A mixed traffic-oriented intersection processing method and device

Abstract

The application discloses a kind of intersection processing methods and devices for mixed traffic, according to road network structure and traffic state information, the best special intersection deployment scheme is determined, wherein, special intersection in road network structure allows network-connected automatic driving vehicle to pass but does not allow human-driven vehicle to pass, the intersection of remaining in road network structure allows network-connected automatic driving vehicle and human-driven vehicle to pass.Through the special intersection deployment scheme of the embodiment of the application, the travel time of network-connected automatic driving vehicle is reduced, and the influence on human-driven vehicle is considered simultaneously, and the efficiency of the whole mixed traffic system is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to, but is not limited to, intelligent transportation technology, in particular to a mixed traffic intersection processing method and device. BACKGROUND

[0002] From the conventional human driving environment to the full automatic driving environment, there will be a long period of time for the mixed traffic environment of two types of vehicles. The management and control of the mixed traffic system of two types of vehicles will also become an important problem of the current intelligent transportation system research.

[0003] In the mixed traffic environment, there are both connected and automated vehicles (CAVs) and human-driven vehicles (HDVs) on the road. Due to more accurate driving behavior and shorter reaction time, connected and automated vehicles show unprecedented efficiency and safety, which has great potential to improve the efficiency of individual vehicles and the operation efficiency of the entire traffic system. However, due to the uncertainty and randomness of human driving behavior, the potential performance of connected and automated vehicles is greatly discounted in the mixed traffic environment. Therefore, for the mixed traffic system, how to control and manage to take advantage of connected and automated vehicles becomes the core problem of improving the efficiency of the mixed traffic system. SUMMARY

[0004] The present application provides a mixed traffic intersection processing method and device, which can reduce the travel time of connected and automated vehicles, while considering the influence on human-driven vehicles, and can improve the efficiency of the entire mixed traffic system.

[0005] The embodiment of the present application provides a mixed traffic intersection processing method, comprising:

[0006] According to the road network structure and traffic state information, a cross in the road network structure is determined as a first intersection; wherein the first intersection allows connected and automated vehicles to pass but does not allow human-driven vehicles to pass, and the remaining intersections in the road network structure are second intersections allowing connected and automated vehicles and human-driven vehicles to pass;

[0007] According to the determined result, the cross in the road network structure is set as the first intersection, and a signal-free cooperative driving algorithm is deployed at the first intersection.

[0008] In an exemplary example, the traffic state information includes traffic demand, penetration rate.

[0009] In an example embodiment, the intersection in the road network structure determined as the first intersection based on the artificial bee colony algorithm.

[0010] In an example embodiment, the intersection in the road network structure determined as the first intersection based on the artificial bee colony algorithm comprises:

[0011] from 2 n a special intersection deployment scheme is randomly selected for each leader bee from the 2 n special intersection deployment schemes, and the counter corresponding to the special intersection deployment scheme of each leader bee is set to zero; wherein n is the number of intersections in the road network structure;

[0012] According to the pre-set neighborhood search rule, the neighborhood search is performed and the special intersection deployment scheme corresponding to each leader bee is determined based on the current special intersection deployment scheme of each leader bee by using the leader bee foraging process.

[0013] According to the pre-set neighborhood search rule, the neighborhood search is performed and the special intersection deployment scheme corresponding to each leader bee is determined based on the current special intersection deployment scheme of each leader bee by using the leader bee foraging process.

[0014] The iteration neighborhood search number is increased by one; for the case that the current iteration neighborhood search number reaches the iteration neighborhood search number threshold, the scout bee randomly selects a special intersection deployment scheme for each leader bee from the 2 n special intersection deployment schemes, and the counter corresponding to the special intersection deployment scheme of each leader bee is set to zero, and the next step is entered; for the case that the current iteration neighborhood search number does not reach the iteration neighborhood search number threshold, the next step is entered.

[0015] The target function values corresponding to the special intersection deployment schemes of all leader bees are compared, and the special intersection deployment scheme with the minimum target function value is recorded.

[0016] The global iteration number is increased by one; for the case that the current global iteration number does not reach the global iteration number threshold, the step of performing neighborhood search and determining the special intersection deployment scheme corresponding to each leader bee by using the leader bee foraging process is returned; for the case that the current global iteration number reaches the global iteration number threshold, the special intersection displayed in the recorded special intersection deployment scheme is displayed as the intersection in the road network structure determined as the first intersection.

[0017] In an example embodiment, the neighborhood search is performed and the special intersection deployment scheme corresponding to each leader bee is determined by using the leader bee foraging process, comprising:

[0018] In a preset range of the current special intersection deployment scheme of each scout bee, neighborhood search is performed according to a preset neighborhood search rule to obtain a more optimal special intersection deployment scheme in a local range.

[0019] In an exemplary instance, the neighborhood search in a preset range of the current special intersection deployment scheme of each scout bee according to a preset neighborhood search rule includes:

[0020] For a certain scout bee, neighborhood search is performed in a preset range of the current special intersection deployment scheme of the scout bee according to the preset neighborhood search rule, and a randomly selected special intersection deployment scheme X k1 is selected as a new special intersection deployment scheme X k2 .

[0021] The new special intersection deployment scheme X j′ is compared with the old, i.e., the current special intersection deployment scheme X upper of the scout bee. j′ ) corresponding to the new special intersection deployment scheme X j (X upper ) corresponding to the old, i.e., the current special intersection deployment scheme X j . If the performance of the new special intersection deployment scheme X j′ is better, the old special intersection deployment scheme X j′ is replaced by the new special intersection deployment scheme X j , and the counter corresponding to the new special intersection deployment scheme X j′ is set to zero; if the performance of the new special intersection deployment scheme X j′ is not better than that of the old, i.e., the current special intersection deployment scheme X j , the old special intersection deployment scheme X j is retained, and the counter corresponding to the old special intersection deployment scheme X j is increased by one.

[0022] The objective function f Upper () is expressed as follows:

[0023]

[0024] wherein, N is a set of intersections in the road network structure, x i / x i‘ is each element in the set N; the decision variable X is represented as X={x1,x2,…,x n}, wherein x i / x i‘= 1 (i / i' = 1, 2,..., n) means that the intersection i / i' is set as the first intersection, x i = 0 (i / i' = 1, 2,..., n) means that the intersection i / i' is set as the second intersection. i‘ = 0 (i / i' = 1, 2,..., n) means that the intersection i / i' is set as the second intersection.

[0025] v ii‘ represents the flow on the edge (i, i'), edge (i, i') e A, A represents the set of edges connecting intersections; t ii‘ , t' ii‘ , t" ii′ , t" ii‘ respectively represent the travel time under four different intersection deployment schemes.

[0026] In an exemplary instance, the searching and determining of the dedicated intersection deployment scheme corresponding to each leader bee by the following-bee honey harvesting process comprises:

[0027] Based on the roulette strategy, a leader bee is selected for each following bee;

[0028] For each leader bee, each following bee of the leader bee performs neighborhood search on the leader bee within a preset range of the current dedicated intersection deployment scheme of the leader bee according to the pre-set neighborhood search rule, so as to obtain the dedicated intersection deployment scheme of each following bee in a local range;

[0029] According to the objective function value corresponding to the dedicated intersection deployment scheme, the performance of the dedicated intersection deployment scheme generated by each following bee and the dedicated intersection deployment scheme of the leader bee is compared respectively, if the performance of the dedicated intersection deployment scheme generated by the following bee is better, the dedicated intersection deployment scheme generated by the following bee is used to replace the dedicated intersection deployment scheme corresponding to the leader bee, until the comparison of all following bees of the leader bee is completed, the counter corresponding to the dedicated intersection deployment scheme of the leader bee is reset to zero; if the performance of the dedicated intersection deployment scheme generated by all following bees is not better than the performance of the dedicated intersection deployment scheme of the leader bee, the dedicated intersection deployment scheme of the leader bee is retained, and the counter corresponding to the dedicated intersection deployment scheme of the leader bee is increased by one.

[0030] In an exemplary instance, the pre-set neighborhood search rule is as follows:

[0031]

[0032] wherein X represents the decision variable of the dedicated intersection solving problem, represents the dedicated intersection deployment scheme X i of the i-th intersection x j; k∈{1,2,…,M} and i∈{1,2,…,n} represent randomly selected subscripts, and k≠j, M represents the colony size in the artificial bee colony algorithm parameters;

[0033] The deployment scheme X from randomly selected dedicated intersections j Dedicated intersection deployment solution X k Choose the better one as the new dedicated intersection deployment scheme. include:

[0034] when At that time, the new solution The value is or

[0035] when When the dedicated intersection deployment scheme X k Objective function value and dedicated intersection deployment scheme X j Objective function value and dedicated cross-section deployment scheme X k The new solution is a random number between 0 and 1 whose sum of objective function values ​​is greater than the ratio of these values. The value is the value of the dedicated intersection deployment scheme. When the dedicated intersection deployment scheme X k The objective function value and the dedicated intersection deployment scheme X j The objective function value and the dedicated intersection deployment scheme X k The ratio of the sum of the objective function values ​​to a random number between 0 and 1 is the new solution. The value is a dedicated intersection deployment scheme.

[0036] In one exemplary instance, the selection of a leader bee for each follower bee based on the roulette wheel strategy includes:

[0037] The following formula is used as the selection probability to select p for each following bee. j Large lead bee:

[0038]

[0039] Where M represents the colony size in the artificial bee colony algorithm parameters.

[0040] This application also provides a computer-readable storage medium storing computer-executable instructions for performing any of the above-described methods for handling mixed traffic intersections.

[0041] This application embodiment further provides an apparatus for implementing intersection processing for mixed traffic, including a memory and a processor, wherein the memory stores the following instructions executable by the processor: steps for performing the intersection processing method for mixed traffic described in any of the above claims.

[0042] This application embodiment provides another intersection processing device for mixed traffic, including: a deployment unit and a setting unit; wherein,

[0043] The deployment unit is used to determine the first intersection in the road network structure based on the road network structure and traffic status information; wherein the first intersection allows connected autonomous vehicles to pass but does not allow human-driven vehicles to pass, and the remaining intersections in the road network structure are the second intersections that allow both connected autonomous vehicles and human-driven vehicles to pass.

[0044] The setting unit is used to set the intersection that is the first intersection in the road network structure as the first intersection according to the determined result, and to deploy the signalless cooperative driving algorithm at the first intersection.

[0045] In one exemplary instance, the deployment unit includes: an initialization module, a first search module, a second search module, a first processing module, a recording module, and a second processing module; wherein,

[0046] The initialization module is used to initialize the parameters of the artificial bee colony algorithm; from 2 n In the dedicated intersection deployment scheme, each leader bee is randomly selected from an intersection deployment scheme, and the counter corresponding to the dedicated intersection deployment scheme of each leader bee is set to zero;

[0047] The first search module is used to perform a neighborhood search based on the current dedicated intersection deployment scheme of each leader bee and the honey-collecting process of the leader bee, according to the pre-set neighborhood search rules.

[0048] The second search module is used to search for and determine the dedicated intersection deployment scheme corresponding to each leader bee based on the determined dedicated intersection deployment scheme corresponding to each leader bee and the honey-collecting process of the follower bees, according to the pre-set neighborhood search rules.

[0049] The first processing module increments the iteration neighborhood search count by one; it then determines whether the current iteration neighborhood search count has reached the iteration neighborhood search count threshold. If the current iteration neighborhood search count has reached the threshold, for each leader bee, the reconnaissance bee increments the count from 2... nIn the dedicated intersection deployment scheme, a dedicated intersection deployment scheme is randomly selected for the leading bee, and the counter corresponding to the dedicated intersection deployment scheme of the leading bee is set to zero, and the process enters the recording module; if the current iteration neighborhood search count has not reached the iteration neighborhood search count threshold, the process enters the recording module.

[0050] The recording module is used to compare the objective function values ​​corresponding to all the dedicated intersection deployment schemes of the leading bees, and record the dedicated intersection deployment scheme with the smallest objective function value;

[0051] The second processing module increments the global iteration count by one and determines whether the current global iteration count has reached the global iteration count threshold. If it has not reached the threshold, the search in the first search module continues; if it has, the neighborhood search is terminated, and the dedicated intersections displayed in the dedicated intersection deployment scheme recorded in the recording module are taken as the first intersections in the road network structure.

[0052] The intersection handling method for mixed traffic provided in this application determines the optimal deployment scheme for dedicated intersections based on road network structure and traffic state information (such as traffic demand and penetration rate). In this scheme, dedicated intersections in the road network structure allow connected autonomous vehicles to pass but not human-driven vehicles, while the remaining intersections in the road network structure allow both connected autonomous vehicles and human-driven vehicles to pass. This dedicated intersection deployment scheme reduces the travel time of connected autonomous vehicles while simultaneously considering the impact on human-driven vehicles, thus improving the efficiency of the entire mixed traffic system.

[0053] Furthermore, the intelligent algorithm based on artificial bee colony proposed in this application, based on traffic state information such as penetration rate and traffic demand, solves for a near-globally optimal dedicated intersection setting scheme within a limited number of iterations, thus achieving the selection of the best dedicated intersection setting scheme for the road network structure. Through this application, while improving the efficiency of connected autonomous vehicles, the negative impact of setting up dedicated intersections within the road network on human-driven vehicles is reduced, achieving global optimization of the traffic system.

[0054] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims, and drawings. Attached Figure Description

[0055] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0056] Figure 1 This is a flowchart illustrating the intersection handling method for mixed traffic in an embodiment of this application.

[0057] Figure 2 This is a schematic diagram illustrating the solution framework for the two-layer dedicated intersection deployment problem in this application embodiment;

[0058] Figure 3 This is a schematic diagram illustrating the process of determining the first intersection in the road network structure based on the artificial bee colony algorithm in an embodiment of this application.

[0059] Figure 4 This is a schematic diagram of a typical road network structure used for simulation verification in the embodiments of this application;

[0060] Figure 5 This is a schematic diagram of the composition of an intersection processing device for mixed traffic in an embodiment of this application. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in detail below with reference to the accompanying drawings. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be arbitrarily combined with each other.

[0062] The steps illustrated in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases the steps shown or described may be performed in a different order than that presented here.

[0063] The performance advantages of connected autonomous vehicles can be divided into two dimensions: a one-dimensional advantage of improved following performance and a two-dimensional advantage of effectively resolving right-of-way conflicts. In recent years, researchers have proposed numerous control strategies for mixed traffic systems to leverage the performance advantages of connected autonomous vehicles. Among these strategies, setting up dedicated lanes for connected autonomous vehicles has become the most popular approach. The dedicated lane strategy for connected autonomous vehicles refers to selecting certain lanes on existing multi-vehicle road sections specifically for their use. These lanes restrict access to human-driven vehicles, reducing interference from human drivers. The dedicated lane strategy separates connected autonomous vehicles from mixed traffic flow, creating a localized homogeneous traffic flow. Then, thanks to the shorter and more stable following distances achieved by connected autonomous vehicles, their driving performance on the road segment can be improved. Essentially, the dedicated lane strategy aims to leverage the one-dimensional performance advantage of autonomous vehicles, resulting in performance improvements in scenarios such as highways where there are no right-of-way conflicts.

[0064] Compared to one-dimensional following distance, two-dimensional right-of-way conflicts have a more dominant impact on road network traffic efficiency. In other words, in urban road network scenarios full of right-of-way conflicts (such as intersections and ramps), shorter following distances have limited effect on improving travel efficiency. Compared to shorter following distances, better intersection control strategies can significantly improve the efficiency of individual vehicles and the entire traffic system.

[0065] In order to fully leverage the advantages of connected and automated vehicles in mixed traffic systems and improve the efficiency of mixed traffic systems, this application provides an intersection handling method for mixed traffic. Figure 1 This is a flowchart illustrating the intersection handling method for mixed traffic in an embodiment of this application, as shown below. Figure 1 As shown, it may include:

[0066] Step 100: Based on the road network structure and traffic status information, determine the intersection in the road network structure that serves as the first intersection; wherein, the first intersection allows connected autonomous vehicles to pass but does not allow human-driven vehicles to pass, and the remaining intersections in the road network structure serve as the second intersections that allow both connected autonomous vehicles and human-driven vehicles to pass.

[0067] In this embodiment, connected autonomous vehicles are allowed to pass through the first intersection (also known as a dedicated intersection), while human-driven vehicles are not allowed to pass through the first intersection; the second intersection (also known as a conventional intersection) allows both connected autonomous vehicles and human-driven vehicles to pass. The second intersection can be a traditional signal-controlled intersection, or it can be another hybrid signal-controlled intersection, such as adding a dedicated color phase that only allows CAVs to pass through the intersection. It should be noted that this embodiment does not limit the control method of the second intersection, nor does the control method of the second intersection limit the scope of protection of this application.

[0068] To leverage the two-dimensional advantages of connected autonomous vehicles in road network traffic, step 100 determines a dedicated intersection deployment scheme within the road network structure based on the road network structure and traffic conditions; that is, which intersections in the road network structure are designated as primary intersections. In one embodiment, the dedicated intersection deployment scheme can refer to: improving the efficiency of connected autonomous vehicles while reducing the negative impact of setting up dedicated intersections within the road network on human-driven vehicles, thereby achieving global optimization of the traffic system. To this end, embodiments of this application propose a two-layer framework for solving the dedicated intersection deployment problem, such as... Figure 2As shown, the upper-level problem is to determine which intersections in the road network structure should be designated as dedicated intersections, while the lower-level problem is to solve the corresponding traffic assignment problem. This involves evaluating the traffic assignment based on the road network structure, including the dedicated intersection deployment scheme determined at the upper level. Factors include traffic flow on roads connecting intersections and travel time. The evaluation results are then used in the implementation of the upper-level dedicated intersection deployment scheme. Thus, the solution to the upper-level problem—the dedicated intersection deployment scheme—constrains the lower-level traffic assignment problem, and the results of the lower-level traffic assignment accurately evaluate the performance of the upper-level dedicated intersection deployment scheme.

[0069] In one exemplary instance, traffic status information may include, but is not limited to, traffic demand, penetration rate, etc.

[0070] In one exemplary instance, the road network structure consists of multiple intersections and roads connecting different intersections. To more clearly illustrate the deployment of dedicated intersections, in one exemplary instance, a graph data structure can be used to describe the urban traffic road network, i.e., the road network structure in this embodiment: G = (N, A), where N represents the set of nodes (intersections), and A represents the set of edges (roads connecting intersections). In one embodiment, i, i' = 1, 2, ..., n are used to represent intersections; then, the edge connecting two intersections, i.e., the lane, can be represented as (i, i') ∈ A.

[0071] In this embodiment, O and D represent the sets of origin and destination, respectively, where O, D ∈ N. For a certain origin o, o ∈ O, and for a certain destination d, d ∈ D; W represents the set of all origin-destination (OD) pairs in the road network structure (an OD pair is denoted as w). For an OD pair w, w ∈ W, an OD pair w will connect an origin o and a destination d.

[0072] Based on the above analysis, the design of dedicated intersections requires a trade-off between the benefits to connected autonomous vehicles and the impact on human-driven vehicles. Therefore, in this embodiment, the objective function f, as shown in formula (1), can be utilized. Upper To achieve overall optimization of the hybrid transportation system:

[0073]

[0074] In formula (1), That is, for each element in the set N of intersections, i.e., each intersection, x i / x i‘ ∈{0,1}. The objective function f in the embodiments of this application UpperIt represents the sum of the traffic flow multiplied by the corresponding travel time on all edges in the road network architecture, which is the same as the objective function for system optimization in the traffic assignment problem.

[0075] In formula (1), {x1, x2, ..., x n Let} be the decision variable X, which can be represented as X = {x1, x2, ..., x} n}, where x i / x i‘ =1 (i / i' = 1, 2, ..., n) indicates that intersection i / i' is set as a dedicated intersection, i.e., the first intersection, x i / x i‘ =0 (i / i' = 1, 2, ..., n) indicates that intersection i / i' is set to a regular traffic light controlled intersection, i.e., the second intersection.

[0076] In formula (1), v ii‘ Let v represent the flow on edge (i, i'), where edge (i, i') ∈ A. In one embodiment, v ii‘ It can be determined through the corresponding underlying traffic allocation. In one embodiment, it can be obtained by simulating and evaluating the road network structure, including the dedicated intersection deployment scheme determined at the upper layer.

[0077] In formula (1), t ii‘ , t′ ii‘ 、t″ ii‘ 、t″′ ii‘ These represent the travel times under four different intersection deployment schemes, where t ii‘ t′ represents the travel time when intersections i and i' are both regular intersections, i.e., the second intersection. ii‘ This indicates that intersection i is a dedicated intersection, i.e., the first intersection, and intersection i' is a regular intersection, i.e., the second intersection, with travel time t″. ii‘ This indicates that intersection i is a regular intersection (i.e., the second intersection), and intersection i' is a dedicated intersection (i.e., the first intersection) with travel time t″′. ii‘ This represents the travel time when both intersections i and i' are dedicated intersections, i.e., the first intersection. t ii‘ , t′ ii‘ 、t″ ii‘ 、t″′ ii‘ It can be determined based on the underlying traffic allocation and road travel time. In one embodiment, it can be obtained by simulating and evaluating the road network structure, including the dedicated intersection deployment scheme determined at the upper level.

[0078] For a road network structure containing n intersections, there are a total of 2 dedicated intersection deployment schemes at the upper layer. nThe solution space for this type of problem grows exponentially with the number of intersections, meaning that the dedicated intersection deployment problem in this application is an NP-hard problem. NP-hard problems refer to a set of problems that are considered at least as difficult as nondeterministic polynomial-time (NP) problems in computational complexity theory.

[0079] In one exemplary instance, the intersection serving as the first intersection in the road network structure can be determined based on the artificial bee colony algorithm. The method for finding the optimal dedicated intersection deployment scheme based on the artificial bee colony algorithm in this application aims to find a near-optimal deployment scheme for dedicated intersections in the road network structure within a finite time. This application embodiment utilizes the artificial bee colony algorithm to iteratively search for better bee sources to obtain a better dedicated intersection deployment scheme. Figure 3 This is a schematic diagram illustrating the process of determining the first intersection in the road network structure based on the artificial bee colony algorithm in an embodiment of this application. Figure 3 As shown, it may include:

[0080] Step 300: Initialize the artificial bee colony algorithm parameters. In one embodiment, parameters such as the colony size M, the global iteration threshold L1, and the iteration neighborhood search threshold L2 for a single nectar source that has not been updated are set.

[0081] In one exemplary instance, the bee colony includes leader bees, follower bees, and scout bees. In practical applications, the size of the bee colony can be set according to the size of the road network structure, that is, the number of intersections n. The more intersections n, the larger the bee colony and the more leader bees; the fewer intersections n, the smaller the bee colony and the fewer leader bees.

[0082] In this embodiment of the application, the nectar sources in the artificial bee colony algorithm include 2 n Each corresponds to 2 in the embodiments of this application. n A dedicated intersection deployment scheme.

[0083] Step 301: Initialize the leader bee. In one embodiment, from 2 n In the dedicated intersection deployment scheme, each leader bee is randomly selected from an intersection deployment scheme, and the counter corresponding to the dedicated intersection deployment scheme of each leader bee is set to zero.

[0084] Step 302: According to the pre-set neighborhood search rules, based on the current dedicated intersection deployment scheme of each leader bee, use the honey-collecting process of the leader bee to perform a neighborhood search and determine the dedicated intersection deployment scheme corresponding to each leader bee.

[0085] In one exemplary instance, a neighborhood search is performed within a preset range around the current dedicated intersection deployment scheme of each leader bee, according to a pre-defined neighborhood search rule, to obtain a better dedicated intersection deployment scheme for each leader bee within a local range. That is, by using an existing dedicated intersection deployment scheme X... j Based on this, the configuration of one intersection in the dedicated intersection deployment scheme is changed (the selection of this intersection is random) to obtain a new dedicated intersection deployment scheme X within the preset range. j′ New Dedicated Cross-Intersection Deployment Solution X j′ This is referred to as the existing dedicated intersection deployment solution X. j A new dedicated intersection deployment scheme “around”.

[0086] In one embodiment, performing a neighborhood search within a preset range around the current dedicated intersection deployment scheme of each leading bee, according to a pre-set neighborhood search rule, may include:

[0087] For a given leader bee, a neighborhood search is performed within a preset range around the leader bee's current dedicated intersection deployment scheme, according to pre-set neighborhood search rules. This search begins with a randomly selected dedicated intersection deployment scheme X. j Dedicated intersection deployment solution X k Choose the better one as the new dedicated intersection deployment scheme.

[0088] Compare the newest dedicated intersection deployment solution X j′ The corresponding objective function value f Upper (X j′ ) and the old, current dedicated intersection deployment scheme X of the leading bee j The corresponding objective function value f Upper (X j If the new dedicated intersection scheme X j′ If the performance is better, then the new dedicated intersection solution X will be adopted. j′ Replacement of the old dedicated intersection solution X j And the new dedicated intersection scheme X j′ The corresponding counter is reset to zero; if the new dedicated cross-section scheme X j′ The performance is inferior to the old dedicated cross-section deployment solution X j For performance reasons, the old dedicated cross-section deployment scheme X is retained. j and replace the old dedicated intersection deployment scheme X j The corresponding counter is incremented by one.

[0089] Compared to the neighborhood search rule in the original artificial bee colony algorithm, the decision variable X in the dedicated intersection problem has different values ​​for x.i Since it is a discrete integer variable of 0-1, rather than a continuous variable, in this embodiment, the pre-set neighborhood search rule can be as shown in formula (2):

[0090] In formula (2), X represents the decision variable for solving the problem of dedicated intersections. x represents the i-th intersection. i Dedicated intersection deployment solution X j k∈{1,2,…,M} and i∈{1,2,…,n} represent randomly selected indices, and k≠j. Upper (X j ) indicates a dedicated intersection deployment scheme X j The objective function value.

[0091] The neighborhood search rule shown in formula (2) represents:

[0092] when At that time, a new solution The value is or That's all;

[0093] when At that time, a new solution The value of X is determined by j and X k The advantages and disadvantages of each option are determined by the probability of choosing a dedicated intersection deployment scheme with better performance. In this case, when dedicated intersection deployment scheme X... k Objective function value and dedicated intersection deployment scheme X j Objective function value and dedicated cross-section deployment scheme X k The ratio of the sum of the objective function values ​​is greater than a random number between 0 and 1, i.e. Indicates the dedicated intersection deployment scheme X j Compared to dedicated intersection deployment solution X k A better, new solution The value is a dedicated intersection deployment scheme. When dedicated intersection deployment scheme X k Objective function value and dedicated intersection deployment scheme X j Objective function value and dedicated cross-section deployment scheme X k The ratio of the sum of the objective function values ​​is less than or equal to a random number between 0 and 1, i.e. Indicates the dedicated intersection deployment scheme X k Compared to dedicated intersection deployment solution X j A better, new solution The value is a dedicated intersection deployment scheme.

[0094] In generating a new dedicated intersection deployment scheme X j′ Then, solve for its corresponding objective function value f. Upper (X j′ Compare the newest dedicated intersection deployment scheme X j′ The corresponding objective function value f Upper (X j′ ) and the old dedicated intersection deployment scheme X j The objective function value f corresponding to (i.e., the current dedicated intersection deployment scheme in step 101) is f. Upper (X j If the new dedicated intersection scheme X j′ If the performance is better, then the new dedicated intersection solution X will be adopted. j′ Replacement of the old dedicated intersection solution X j And the new dedicated intersection scheme X j′ The corresponding counter is reset to zero; if the new dedicated cross-section scheme X j′ The performance is inferior to the old dedicated cross-section deployment solution X j For performance reasons, the old dedicated cross-section deployment scheme X is retained. j and replace the old dedicated intersection deployment scheme X j The corresponding counter is incremented by one.

[0095] Step 303: According to the pre-set neighborhood search rules, based on the determined dedicated intersection deployment scheme corresponding to each leader bee, use the honey-collecting process of the follower bees to search for and determine the dedicated intersection deployment scheme corresponding to each leader bee.

[0096] In one exemplary instance, step 303 may include:

[0097] Based on a roulette wheel strategy, a leader bee is selected for each follower bee;

[0098] For each leader bee, each follower bee of the leader bee performs a neighborhood search around the leader bee’s current dedicated intersection deployment scheme within a preset range, according to a pre-set neighborhood search rule, so as to obtain the dedicated intersection deployment scheme of each follower bee in a local range.

[0099] Based on the objective function value corresponding to the dedicated intersection deployment scheme, the performance of each dedicated intersection deployment scheme generated by a follower bee is compared with that of the leader bee. If the performance of the dedicated intersection deployment scheme generated by the follower bee is better, then the dedicated intersection deployment scheme generated by the follower bee is used to replace the dedicated intersection deployment scheme corresponding to the leader bee, until all the follower bees of the leader bee have completed the comparison, and then the counter corresponding to the dedicated intersection deployment scheme of the leader bee is set to zero; if the performance of the dedicated intersection deployment schemes generated by all the follower bees is worse than that of the dedicated intersection deployment scheme of the leader bee, then the dedicated intersection deployment scheme of the leader bee is retained, and the counter corresponding to the dedicated intersection deployment scheme of the leader bee is incremented by one.

[0100] In one exemplary instance, roulette wheel selection is the simplest and most commonly used selection method, in which the selection probability of each individual is proportional to its fitness value; the higher the fitness, the greater the selection probability. In one embodiment, formula (3) can be used as the selection probability to select a leader bee for each follower bee.

[0101]

[0102] As shown in formula (3), M represents the size of the bee colony. Follower bees will choose p. j Larger leader bees will follow. It should be noted that each leader bee may have no followers, or it may be chosen by one or more followers.

[0103] The specific implementation of each follower bee in step 303 performing a neighborhood search for the lead bee within a preset range around its current dedicated intersection deployment scheme, according to a pre-set neighborhood search rule, is the same as performing a neighborhood search within a preset range around each lead bee's current dedicated intersection deployment scheme in step 302, and will not be repeated here.

[0104] Step 304: Increment the iteration neighborhood search count; determine whether the current iteration neighborhood search count has reached the iteration neighborhood search count threshold L2. If the current iteration neighborhood search count has reached the iteration neighborhood search count threshold L2, proceed to step 305; if the current iteration neighborhood search count has not reached the iteration neighborhood search count threshold L2, proceed to step 306.

[0105] Step 305: For each leader bee, the scout bees start from 2 n In the dedicated intersection deployment scheme, a dedicated intersection deployment scheme is randomly selected for the leading bee, and the counter corresponding to the dedicated intersection deployment scheme of the leading bee is set to zero.

[0106] In this step, if the counter value corresponding to the leading bee is zero, it means that the dedicated intersection deployment scheme of the leading bee has not been updated and is not considered the best dedicated intersection deployment scheme at present. It is necessary to search a different area for the leading bee.

[0107] Step 306: Compare the objective function values ​​corresponding to all the dedicated intersection deployment schemes of the leading bees, and record the dedicated intersection deployment scheme with the smallest objective function value (i.e., the best).

[0108] Steps 307-308: Increment the global iteration count by one, and determine whether the current global iteration count has reached the global iteration count threshold L1. If it has not reached the threshold, return to step 302; if it has reached the threshold, terminate the neighborhood search and use the dedicated intersections shown in the recorded dedicated intersection deployment scheme as the first intersections in the road network structure.

[0109] The intelligent algorithm based on artificial bee colony proposed in this application, based on traffic state information such as penetration rate and traffic demand, solves for a near-globally optimal dedicated intersection setting scheme within a limited number of iterations, thus achieving the selection of the best dedicated intersection setting scheme for the road network structure. Through this application, while improving the efficiency of connected autonomous vehicles, the negative impact of setting up dedicated intersections within the road network on human-driven vehicles is reduced, achieving global optimization of the traffic system.

[0110] Step 101: Based on the determined results, set the intersection in the road network structure that is the first intersection as the first intersection, and deploy the signalless cooperative driving algorithm at the first intersection.

[0111] In this embodiment of the application, the signalless cooperative driving algorithm deployed at the first intersection is not limited, and its specific implementation is not intended to limit the scope of protection of this application.

[0112] The intersection handling method for mixed traffic provided in this application determines the optimal deployment scheme for dedicated intersections based on road network structure and traffic state information (such as traffic demand and penetration rate). In this scheme, dedicated intersections in the road network structure allow connected autonomous vehicles to pass but not human-driven vehicles, while the remaining intersections in the road network structure allow both connected autonomous vehicles and human-driven vehicles to pass. This dedicated intersection deployment scheme reduces the travel time of connected autonomous vehicles while simultaneously considering the impact on human-driven vehicles, thus improving the efficiency of the entire mixed traffic system.

[0113] The intersection handling method for mixed traffic provided in this application is general for the management and control of mixed traffic systems. With the increasing penetration rate of autonomous vehicles, deploying dedicated autonomous driving intersections in urban road networks will realize the spatiotemporal separation of autonomous vehicles and human-driven vehicles, thereby avoiding disturbances from human-driven vehicles, effectively leveraging the advantages of autonomous vehicles, and improving traffic efficiency.

[0114] This application also provides a computer-readable storage medium storing computer-executable instructions for performing the intersection handling method for mixed traffic described in any of the preceding claims.

[0115] This application further provides an apparatus for implementing intersection processing for mixed traffic, including a memory and a processor, wherein the memory stores the following instructions executable by the processor for performing the steps of the intersection processing method for mixed traffic described in any of the preceding claims.

[0116] The dedicated intersection strategy for mixed traffic intersection handling proposed in this application significantly improves the traffic efficiency of the entire road network, especially the efficiency of autonomous vehicles. Figure 4 This is a schematic diagram of a typical road network structure used for simulation verification in the embodiments of this application. Figure 4 The intersection in the middle of the road network structure shown is set as a dedicated intersection. The average travel time of vehicles under two different methods, namely the conventional intersection strategy in related technologies and the dedicated intersection strategy provided in the embodiments of this application, is shown in Table 1.

[0117]

[0118]

[0119] Table 1

[0120] As shown in Table 1, by adopting the dedicated intersection strategy provided in the embodiments of this application, the dedicated intersection significantly reduces the average travel time of mixed traffic flow under any arrival rate, which means that the efficiency of the entire traffic system is improved.

[0121] Figure 5 This is a schematic diagram of the composition structure of an intersection processing device for mixed traffic in an embodiment of this application, as shown below. Figure 5 As shown, it may include: a deployment unit and a setting unit; wherein,

[0122] The deployment unit is used to determine the first intersection in the road network structure based on the road network structure and traffic status information; wherein the first intersection allows connected autonomous vehicles to pass but does not allow human-driven vehicles to pass, and the remaining intersections in the road network structure are the second intersections that allow both connected autonomous vehicles and human-driven vehicles to pass.

[0123] The setting unit is used to set the intersection that is the first intersection in the road network structure as the first intersection according to the determined result, and to deploy the signalless cooperative driving algorithm at the first intersection.

[0124] In one exemplary instance, the deployment unit may include: an initialization module, a first search module, a second search module, a first processing module, a recording module, and a second processing module; wherein,

[0125] The initialization module is used to initialize the parameters of the artificial bee colony algorithm; from 2 n In the dedicated intersection deployment scheme, each leader bee is randomly selected from an intersection deployment scheme, and the counter corresponding to the dedicated intersection deployment scheme of each leader bee is set to zero;

[0126] The first search module is used to perform a neighborhood search based on the current dedicated intersection deployment scheme of each leader bee and the honey-collecting process of the leader bee, according to the pre-set neighborhood search rules.

[0127] The second search module is used to search for and determine the dedicated intersection deployment scheme corresponding to each leader bee based on the determined dedicated intersection deployment scheme corresponding to each leader bee and the honey-collecting process of the follower bees, according to the pre-set neighborhood search rules.

[0128] The first processing module increments the iteration neighborhood search count by one; it then determines whether the current iteration neighborhood search count has reached the iteration neighborhood search count threshold L2. If the current iteration neighborhood search count has reached the iteration neighborhood search count threshold L2, for each leader bee, the scout bee starts from 2... n In the dedicated intersection deployment scheme, a dedicated intersection deployment scheme is randomly selected for the leading bee, and the counter corresponding to the dedicated intersection deployment scheme of the leading bee is set to zero, and the process enters the recording module; if the current iteration neighborhood search count has not reached the iteration neighborhood search count threshold L2, the process enters the recording module.

[0129] The recording module is used to compare the objective function values ​​corresponding to all the dedicated intersection deployment schemes of the leading bees and record the dedicated intersection deployment scheme with the smallest (i.e., the best) objective function value.

[0130] The second processing module increments the global iteration count by one and determines whether the current global iteration count has reached the global iteration count threshold L1. If it has not reached the threshold, the search in the first search module continues; if it has reached the threshold, the neighborhood search is terminated, and the dedicated intersections displayed in the dedicated intersection deployment scheme recorded in the recording module are taken as the first intersections in the road network structure.

[0131] In one exemplary instance, the first search module can be used to:

[0132] Within a preset range around the current dedicated intersection deployment scheme of each leader bee, a neighborhood search is performed according to a pre-set neighborhood search rule to obtain a better dedicated intersection deployment scheme for each leader bee in a local range.

[0133] In one exemplary instance, the first search module performs a neighborhood search within a preset range around the current dedicated intersection deployment scheme of each leading bee, according to pre-set neighborhood search rules, including:

[0134] For a given leader bee, a neighborhood search is performed within a preset range around the leader bee's current dedicated intersection deployment scheme, according to pre-set neighborhood search rules. This search begins with a randomly selected dedicated intersection deployment scheme X. k1 Dedicated intersection deployment solution X k2 Choose the better one as the new dedicated intersection deployment scheme.

[0135] Compare the newest dedicated intersection deployment solution X j′ The corresponding objective function value f Upper (X j′ ) and the old, current dedicated intersection deployment scheme X of the leading bee j The corresponding objective function value f Upper (X j If the new dedicated intersection scheme X j′ If the performance is better, then the new dedicated intersection solution X will be adopted. j′ Replacement of the old dedicated intersection solution X j And the new dedicated intersection scheme X j′ The corresponding counter is reset to zero; if the new dedicated cross-section scheme X j′ The performance is inferior to the old dedicated cross-section deployment solution X j For performance reasons, the old dedicated cross-section deployment scheme X is retained. j and replace the old dedicated intersection deployment scheme X j The corresponding counter is incremented by one.

[0136] In one exemplary instance, the second search module can be used to:

[0137] Based on a roulette wheel strategy, a leader bee is selected for each follower bee;

[0138] For each leader bee, each follower bee of the leader bee performs a neighborhood search around the leader bee’s current dedicated intersection deployment scheme within a preset range, according to a pre-set neighborhood search rule, so as to obtain the dedicated intersection deployment scheme of each follower bee in a local range.

[0139] Based on the objective function value corresponding to the dedicated intersection deployment scheme, the performance of each dedicated intersection deployment scheme generated by a follower bee is compared with that of the leader bee. If the performance of the dedicated intersection deployment scheme generated by the follower bee is better, then the dedicated intersection deployment scheme generated by the follower bee is used to replace the dedicated intersection deployment scheme corresponding to the leader bee, until all the follower bees of the leader bee have completed the comparison, and then the counter corresponding to the dedicated intersection deployment scheme of the leader bee is set to zero. If the performance of the dedicated intersection deployment schemes generated by all the follower bees is worse than that of the dedicated intersection deployment scheme of the leader bee, then the dedicated intersection deployment scheme of the leader bee is retained, and the counter corresponding to the dedicated intersection deployment scheme of the leader bee is incremented by one.

[0140] The intersection processing device for mixed traffic provided in this application determines the optimal deployment scheme for dedicated intersections based on road network structure and traffic state information (such as traffic demand, penetration rate, etc.), reducing the travel time of connected autonomous vehicles, while also taking into account the impact on human-driven vehicles, thereby improving the efficiency of the entire mixed traffic system.

[0141] Although the embodiments disclosed in this application are as described above, the content described is merely for the purpose of understanding this application and is not intended to limit this application. Any person skilled in the art to which this application pertains may make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in this application; however, the scope of patent protection of this application shall still be determined by the scope defined in the appended claims.

Claims

1. A method for handling intersections with mixed traffic, characterized in that, include: Based on the road network structure and traffic status information, the intersections in the road network structure are identified as the first intersections; wherein, the first intersections allow connected autonomous vehicles to pass but do not allow human-driven vehicles to pass, and the remaining intersections in the road network structure are the second intersections that allow both connected autonomous vehicles and human-driven vehicles to pass. Based on the determined results, the intersection in the road network structure that serves as the first intersection is designated as the first intersection, and a traffic light-free cooperative driving algorithm is deployed at the first intersection. The intersections that serve as the first intersections in the road network structure are determined based on the artificial bee colony algorithm, including: From 2 n In the dedicated intersection deployment scheme, each leader bee is randomly selected to choose an intersection deployment scheme, and the counter corresponding to the dedicated intersection deployment scheme of each leader bee is set to zero; where n is the number of intersections in the road network structure; According to the pre-set neighborhood search rules, based on the current dedicated intersection deployment scheme of each leader bee, the neighborhood search is performed using the honey-collecting process of the leader bees to determine the dedicated intersection deployment scheme corresponding to each leader bee. According to the pre-set neighborhood search rules, based on the dedicated intersection deployment scheme corresponding to each leader bee, the dedicated intersection deployment scheme corresponding to each leader bee is searched and determined by the honey-collecting process of the follower bees. Increment the iteration neighborhood search count by one; if the current iteration neighborhood search count reaches the iteration neighborhood search count threshold, for each leader bee, the scout bee increments the count by 2. n In the dedicated intersection deployment scheme, a dedicated intersection deployment scheme is randomly selected for the leading bee, and the counter corresponding to the dedicated intersection deployment scheme of the leading bee is set to zero, and the process proceeds to the next step; if the current iteration neighborhood search count has not reached the iteration neighborhood search count threshold, the process proceeds to the next step. Compare the objective function values ​​corresponding to all the dedicated intersection deployment schemes of the leading bees, and record the dedicated intersection deployment scheme with the smallest objective function value; Increment the global iteration count by one. If the current global iteration count has not reached the global iteration count threshold, return to the step of using the nectar-collecting process of the lead bee to perform neighborhood search and determine the dedicated intersection deployment scheme corresponding to each lead bee. If the current global iteration count has reached the global iteration count threshold, use the dedicated intersection shown by the recorded dedicated intersection deployment scheme as the intersection that is the first intersection in the road network structure. The method of utilizing the honey-collecting process of lead bees to perform neighborhood search and determine the dedicated intersection deployment scheme corresponding to each lead bee includes: Within a preset range of the current dedicated intersection deployment scheme for each leader bee, a neighborhood search is performed according to the pre-set neighborhood search rules to obtain a better dedicated intersection deployment scheme for each leader bee within a local range; wherein, the neighborhood search within the preset range of the current dedicated intersection deployment scheme for each leader bee, according to the pre-set neighborhood search rules, includes: For a given leader bee, a neighborhood search is performed within a preset range of the leader bee's current dedicated intersection deployment scheme according to the pre-set neighborhood search rules, starting from a randomly selected dedicated intersection deployment scheme X. j Dedicated intersection deployment solution X k Choose the better one as the new dedicated intersection deployment scheme. Compare the newest dedicated intersection deployment solution X j′ The corresponding objective function value f Upper (X j′ ) and the old, current dedicated intersection deployment scheme X of the leading bee j The corresponding objective function value f Upper (X j If the new dedicated intersection scheme X j′ If the performance is better, then the new dedicated intersection solution X will be adopted. j′ Replacement of the old dedicated intersection solution X j And the new dedicated intersection scheme X j′ The corresponding counter is reset to zero; if the new dedicated cross-section scheme X j′ The performance is inferior to the old dedicated cross-section deployment solution X j For performance reasons, the old dedicated cross-section deployment scheme X is retained. j and replace the old dedicated intersection deployment scheme X j The corresponding counter increments by one; Wherein, the objective function f Upper The expression in parentheses is as follows: Where, x i x ∈{0,1} i‘ ∈{0,1}, N is the set of intersections in the road network structure, x i x i‘ For each element in set N; the decision variable X is represented as X = {x1, x2, ..., x...} n }, where x i =1 indicates that intersection i is set as the first intersection, where i = 1, 2, ..., n, x i‘ =1 indicates that intersection i' is set as the first intersection, where i' = 1, 2, ..., n; x i =0 indicates that intersection i is set as the second intersection, i = 1, 2, ..., n, x i‘ =0 indicates that intersection i' is set as the second intersection, where i' = 1, 2, ..., n; v ii‘ Let A represent the flow on edge (i, i'), where edge (i, i') ∈ A, and A represents the set of edges connecting the intersections. t ii‘ , t′ ii‘ 、t″ ii‘ ,t″′ ii‘ These represent the travel time under four different intersection deployment schemes.

2. The intersection handling method according to claim 1, wherein, The traffic status information includes traffic demand and penetration rate.

3. The intersection processing method according to claim 1, wherein, The method of using the honey-foraging process of follower bees to search for and determine the dedicated intersection deployment scheme corresponding to each leader bee includes: Based on a roulette wheel strategy, a leader bee is selected for each follower bee; For each leader bee, each follower bee of the leader bee performs a neighborhood search on the leader bee within a preset range of the leader bee's current dedicated intersection deployment scheme, in order to obtain the dedicated intersection deployment scheme of each follower bee in a local range. Based on the objective function value corresponding to the dedicated intersection deployment scheme, the performance of each dedicated intersection deployment scheme generated by a follower bee is compared with that of the leader bee. If the performance of the dedicated intersection deployment scheme generated by the follower bee is better, then the dedicated intersection deployment scheme generated by the follower bee is used to replace the dedicated intersection deployment scheme corresponding to the leader bee, until all the follower bees of the leader bee have completed the comparison, and then the counter corresponding to the dedicated intersection deployment scheme of the leader bee is set to zero; if the performance of the dedicated intersection deployment schemes generated by all the follower bees is worse than that of the dedicated intersection deployment scheme of the leader bee, then the dedicated intersection deployment scheme of the leader bee is retained, and the counter corresponding to the dedicated intersection deployment scheme of the leader bee is incremented by one.

4. The intersection handling method according to claim 1 or 3, wherein, The pre-set neighborhood search rules are as follows: Where X represents the decision variable for solving the dedicated intersection problem. x represents the i-th intersection. i Dedicated intersection deployment solution X j k∈{1,2,…,M} and i∈{1,2,…,n} represent randomly selected subscripts, and k≠j, M represents the colony size in the artificial bee colony algorithm parameters; The deployment scheme X from randomly selected dedicated intersections j Dedicated intersection deployment solution X k Choose the better one as the new dedicated intersection deployment scheme. include: when At that time, the new dedicated intersection deployment scheme The value is or when When the dedicated intersection deployment scheme X k Objective function value and dedicated intersection deployment scheme X j Objective function value and dedicated cross-section deployment scheme X k The ratio of the sum of the objective function values ​​is greater than a random number between 0 and 1, in the new dedicated cross-section deployment scheme. The value is a dedicated intersection deployment scheme. When the dedicated intersection deployment scheme X k The objective function value and the dedicated intersection deployment scheme X j The objective function value and the dedicated intersection deployment scheme X k The ratio of the sum of the objective function values ​​is less than or equal to a random number between 0 and 1, in the new dedicated cross-section deployment scheme. The value is a dedicated intersection deployment scheme.

5. The intersection handling method according to claim 3, wherein, The roulette-based strategy for selecting a leader bee for each follower bee includes: The following formula is used as the selection probability to select p for each following bee. j Large lead bee: Where M represents the colony size in the artificial bee colony algorithm parameters.

6. A computer-readable storage medium storing computer-executable instructions for performing the intersection handling method for mixed traffic as described in any one of claims 1 to 5.

7. An apparatus for handling mixed traffic intersections, comprising a memory and a processor, wherein, The memory stores the following instructions that can be executed by the processor: steps for performing the intersection handling method for mixed traffic as described in any one of claims 1 to 5.

8. A cross-traffic processing device for mixed traffic, characterized in that, include: Deployment unit, configuration unit; among which, The deployment unit is used to determine the first intersection in the road network structure based on the road network structure and traffic status information; wherein the first intersection allows connected autonomous vehicles to pass but does not allow human-driven vehicles to pass, and the remaining intersections in the road network structure are the second intersections that allow both connected autonomous vehicles and human-driven vehicles to pass. The setting unit is used to set the intersection that is the first intersection in the road network structure as the first intersection according to the determined result, and to deploy the signalless cooperative driving algorithm at the first intersection. The deployment unit includes: an initialization module, a first search module, a second search module, a first processing module, a recording module, and a second processing module; wherein, The initialization module is used to initialize the parameters of the artificial bee colony algorithm; from 2 n In the dedicated intersection deployment scheme, each leader bee is randomly selected from an intersection deployment scheme, and the counter corresponding to the dedicated intersection deployment scheme of each leader bee is set to zero; The first search module is used to perform a neighborhood search based on the current dedicated intersection deployment scheme of each leader bee and the honey-collecting process of the leader bee, according to the pre-set neighborhood search rules. The second search module is used to search for and determine the dedicated intersection deployment scheme corresponding to each leader bee based on the determined dedicated intersection deployment scheme corresponding to each leader bee and the honey-collecting process of the follower bees, according to the pre-set neighborhood search rules. The first processing module increments the iteration neighborhood search count by one; it then determines whether the current iteration neighborhood search count has reached the iteration neighborhood search count threshold. If the current iteration neighborhood search count has reached the threshold, for each leader bee, the reconnaissance bee increments the count from 2... n In the dedicated intersection deployment scheme, a dedicated intersection deployment scheme is randomly selected for the leading bee, and the counter corresponding to the dedicated intersection deployment scheme of the leading bee is set to zero, and the process enters the recording module; if the current iteration neighborhood search count has not reached the iteration neighborhood search count threshold, the process enters the recording module; where n is the number of intersections in the road network structure; The recording module is used to compare the objective function values ​​corresponding to all the dedicated intersection deployment schemes of the leading bees, and record the dedicated intersection deployment scheme with the smallest objective function value; The second processing module increments the global iteration count by one and determines whether the current global iteration count has reached the global iteration count threshold. If it has not reached the threshold, the search in the first search module continues. If it has reached the threshold, the neighborhood search is terminated, and the dedicated intersections displayed in the dedicated intersection deployment scheme recorded in the recording module are taken as the first intersections in the road network structure. The second search module utilizes the nectar-collecting process of lead bees to perform neighborhood searches and determine the dedicated intersection deployment scheme corresponding to each lead bee, including: Within a preset range of the current dedicated intersection deployment scheme for each leader bee, a neighborhood search is performed according to the pre-set neighborhood search rules to obtain a better dedicated intersection deployment scheme for each leader bee within a local range; wherein, the neighborhood search within the preset range of the current dedicated intersection deployment scheme for each leader bee, according to the pre-set neighborhood search rules, includes: For a given leader bee, a neighborhood search is performed within a preset range of the leader bee's current dedicated intersection deployment scheme according to the pre-set neighborhood search rules, starting from a randomly selected dedicated intersection deployment scheme X. j Dedicated intersection deployment solution X k Choose the better one as the new dedicated intersection deployment scheme. Compare the newest dedicated intersection deployment solution X j′ The corresponding objective function value f Upper (X j′ ) and the old, current dedicated intersection deployment scheme X of the leading bee j The corresponding objective function value f Upper (X j If the new dedicated intersection scheme X j′ If the performance is better, then the new dedicated intersection solution X will be adopted. j′ Replacement of the old dedicated intersection solution X j And the new dedicated intersection scheme X j′ The corresponding counter is reset to zero; if the new dedicated cross-section scheme X j′ The performance is inferior to the old dedicated cross-section deployment solution X j For performance reasons, the old dedicated cross-section deployment scheme X is retained. j and replace the old dedicated intersection deployment scheme X j The corresponding counter increments by one; Wherein, the objective function f Upper The expression in parentheses is as follows: Where, x i x ∈{0,1} i‘ ∈{0,1}, N is the set of intersections in the road network structure, x i x i‘ For each element in set N; the decision variable X is represented as X = {x1, x2, ..., x...} n }, where x i =1 indicates that intersection i is set as the first intersection, i = 1, 2, ..., n, x i‘ =1 indicates that intersection i' is set as the first intersection, where i' = 1, 2, ..., n; x i =0 indicates that intersection i is set as the second intersection, i = 1, 2, ..., n, x i‘ =0 indicates that intersection i' is set as the second intersection, where i' = 1, 2, ..., n; v ii‘ Let A represent the flow on edge (i, i'), where edge (i, i') ∈ A, and A represents the set of edges connecting the intersections. t ii‘ , t′ ii‘ 、t″ ii‘ ,t″′ ii‘ These represent the travel time under four different intersection deployment schemes.

Images