure that each chromosome in the population is free of deadlocks;

[0023] S7, calculate the maximum transit time and fitness value of a single chromosome, and output the optimal individual of the current population;

[0024] S8, determine whether the termination condition gen>Maxgen is met, where gen is the number of iterations of the current population and Maxgen is the maximum number of iterations; if the termination condition is met, the optimal individual is output, if not, execute S9;

[0025] S9, genetic operation: perform the three genetic operations of selection, crossover and mutation on the current population, and execute S5 after obtaining the new generation of population.

[0026] As a preferred technical solution of the present invention, the specific steps of S3 are as follows:

[0027] S3-1. Number the vehicles according to their travel paths and the order in which they arrive at the intersection. The number is a two-digit integer ID, where I represents the path that the vehicle is going to travel and D represents the order in which the same vehicle arrives at the intersection.

[0028] S3-2, use the traffic sequence to indicate the traffic order of all vehicles, vehicle number ID i The ID of the i-th vehicle corresponding to path I is i The x-th occurrence of corresponds to its x-th passing process, and the length of the passing sequence m is the total number of passing sequences of all vehicles;

[0029] S3-3, decode the encoded pass sequence into a transition sequence, and the ID in the pass sequence i The yth occurrence of corresponds to the yth transition on path I.

[0030] As a preferred technical solution of the present invention, the specific steps of S4 are as follows:

[0031] S4-1, record the total number of vehicles that are about to pass through the intersection as θ, and arrange the θ vehicles in ascending order of number into an array a0, and then expand the sequence according to the number of transitions that each vehicle needs to pass in the path of the Petri net to obtain the expanded array a1, and record the total number of numbers in the array a1 as m;

[0032] S4-2, initialize the population array PopChrom, which contains unit_num empty arrays;

[0033] S4-3, let v=1, record the chromosome number to be generated;

[0034] S4-4, let i=1, record the number of iterations of the chaotic mapping;

[0035] S4-5. Generate a random number B (1) ∈(0,1);

[0036] S4-6. According to the Circle chaos mapping formula, we get C (i) , the Circle chaos mapping formula is as follows:

[0037]

[0038] C (i) =int(m×B (i) )+1

[0039] In the formula, λ is the control parameter, B (i) is the random number of the Circle chaotic map generated after the i-th iteration, C (i) Indicates the position number selected in the chromosome, int() is the rounding function;

[0040] S4-7, the vehicle number ID in array a1 i Put it in PopChrom[v][C (i) ], i=i+1;

[0041] S4-8, if i>m, execute S4-9, otherwise execute S4-6;

[0042] S4-9. If v>unit_num, output the generated initial population PopChrom, otherwise v=v+1, and execute S4-4.

[0043] As a preferred technical solution of the present invention, chromosomes are standardized in S5, and the specific steps are as follows:

[0044] S5-1, i=1, represents the sequence number of the chromosome in the population;

[0045] S5-2. Initialize a new chromosome array S5-3, decode PopChrom[i] into transition sequence Tran;

[0046] S5-4. Initialize an array with the same length as Tran To store the changes that have been judged;

[0047] S5-5, j=1, indicating the sequence number of the transition in the transition sequence Tran;

[0048] S5-6, traverse all paths, find the path corresponding to transition Tran[j], obtain the list of vehicles to be passed by the path, and sort the vehicle numbers from small to large into an array Cars;

[0049] S5-7, let temp[j] = Tran[j], store the determined transition;

[0050] S5-8. traverse the number of times Tran[j] appears in temp j , then Gene i [j] = Cars[count j ]; j = j + 1;

[0051] S5-9, if j> the length of the array Tran, then the standardization of the current chromosome is completed, i=i+1, and S5-10 is executed; otherwise, S5-6 is executed;

[0052] S5-10. If i> the length of the population PopChrom, complete the standardization of all chromosomes in the entire population and output the standardized population; otherwise, execute S5-2.

[0053] As a preferred technical solution of the present invention, the specific steps of S6 are as follows:

[0054] S6-1, set u=1, record the transition number currently detected;

[0055] S6-2. If u is greater than the length of the transition sequence, execute S7. Otherwise, set the uth transition in the transition sequence to t. α , execute S6-3; S6-3, check the transition t α Is it enabled under the current flag? If it is enabled, execute S6-4. Otherwise, start from t α Then select an enabled transition and place it at t α Before, and update t α ;

[0056] S6-4. Use the one-step forward method to check the change t αCan it cause deadlock? If it will cause deadlock, select t α The subsequent enabling transition is placed before this transition and updates t α , re-execute S6-4; otherwise, trigger t α , and update the current flag, set u=u+1, and execute S6-2.

[0057] As a preferred technical solution of the present invention, the specific steps of S7 are as follows:

[0058] S7-1. Calculate the maximum travel time: determine the idle time of the next right-of-way point that the vehicle is about to travel, compare it with the estimated departure time of the vehicle's previous right-of-way point, and take the larger value of the two as the start time of the current vehicle entering the next right-of-way point; after calculating all the travel processes of all vehicles, the maximum departure time in the system is the maximum travel time T of the entire travel sequence max ;

[0059] S7-2, calculate the fitness value: calculate the fitness value Adapt, the calculation formula is as follows:

[0060]

[0061] Among them, MaxT is the maximum transit time in the chromosome of the current population, MinT is the minimum transit time in the chromosome of the current population, and T max is the transit time of the current chromosome, k is a constant;

[0062] S7-3. Output the best individual: Output the best chromosome in the population and the corresponding T max .

[0063] As a preferred technical solution of the present invention, the genetic operation in S9 comprises the following specific steps:

[0064] S9-1, selection operation: Arrange the chromosomes in the population from large to small according to the fitness value, select the first Selectnum×unit_num chromosomes to join the next generation, where Selectnum is the selection factor and unit_num is the population size;

[0065] S9-2, Crossover operation: In order to promote the preservation of excellent genes and increase the diversity of the population, an adaptive crossover operator is used to perform a chromosome crossover operation. The specific steps are as follows:

[0066] S9-2-1, i=1, represents the chromosome number in the population;

[0067] S9-2-2. Calculate the fitness value f of each chromosome in the population i ;

[0068] S9-2-3. Calculate the average fitness value f of the population avg , maximum fitness value f max ;

[0069] S9-2-4. According to the improved adaptive crossover rate formula, calculate the crossover probability P of the i-th chromosome ci , the formula is as follows:

[0070]

[0071] Among them, P c1 , P c2 is the control parameter, P c1 , P c2 ∈(0,1);γ is the decay rate;

[0072] S9-2-5, generate a random number ran between (0,1), if ran <P ci , then perform partial mapping crossover on the chromosome, randomly select the chromosome and one of the chromosomes selected in S9-1 as the parent chromosome, randomly select two insertion points between the chromosomes, and exchange the fragments of the two parent chromosomes between the insertion points; move the fragments between the insertion points in the original chromosome to the front of the process sequence, and then delete the genes that are the same as the inserted fragment genes from the front to the back; let i=i+1; S9-2-6, if i>unit_num, output the new generation population, otherwise execute S9-2-4;

[0073] S9-3, mutation operation: In order to avoid the increase of the number of similar individuals in the population after the crossover operation, the chromosomes with differences less than a certain threshold are mutated. The specific steps are as follows:

[0074] S9-3-1. Calculate the Hamming Distance between chromosomes in the population, which indicates the number of different numbers at the corresponding positions of two chromosomes;

[0075] S9-3-2, set the control threshold δ. If the Hamming Distance of two chromosomes is less than δ, one of them will be randomly retained and the other will execute S9-3-3.

[0076] S9-3-3. Randomly select a mutation point in the chromosome and randomly determine a mutation length, and swap the genes of the mutation length before and after the mutation point.

[0077] The vehicle dispatching method for an unsignalized intersection in an intelligent network environment described in the present invention adopts the above technical solution and has the following technical effects compared with the prior art:

[0078] (1) The present invention is based on Petri net modeling, providing an intuitive graphical representation of the intersection structure and vehicle driving process, which is convenient for formal analysis and verification.

[0079] (2) The present invention repairs the sequence through a deadlock detection algorithm, which can ensure that all scheduling sequences will not cause vehicles to wait in a loop while the algorithm is optimizing, thereby solving the problem of vehicle congestion at unsignaled intersections in an intelligent network environment.

[0080] (3) The present invention uses an improved genetic algorithm for optimization, combined with Circle chaos mapping to initialize the population, chromosome standardization operations, improved adaptive crossover formulas, and operations to mutate similar individuals, which can quickly obtain the optimal or better scheduling strategy, thereby improving the traffic efficiency of the intersection. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 It is a flow chart of a method for dispatching vehicles at an unsignalized intersection in an intelligent network environment according to the present invention;

[0082] Figure 2 is a discretization model of a two-way four-lane intersection in an embodiment of the present invention;

[0083] Figure 3 It is a straight-going and left-turning vehicle route of a two-way four-lane vehicle in an embodiment of the present invention;

[0084] Figure 4 It is a Petri net model of a two-way four-lane intersection in an embodiment of the present invention. DETAILED DESCRIPTION

[0085] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present application, some symbols in the embodiments of the present application are first explained to facilitate understanding by those skilled in the art.

[0086] This embodiment is a specific application of a vehicle dispatching method for an unsignalized intersection in an intelligent network environment at a two-way four-lane intersection. The two-way four-lane intersection is discretized into 24 right-of-way points, and is modeled using Petri nets. Vehicles that are about to pass through the intersection are numbered to generate a traffic sequence. Different traffic sequences correspond to the order in which vehicles pass through each right-of-way point. The goal of its dispatching strategy is to achieve deadlock detection and repair of the traffic sequence while ensuring that the passage process of each vehicle is not missed, and to minimize the maximum passage time as the optimization goal, thereby avoiding the cyclic waiting phenomenon in the intersection area and improving the traffic efficiency of the intersection. The specific steps are as follows:

[0087] Step 1) (Modeling): Discretize the internal area of ​​the intersection, such as Figure 2As shown in the figure, the discretized intersection consists of 24 small cells. Each cell is set to accommodate only one vehicle, and the space of the cell can ensure the safe distance between this vehicle and other vehicles. The discretized cells are called right-of-way points. The straight and left-turn vehicle routes of a two-way four-lane road are as follows: Figure 3 As shown in Figure 1, the vehicle has 8 paths, which are straight paths in four directions and left-turn paths. 1, 2, 3, and 4 are straight paths in four directions, and 5, 6, 7, and 8 are left-turn paths in four directions. i The Petri net modeling includes the place set P, transition set T, directed arc set F, and system operation state M. The vehicle starts from the starting place P. is Enter the inner area of ​​the intersection and pass through the library P ij , and finally by the end library P if The entire Petri net model consists of 8 single paths Road i The Petri net model is composed of Figure 4 The definitions of symbols in the Petri net are as follows:

[0088] P is (i=1, 2, 3, 4, 5, 6, 7, 8) is the starting place of path i, from which the vehicle enters the inner area of ​​the intersection, and enters the ending place P after passing all right-of-way points. if , P is The number in represents the total number of vehicles to pass through the path;

[0089] P = {p ij , i=1, 2, 3, 4; j=1, 2, 3, 4}∪{p ij , i = 5, 6, 7, 8; j = 1, 2, 3, 4, 5, 6} is the

[0090] represents the library set of the vehicle driving process, where p ij Indicates that the vehicle travels to the jth right-of-way point on path i, p ij The number in represents the current number of vehicles (can only be 0 or 1);

[0091] P c = {r k , k = 1, 2, 3, ..., 22, 23, 24} is the resource library, where r k The number in indicates whether the current right-of-way point is occupied, 0 means it is occupied, and 1 means it is not occupied;

[0092] T=P={t ij , i=1, 2, 3, 4; j=1, 2, 3, 4, 5}∪{t ij , i=5, 6, 7, 8; j=1, 2, 3, 4, 5, 6,

[0093] 7} is a transition set, where t ij It means that the vehicle on path i ends the j-1th right-of-way point and starts the jth right-of-way point.

[0094] Specifically, 11 The vehicle on path 1 starts from the starting place p 1s Entering the inner area of ​​the intersection, p 11 Indicates that the vehicle enters the first right-of-way point r 10 , the process takes r 10 The right-of-way resources in the resource library are 10 The number of right-of-way resources in is 0; 12 Indicates p 11 The vehicle in the end of the occupation of the previous right-of-way point, began to enter the next right-of-way point p 12 , at this time r 10 The number in is 1, the number in r9 is 0, and so on, until the vehicle enters the end library p 1f , indicating that the vehicle has exited the intersection area.

[0095] Table 1 Physical meaning of places and transitions in the Petri net model

[0096] element meaning <![CDATA[P is ]]> The vehicle on the i-th path enters the vehicle waiting area at the intersection entrance <![CDATA[t ij ]]> The vehicle on the i-th path ends at the j-1th right-of-way point and starts at the j-th right-of-way point <![CDATA[P ij ]]> The vehicle enters the jth right-of-way point on the i-th path <![CDATA[P if ]]> The vehicle on the i-th path leaves the intersection area <![CDATA[r i ]]> The i-th right-of-way point

[0097] Step 2) The intelligent roadside dispatching system obtains vehicle information in the unsignalized intersection area, including the number of vehicles, speed, and lane number;

[0098] Specifically, this example uses the traffic flow information of a two-way four-lane intersection area within a certain period of time, in which there are 29 vehicles in eight lanes. The number of vehicles in lanes 1 to 8 are 6, 5, 4, 5, 3, 1, 4, 1. The speeds of vehicles 1 to 29 are 7.5, 5, 6.5, 4, 7.5, 8.5, 2.5, 6, 5, 7.5, 3, 3.5, 7, 7.5, 4, 5, 5.5, 4, 7.5, 6.5, 2.5, 4.5, 3, 2, 3, 3.5, 7, 5.5, 5, in m / s.

[0099] Step 3) (Encoding and decoding): Number the vehicles that are about to arrive at the intersection area and generate a passage sequence based on the passage process of each vehicle passing through each road right point on its path, and decode it into a transition sequence. The specific steps are as follows:

[0100] Step 3-1): The vehicles are numbered according to their travel paths and the order in which they arrive at the intersections. The numbering is a two-digit integer ID, where I represents the path that the vehicle is going to travel and D represents the order in which the same vehicle arrives at the intersection.

[0101] Step 3-2): Use the traffic sequence to indicate the traffic order of all vehicles, vehicle number ID i The ID of the i-th vehicle corresponding to path I is i The x-th occurrence of corresponds to its x-th passing process, and the length m of the passing sequence is the total number of passing sequences of all vehicles.

[0102] Step 3-3): Decode the encoded pass sequence into a transition sequence. i The yth occurrence of corresponds to the yth transition on path I.

[0103] Specifically, there are 6 vehicles about to enter the intersection in lane 1, and their I values ​​are all 1. The first vehicle to arrive at the buffer zone has a D value of 1 and an ID of 11; the second vehicle to arrive at the buffer zone has a D value of 2 and an ID of 12; the third vehicle to arrive at the buffer zone has a D value of 3 and an ID of 13, and so on. After obtaining the vehicle IDs, they are arranged in order of size into an array a0 of length 14: [11, 12, 13, 14, 15, 16, 21, 22, 23, 24, 25, 31, 32, 33, 34, 41, 42, 43, 44, 45, 51, 52, 53, 61, 71, 72, 73, 74, 81];

[0104] According to the decoding rules, the ID number in the pass sequence i The yth occurrence of corresponds to the yth transition on path I in the Petri net model. For example, the encoding array [11,11,12,21,22] can be decoded into the transition sequence [t 11 ,t 12 ,t 11 ,t 21 ,t 21 ];

[0105] Step 4) (Generate the initial chaotic population): Under the requirements of encoding and decoding in step 3), use the Circle chaotic map to generate the initial population with the number of chromosomes unit_num. The steps are as follows:

[0106] Step 4-1): Let the total number of vehicles that are about to pass through the intersection be θ, and arrange the θ vehicles in ascending order of number into array a0, and then expand the sequence according to the number of transitions that each vehicle needs to pass through in the Petri net path to obtain the expanded array a1. Let the total number in array a1 be m;

[0107] Step 4-2): Initialize the population array PopChrom, which contains unit_num empty arrays;

[0108] Step 4-3): Let v=1, and record the chromosome number to be generated;

[0109] Step 4-4): Let i=1 and record the number of iterations of the chaotic mapping;

[0110] Step 4-5): Generate a random number B (1) ∈(0,1);

[0111] Step 4-6): According to the Circle chaos mapping formula, we get C (i) , the Circle chaos mapping formula is as follows:

[0112]

[0113] C (i) =int(m×B (i) )+1

[0114] In the formula, λ is the control parameter, B (i) is the random number of the Circle chaotic map generated after the i-th iteration, C (i) Indicates the position number selected in the chromosome, int() is the rounding function;

[0115] Step 4-7): Replace the vehicle ID in array a1 with i Put it in PopChrom[v][C (i) ], i=i+1;

[0116] Step 4-8): If i>m, execute step 4-9), otherwise execute step 4-6);

[0117] Step 4-9): If v>unit_num, output the generated initial population PopChrom, otherwise v=v+1, and execute step 4-4);

[0118] Specifically, the 29 vehicles are sorted according to their encoded sizes as a0 = [11, 12, 13, 14, 15, 16, 21, 22, 23, 24, 25, 31, 32, 33, 34, 41, 42, 43, 44, 45, 51, 52, 53, 61, 71, 72, 73, 74, 81], and according to the traffic flow process, it is expanded to a1 = [11, 11, 11, 11, 12, 12, 12, 12, 12,13,13,13,13,13,14,14,14,14,14,15,15,15,15,15,16,16,16,16,16,21,21,21,21,21,22,22,22,22,23,23,23,23,23,24,24,24,24,25,25,25,25,31,31,31,31,31,3 2,32,32,32,32,33,33,33,33,33,34,34,34,34,34,41,41,41,41,41,42,42,42,42,43,43,43,43,43,44,44,44,45,45,45,45,51,51,51,51,51,51,51,52,52,52,52 ,52,52,53,53,53,53,53,53,53,61,61,61,61,61,61,61,71,71,71,71,71,71,71,72,72,72,72,72,73,73,73,73,73,73,74,74,74,74,74,81,81,81,81,81,81,81].Then, according to the Circle chaos mapping formula, a random chromosome is generated, one of which is: [52,13,72,24,34,43,16,61,12,71,32,44,53,15,72,21,31,41,52,14,74,22,45,51,11,73,23,81,33,42,52,13,74,25, 34,43,16,61,12,71,32,44,53,15,72,24,31,41,61,14,74,22,45,51,11,73,21,81,33,42,52,13,74,25,34,51,16,61,23,71,32,44,53,15,72,24,31,43,61,12,7 1,22,45,51,11,73,21,81,33,41,52,14,74,25,34,51,16,73,23,71,32,42,53,13,72,24,31,43,61,12,71,22,44,53,15,73,21,81,33,41,52,14,74,25,45,51,11 ,73,23,81,32,42,53,13,72,24,34,43,16,61,12,71,22,44,53,15,72,21,81,31,41,52,14,74,25,45,51,11,73,23,81,33,42]; Repeat this step unit_num times to obtain the initial population PopChrom.

[0119] Step 5) (Standardize the chromosomes):

[0120] Step 5-1): i=1, indicating the number of chromosomes in the population;

[0121] Step 5-2): Initialize a new chromosome array

[0122] Step 5-3): Decode PopChrom[i] into a transition sequence Tran;

[0123] Step 5-4): Initialize an array of the same length as Tran To store the changes that have been judged;

[0124] Step 5-5): j=1, indicating the sequence number of the transition in the transition sequence Tran;

[0125] Step 5-6): Traverse all paths, find the path corresponding to transition Tran[j], obtain the list of vehicles to be passed by the path, and sort the vehicle numbers from small to large into an array Cars;

[0126] Step 5-7): Let temp[j] = Tran[j] to store the determined transitions;

[0127] Step 5-8): Traverse the number of times Tran[j] appears in temp j , then Gene i [j] = Cars[count j ]; j = j + 1;

[0128] Step 5-9): If j>the length of the array Tran, the standardization of the current chromosome is completed, i=i+1, and step 5-10) is executed; otherwise, step 5-6) is executed;

[0129] Step 5-10): If i> the length of the population PopChrom, complete the standardization of all chromosomes in the entire population and output the standardized population; otherwise, execute step 5-2);

[0130] Specifically, a chromosome in the population is [71,23,41,51,15,72,24,81,31,43,52,14,74,21,33,44,13,61,16,73,32,42,53,11,74,25,34,45,52,12,71,22,41,51,15,72,24,81,31,43,52,14,71,2 3,33,44,13,61,16,73,32,42,53,11,74,21,34,45,61,12,71,22,41,51,15,72,25,81,31,43,52,14,71,23,33,51,13,61,24,73,32,42,53,11,74,21,34,44,61,16,73,2 2,41,51,15,72,25,81,31,45,52,12,71,23,33,51,13,72,24,73,32,43,53,14,74,21,34,44,61,16,73,22,42,53,11,72,25,81,31,45,52,12,71,23,41,51,15,72,24,8 1,32,43,53,14,74,21,33,44,13,61,16,73,22,42,53,11,74,25,81,34,45,52,12], in order to make the order of vehicle passage in the chromosome array consistent with reality and facilitate the detection of identical or similar chromosomes in subsequent operations, they are standardized so that the vehicle that arrives at the intersection first starts the passage process first.After the algorithm in step 5, the chromosome is standardized to [71,21,41,51,11,72,22,81,31,42,52,12,73,23,32,43,13,61,14,74,33,44,53,15,71,24,34,45,51,16,72,25,41,52,11,73,21,81, 31,42,51,12,71,22,32,43,13,61,14,74,33,44,53,15,72,23,34,45,61,16,71,24,41,52,11,73,25,81,31,42,51,12,71,21,32,52,13,61,22,74,33,43,5 3,14,72,23,34,44,61,15,73,24,41,51,11,74,25,81,31,45,52,16,71,21,32,51,12,72,22,73,33,42,53,13,74,23,34,43,61,14,72,24,44,53,15,73,25 ,81,31,45,52,16,71,21,41,51,11,72,22,81,32,42,53,12,74,23,33,43,13,61,14,73,24,44,52,15,74,25,81,34,45,53,16], the chromosomes after standardized processing are more in line with the actual vehicle traffic conditions;.

[0131] Step 6) (Detection and Repair): By iteratively analyzing the transitions and resources in the Petri net, deadlocks are detected and repaired to ensure that each chromosome in the population is deadlock-free.

[0132] Specifically, the transition sequence a corresponding to the above chromosome t =['t 71 ','t 21 ','t 41 ','t 51 ','t 11 ','t 71 ','t 21 ','t 81 ','t 31 ','t 41 ','t 51 ','t 11 ','t 71 ','t 21 ','t 31 ','t 41 ','t 11 ','t 61 ','t 11 ','t71 ','t 31 ','t 41 ','t 51 ','t 11 ','t 72 ','t 21 ','t 31 ','t 41 ','t 52 ','t 11 ','t 72 ','t 21 ','t 42 ','t 52 ','t 12 ','t 72 ','t 22 ','t 82 ','t 32 ','t 42 ','t 53 ','t 12 ','t 73 ','t 22 ','t 32 ','t 42 ','t 12 ','t 62 ','t 12 ','t 72 ','t 32 ','t 42 ','t 52 ','t 12 ','t 73 ','t 22 ','t 32 ','t 42 ','t 63 ','t 12 ','t 74 ','t 22 ','t 43 ','t 53 ','t 13 ','t 73 ','t 22 ','t 83 ','t 33 ','t 43 ','t 54 ','t 13 ','t 75 ','t 23 ','t 33 ','t 54 ','t 13 ','t 64 ','t23 ','t 73 ','t 33 ','t 43 ','t 53 ','t 13 ','t 74 ','t 23 ','t 33 ','t 43 ','t 65 ','t 13 ','t 74 ','t 23 ','t 44 ','t 55 ','t 14 ','t 74 ','t 23 ','t 84 ','t 34 ','t 43 ','t 55 ','t 13 ','t 76 ','t 24 ','t 34 ','t 56 ','t 14 ','t 75 ','t 24 ','t 75 ','t 34 ','t 44 ','t 54 ','t 14 ','t 75 ','t 24 ','t 34 ','t 44 ','t 66 ','t 14 ','t 76 ','t 24 ','t 44 ','t 55 ','t 14 ','t 76 ','t 24 ','t 85 ','t 35 ','t 44 ','t 56 ','t 14 ','t 77 ','t 25 ','t 45 ','t 57 ','t 15 ','t77 ','t 25 ','t 86 ','t 35 ','t 45 ','t 56 ','t 15 ','t 76 ','t 25 ','t 35 ','t 45 ','t 15 ','t 67 ','t 15 ','t 77 ','t 25 ','t 45 ','t 57 ','t 15 ','t 77 ','t 25 ','t 87 ','t 35 ','t 45 ','t 57 ','t 15 '], initial identification M0 = (6,0,0,0,0,0,5,0,0,0,0,0,4,0,0,0,0,0,5,0,0,0,0,0,3,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,4,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1), under the initial identification M0, check the enabling status of each transition and determine whether it will cause deadlock after initiation. Starting from u = 1, determine the transition t α =a t [1] = t 71 , find the transition t 71 When triggered, the token will decrease in the library r8. At this time, M0[r8]=1, then the transition t 71 Enable, and trigger t 71 Will not cause deadlock, so it triggers t 71 , and update the current identifier to M1=(6,0,0,0,0,0,5,0,0,0,0,0,4,0,0,0,0,0,5,0,0,0,0,0,3,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,3,1,0,0,0,0,0,0,1,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,1,1,1,1,1,1,1,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1), u=u+1. After judging the transition t71 -t 21 -t 41 -t 51 -t 11 After that, M0[t 71 -t 21 -t 41 -t 51 -t 11 >M1, judge t α =a t [u]=t 71 Is it allowed to be triggered? Find the transition t 71 When the trigger is triggered, the right-of-way resource of the library r8 will be reduced. At this time, M1[r8]=0, then t 71 If it is not enabled, select the transition t that is enabled under the flag M1 from the following transitions. 21 After the iteration loop, after judging each transition, a set of deadlock-free transition sequences a can be obtained. t =['t 71 ','t 21 ','t 41 ','t 51 ','t 11 ','t 22 ','t 23 ','t 24 ','t 25 ','t 12 ','t 42 ','t 43 ','t 44 ','t 45 ','t 52 ','t 53 ','t 54 ','t 55 ','t 56 ','t 57 ','t 72 ','t 71 ','t 21 ','t 81 ','t 31 ','t 41 ','t 51 ','t 11 ','t 73 ','t 74 ','t 75 ','t 76 ','t 77 ','t 22 ','t 32 ','t 42 ','t 43 ','t82 ','t 23 ','t 24 ','t 25 ','t 83 ','t 84 ','t 85 ','t 72 ','t 71 ','t 21 ','t 31 ','t 41 ','t 73 ','t 74 ','t 75 ','t 76 ','t 77 ','t 22 ','t 23 ','t 42 ','t 52 ','t 33 ','t 34 ','t 35 ','t 53 ','t 54 ','t 55 ','t 56 ','t 57 ','t 32 ','t 24 ','t 25 ','t 33 ','t 34 ','t 35 ','t 21 ','t 22 ','t 23 ','t 24 ','t 25 ','t 72 ','t 13 ','t 86 ','t 87 ','t 14 ','t 15 ','t 73 ','t 74 ','t 75 ','t 12 ','t 11 ','t 13 ','t 14 ','t 15 ','t 12 ','t 44 ','t 61 ','t 45 ','t 11 ','t71 ','t 31 ','t 41 ','t 51 ','t 62 ','t 43 ','t 76 ','t 77 ','t 63 ','t 64 ','t 65 ','t 66 ','t 67 ','t 32 ','t 42 ','t 52 ','t 33 ','t 34 ','t 35 ','t 53 ','t 54 ','t 55 ','t 56 ','t 57 ','t 41 ','t 31 ','t 32 ','t 33 ','t 21 ','t 22 ','t 23 ','t 24 ','t 25 ','t 72 ','t 13 ','t 14 ','t 15 ','t 12 ','t 11 ','t 13 ','t 14 ','t 15 ','t 73 ','t 74 ','t 75 ','t 12 ','t 11 ','t 13 ','t 14 ','t 15 ','t 12 ','t 44 ','t 45 ','t 43 ','t 42 ','t 44 ','t 43 ','t 13 ','t 45 ','t44 ','t 14 ','t 34 ','t 76 ','t 35 ','t 77 ','t 45 ','t 15 '], so the deadlock-free vehicle passage sequence is [71,21,41,51,11,21,21,21,21,11,41,41,41,41,51,51,51,51,51,51,71,72,22,81,31,42,52,12,71,71,71,71,71,71,22,31,42, 42,81,22,22,22,81,81,81,72,73,23,32,43,72,72,72,72,72,23,23,43,52,31,31,31,52,52,52,52,32,23,23,32,32,32,24,24,24,24,24,73,1 1,81,81,11,11,73,73,73,12,13,12,12,12,13,42,61,42,14,74,33,44,53,61,43,73,73,61,61,61,61,33,44,53,33,33,33,53,53,53,53,53,45 ,34,34,34,25,25,25,25,25,74,13,13,13,14,15,14,14,14,74,74,74,15,16,15,15,15,16,43,43,44,45,44,45,16,44,45,16,34,74,34,74,45,16].

[0133] Step 7): Calculate the maximum transit time and fitness value of a single chromosome, and output the optimal individual in the current population.

[0134] Specifically, the objective function of this scheduling method is to find the minimum value of the maximum travel time, so it is necessary to solve the maximum travel time and fitness value corresponding to all chromosomes in each generation of the population. Determine the idle time of the next right-of-way point that the vehicle is about to travel, compare it with the estimated departure time of the vehicle's previous right-of-way point, and take the larger value of the two as the start time of the current vehicle entering the next right-of-way point. After calculating all the travel processes of all vehicles, the maximum departure time in the system is the maximum travel time T of the entire travel sequence. max ; After obtaining the T corresponding to all chromosomes max After that, the maximum time T max That is MaxT, the minimum T maxThat is MinT, set the constant k to 0.2, and find the fitness value of each chromosome. The calculation formula is as follows:

[0135]

[0136] Step 8): Determine whether the termination condition gen>Maxgen is met, where gen is the number of iterations of the current population and Maxgen is the maximum number of iterations. If the termination condition is met, the optimal individual is output; if not, step 9) is executed;

[0137] Specifically, this example takes into account the scale of traffic flow and the scale of feasible solutions corresponding to the number of right-of-way points at a two-way four-lane intersection, and sets Maxgen to 500, that is, the algorithm iterates 500 times to reach the termination condition;

[0138] Step 9) (Genetic Operation): Perform three genetic operations of selection, crossover and mutation on the current population, and execute step 5) after obtaining a new generation of population;

[0139] Specifically, the chromosomes in the population are arranged from large to small according to the fitness value, and the first 10 chromosomes are selected, and then the crossover operation is performed. In order to promote the preservation of excellent genes and increase the diversity of the population, the adaptive crossover operator is used to perform the crossover operation of chromosomes, and the crossover probability P of the i-th chromosome is calculated. ci , the improved adaptive crossover rate formula is as follows:

[0140]

[0141] Among them, P c1 , P c2 is the control parameter, P c1 , P c2 ∈(0,1), in this example, P c1 Set to 0.8, P c2 Set to 0.2; γ is the decay rate, and in this example, γ is set to 2; adjust P according to the degree of chromosome aggregation in the population c1 , P c2 ,γ,, can realize a reasonable crossover strategy;

[0142] Calculate the crossover probability P of the i-th chromosome ci After that, generate a random number ran between (0,1). If ran <P ci , then perform partial mapping crossover on the chromosome, randomly select one of the chromosomes selected in the selection operation as the parent chromosome, randomly select two insertion points between the chromosomes, and exchange the fragments of the two parent chromosomes between the insertion points. Move the fragments between the insertion points in the original chromosome to the front of the process sequence, and then delete the genes that are the same as the inserted fragment genes from the front to the back;

[0143] In order to avoid the increase of the number of similar individuals in the population after the crossover operation, the Hamming Distance between chromosomes in the population is calculated, and the control threshold δ is set to 20. If the Hamming Distance of two chromosomes is less than 20, one of them is retained and the other is subjected to single-point mutation. A mutation point is randomly selected, and a mutation length is randomly determined. The genes with mutation lengths before and after the mutation point are swapped.

[0144] After iterating the above steps 500 times, the optimal sequence finally obtained in this example is: [11,11,11,21,21,12,51,51,11,12,11,22,41,41,41,41,41,31,31,31,31,31,32,32,32,32,33,33,33,33,33,34,34,3 4,34,34,21,21,21,22,12,12,22,22,22,51,51,51,12,51,51,13,13,13,13,13,81,61,61,61,81,14,14,71,71,14,14,14,71,42,42,42,23,23,23,23,23,24 ,24,24,24,24,43,43,72,72,42,42,43,43,43,25,25,25,25,61,71,71,71,71,71,61,72,72,72,72,61,61,72,52,52,44,53,53,81,81,73,73,15,15,16, 81,44,45,44,45,74,52,52,73,73,73,74,15,15,15,52,52,53,53,53,53,53,81,81,16,44,45,73,73,16,16,16,74,44,45,74,74,74,45,74], the corresponding transition sequence is: ['t 11 ','t 12 ','t 13 ','t 21 ','t 22 ','t 11 ','t 51 ','t 52 ','t 14 ','t 12 ','t 15 ','t 21 ','t 41 ','t 42 ','t 43 ','t 44','t 45 ','t 31 ','t 32 ','t 33 ','t 34 ','t 35 ','t 31 ','t 32 ','t 33 ','t 34 ','t 35 ','t 31 ','t 32 ','t 33 ','t 34 ','t 35 ','t 31 ','t 32 ','t 33 ','t 34 ','t 35 ','t 23 ','t 24 ','t 25 ','t 22 ','t 13 ','t 14 ','t 23 ','t 24 ','t 25 ','t 53 ','t 54 ','t 55 ','t 15 ','t 56 ','t 57 ','t 11 ','t 12 ','t 13 ','t 14 ','t 15 ','t 81 ','t 61 ','t 62 ','t 63 ','t 82 ','t 11 ','t 12 ','t 71 ','t 72 ','t 13 ','t 14 ','t 15 ','t 73 ','t 41 ','t 42 ','t 43 ','t 21 ','t 22','t 23 ','t 24 ','t 25 ','t 21 ','t 22 ','t 23 ','t 24 ','t 25 ','t 41 ','t 42 ','t 71 ','t 72 ','t 44 ','t 45 ','t 43 ','t 44 ','t 45 ','t 21 ','t 22 ','t 23 ','t 24 ','t 25 ','t 64 ','t 74 ','t 75 ','t 76 ','t 77 ','t 65 ','t 73 ','t 74 ','t 75 ','t 76 ','t 66 ','t 67 ','t 77 ','t 51 ','t 52 ','t 53 ','t 41 ','t 51 ','t 52 ','t 83 ','t 84 ','t 71 ','t 72 ','t 11 ','t 12 ','t 11 ','t 85 ','t 42 ','t 41 ','t 43 ','t 42 ','t 71 ','t 54 ','t 55 ','t 73 ','t 74 ','t 75','t 72 ','t 13 ','t 14 ','t 15 ','t 56 ','t 57 ','t 53 ','t 54 ','t 55 ','t 56 ','t 57 ','t 86 ','t 87 ','t 12 ','t 44 ','t 43 ','t 76 ','t 45 ','t 44 ','t 45 ','t 13 ','t 14 ','t 15 ','t 73 ','t 74 ','t 75 ','t 77 ','t 76 ','t 77 '], and the corresponding vehicle travel time is 15.98s.

[0145] This method and the method in Reference 1 were used to perform optimization five times in the same scenario, and the results are as follows:

[0146] Table 2 Comparison of multiple optimization results of two algorithms

[0147]

[0148]

[0149] After verification by examples, the quality of the solution obtained by the scheduling strategy based on reference 1 is lower than that of this method. The significant advantage of this method is that it uses a deadlock detection and repair algorithm to promptly repair the randomly generated solution, ensuring that there will be no circular waiting of vehicles during the scheduling process. At the same time, through the chromosome standardization algorithm and the improved genetic algorithm, this method can search the solution space more efficiently and quickly find the optimal solution. This scheduling strategy allows the right-of-way resources of the intersection to be utilized to the greatest extent, thereby significantly improving the traffic efficiency of the intersection.

[0150] The specific implementation scheme described above further describes in detail the purpose, technical scheme and beneficial effects of the present invention. It should be understood that the above is only a specific implementation scheme of the present invention and is not intended to limit the scope of the present invention. Any equivalent changes and modifications made by any technician in the field without departing from the concept and principle of the present invention should fall within the scope of protection of the present invention.

Claims

1. A method for dispatching vehicles at unsignalized intersections in an intelligent network environment, characterized in that: The following steps are involved: S1. Modeling: Based on the spatial resources in the internal area of ​​the intersection and the occupation of various resources in the continuous process of vehicles passing through the intersection, a Petri net model (N, M0) of the unsignalized intersection system is constructed, and the following symbols are defined: N: A Petri net N = (P, T, F) consisting of circular nodes, square nodes and directed arcs, representing an intersection traffic system consisting of k right-of-way points and allowing vehicles on n paths to pass through; P: Place set, which is the set of all circular nodes in N. Where Pi0=Pis∪Pif, Pis is the buffer zone before the vehicle enters the intersection, and Pif is the buffer zone when the vehicle leaves the intersection; d represents the total number of right-of-way points on path i, j represents the jth right-of-way point passed by the vehicle, and Pij represents the jth passage process of the vehicle on path i through the intersection; PR = {Pr}, Pr represents the library set composed of all right-of-way points; T: transition set, consisting of all square nodes in N, in d represents the total number of right-of-way points on path i, and the initiation of tij means that the vehicle on the i-th path travels from the j-1th right-of-way point to the jth right-of-way point; F: directed arc set, representing the driving conditions of each vehicle in the system and the demand and release of resources when passing each right-of-way point during driving; M: P→N is the identifier, N is a non-negative integer set, indicating the operating status of the system, and each number of the identifier represents the number of vehicles or resources contained in each library; M0 is the initial identifier, indicating that the current vehicles have not entered the inner area of ​​the intersection, the vehicles are in the buffer zone, and the resources are not occupied; A: The incidence matrix, which represents the changing pattern of the number of tokens in each repository under each transition in N. It is a matrix with |T| rows and |P| columns. For the unsignalized intersection system, the objective function of system scheduling is to minimize the maximum travel time, that is: F=min{max{C l }} T max =max{C l } Among them, C l is the time when the kth vehicle leaves the intersection, T max The maximum travel time for all vehicles to pass through; S2, the intelligent roadside dispatching system obtains vehicle information in the unsignalized intersection area, including the number of vehicles, speed, and lane number; S3, encoding and decoding: number the vehicles that are about to arrive at the intersection area and generate a passage sequence according to the passage process of each vehicle passing through each road right point on its path, and decode it into a transition sequence; S4, generate chaotic initial population: in accordance with the requirements of S3 encoding and decoding, use Circle chaotic mapping to generate an initial population with chromosome number unit_num; S5, standardizing the chromosomes: In order to reduce the algorithm search space, all chromosomes are limited to the coding subspace containing all feasible scheduling solutions, and the chromosomes are standardized to reduce the number of equivalent chromosomes, and then S6 is executed; S6, Detection and Repair: By iteratively analyzing the transitions and resources in the Petri net, deadlocks are detected and repaired to ensure that each chromosome in the population is free of deadlocks; S7, calculate the maximum transit time and fitness value of a single chromosome, and output the optimal individual of the current population; S8, determine whether the termination condition gen>Maxgen is met, where gen is the number of iterations of the current population and Maxgen is the maximum number of iterations; if the termination condition is met, the optimal individual is output, if not, execute S9; S9, genetic operation: perform the three genetic operations of selection, crossover and mutation on the current population, and execute S5 after obtaining the new generation of population.

2. The method for dispatching vehicles at unsignalized intersections in an intelligent network environment according to claim 1, characterized in that: The specific steps of S3 are as follows: S3-1. Number the vehicles according to their travel paths and the order in which they arrive at the intersection. The number is a two-digit integer ID, where I represents the path that the vehicle is going to travel and D represents the order in which the same vehicle arrives at the intersection. S3-2, use the traffic sequence to indicate the traffic order of all vehicles, vehicle number ID i The ID of the i-th vehicle corresponding to path I is i The x-th occurrence of corresponds to its x-th passing process, and the length of the passing sequence m is the total number of passing sequences of all vehicles; S3-3, decode the encoded pass sequence into a transition sequence, and the ID in the pass sequence i The yth occurrence of corresponds to the yth transition on path I.

3. The method for dispatching vehicles at unsignalized intersections in an intelligent network environment according to claim 2, characterized in that: The specific steps of S4 are as follows: S4-1, record the total number of vehicles that are about to pass through the intersection as θ, and arrange the θ vehicles in ascending order of number into an array a0, and then expand the sequence according to the number of transitions that each vehicle needs to pass in the path of the Petri net to obtain the expanded array a1, and record the total number of numbers in the array a1 as m; S4-2, initialize the population array PopChrom, which contains unit_num empty arrays; S4-3, let v=1, record the chromosome number to be generated; S4-4, let i=1, record the number of iterations of the chaotic mapping; S4-5. Generate a random number B (1) ∈(0,1); S4-6. According to the Circle chaos mapping formula, we get C (i) , the Circle chaos mapping formula is as follows: C (i) =int(m×B (i) )+1 In the formula, λ is the control parameter, B (i) is the random number of the Circle chaotic map generated after the i-th iteration, C (i) Indicates the position number selected in the chromosome, int() is the rounding function; S4-7, the vehicle number ID in array a1 i Put it in PopChrom[v][C (i) ], i=i+1; S4-8, if i>m, execute S4-9, otherwise execute S4-6; S4-9. If v>unit_num, output the generated initial population PopChrom, otherwise v=v+1, and execute S4-4.

4. The method for dispatching vehicles at unsignalized intersections in an intelligent network environment according to claim 3, characterized in that: In S5, chromosomes are standardized. The specific steps are as follows: S5-1, i=1, represents the sequence number of the chromosome in the population; S5-2. Initialize a new chromosome array S5-3, decode PopChrom[i] into transition sequence Tran; S5-4. Initialize an array with the same length as Tran To store the changes that have been judged; S5-5, j=1, indicating the sequence number of the transition in the transition sequence Tran; S5-6, traverse all paths, find the path corresponding to transition Tran[j], obtain the list of vehicles to be passed by the path, and sort the vehicle numbers from small to large into an array Cars; S5-7, let temp[j] = Tran[j], store the determined transition; S5-8. traverse the number of times Tran[j] appears in temp j , then Gene i [j] = Cars[count j ]; j = j + 1; S5-9, if j> the length of the array Tran, then the standardization of the current chromosome is completed, i=i+1, and S5-10 is executed; otherwise, S5-6 is executed; S5-10. If i> the length of the population PopChrom, complete the standardization of all chromosomes in the entire population and output the standardized population; otherwise, execute S5-2.

5. The method for dispatching vehicles at unsignalized intersections in an intelligent network environment according to claim 4, characterized in that: The specific steps of S6 are as follows: S6-1, set u=1, record the transition number currently detected; S6-2. If u is greater than the length of the transition sequence, execute S7. Otherwise, set the uth transition in the transition sequence to t. α , execute S6-3; S6-3, check the transition t α Is it enabled under the current flag? If it is enabled, execute S6-4. Otherwise, start from t α Then select an enabled transition and place it at t α Before, and update t α ; S6-4. Use the one-step forward method to check the change t α Can it cause deadlock? If it will cause deadlock, select t α The subsequent enabling transition is placed before this transition and updates t α , re-execute S6-4; otherwise, trigger t α , and update the current flag, set u=u+1, and execute S6-2.

6. The method for dispatching vehicles at unsignalized intersections in an intelligent network environment according to claim 5, characterized in that: S7 specific steps are as follows: S7-1. Calculate the maximum travel time: determine the idle time of the next right-of-way point that the vehicle is about to travel, compare it with the estimated departure time of the vehicle's previous right-of-way point, and take the larger value of the two as the start time of the current vehicle entering the next right-of-way point; after calculating all the travel processes of all vehicles, the maximum departure time in the system is the maximum travel time T of the entire travel sequence max ; S7-2, calculate the fitness value: calculate the fitness value Adapt, the calculation formula is as follows: Among them, MaxT is the maximum transit time in the chromosome of the current population, MinT is the minimum transit time in the chromosome of the current population, and T max is the transit time of the current chromosome, k is a constant; S7-3. Output the best individual: Output the best chromosome in the population and the corresponding T max .

7. The method for dispatching vehicles at unsignalized intersections in an intelligent network environment according to claim 6, characterized in that: The genetic operation in S9 has the following specific steps: S9-1, selection operation: Arrange the chromosomes in the population from large to small according to the fitness value, select the first Selectnum×unit_num chromosomes to join the next generation, where Selectnum is the selection factor and unit_num is the population size; S9-2, Crossover operation: In order to promote the preservation of excellent genes and increase the diversity of the population, an adaptive crossover operator is used to perform a chromosome crossover operation. The specific steps are as follows: S9-2-1, i=1, represents the chromosome number in the population; S9-2-2. Calculate the fitness value f of each chromosome in the population i ; S9-2-3. Calculate the average fitness value f of the population avg , maximum fitness value f max ; S9-2-4. According to the improved adaptive crossover rate formula, calculate the crossover probability P of the i-th chromosome ci , the formula is as follows: Among them, P c1 , P c2 is the control parameter, P c1 , P c2 ∈(0,1);γ is the decay rate; S9-2-5, generate a random number ran between (0,1), if ran <P ci , then perform partial mapping crossover on the chromosome, randomly select the chromosome and one of the chromosomes selected in S9-1 as the parent chromosome, randomly select two insertion points between the chromosomes, and exchange the fragments of the two parent chromosomes between the insertion points; move the fragments between the insertion points in the original chromosome to the front of the process sequence, and then delete the genes that are the same as the inserted fragment genes from the front to the back; let i=i+1; S9-2-6, if i>unit_num, output the new generation population, otherwise execute S9-2-4; S9-3, mutation operation: In order to avoid the increase of the number of similar individuals in the population after the crossover operation, the chromosomes with differences less than a certain threshold are mutated. The specific steps are as follows: S9-3-1. Calculate the Hamming Distance between chromosomes in the population, which indicates the number of different numbers at the corresponding positions of two chromosomes; S9-3-2, set the control threshold δ. If the Hamming Distance of two chromosomes is less than δ, one of them will be randomly retained and the other will execute S9-3-3. S9-3-3. Randomly select a mutation point in the chromosome and randomly determine a mutation length, and swap the genes of the mutation length before and after the mutation point.

Images