Intersection group collaborative optimization method and device in man-machine mixed driving scene, and storage medium

Abstract

The invention relates to an intersection group collaborative optimization method and device in a man-machine mixed driving scene, and a storage medium. The method comprises the following steps: dividing a to-be-optimized urban area into a plurality of traffic zones; predicting the ownership of automatic driving vehicles and human driving vehicles in the traffic zone according to the vehicle type purchase intention data and a pre-constructed multiple linear regression model; according to a binomial Logit model, calculating a human driving travel ratio and an automatic driving travel ratio of the traffic zone, according to the travel ratio and the inventory, calculating the travel occurrence amount of human driving vehicles and automatic driving vehicles of the traffic zone, and obtaining a preset attraction amount of the traffic zone; calculating traffic demands of human-driven vehicles and automatic-driven vehicles between the starting point and the ending point of the traffic zone according to the travel occurrence amount and the attraction amount; constructing a flow distribution model according to the road network of the traffic zone; and solving the intersection group collaborative optimization model according to the flow distribution model to obtain a final signal optimization scheme.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of traffic signal control, and in particular to a method, device and storage medium for intersection group cooperative optimization in a human-machine mixed driving scenario. BACKGROUND

[0002] With the rapid development of automatic driving and intelligent networking technology, there will be more and more urban roads with mixed traffic flow scenarios of intelligent networking, that is, the state of mixed driving of human-driven vehicles and intelligent automatic driving vehicles. In order to ensure the traffic efficiency of the road, an intersection optimization design method based on time elements, space elements and space-time joint optimization is needed to provide a theoretical basis and optimization strategy for intersection management in the mixed traffic environment of intelligent networking.

[0003] At present, some vehicle networking pilot areas have built a wide range of intelligent networking roadside facilities and a batch of automatic driving and intelligent networking application scenarios, realizing relatively accurate traffic perception and control. Based on this, if further optimization design of the signal control intersection itself is made according to the traffic mechanism and characteristics of intelligent networking mixed traffic in the next step of construction, it will have better forward-looking value and application effect. Therefore, in view of different penetration rate scenarios of future automatic driving vehicles, effective reduction of vehicle delay at intersections and other situations, there is an urgent need for a signal control optimization method in a human-machine mixed driving scenario to improve the traffic efficiency of traffic flow. SUMMARY

[0004] The embodiments of the present application provide a method, device and storage medium for intersection group cooperative optimization in a human-machine mixed driving scenario, to at least solve the problem of low road traffic efficiency in the related art in a human-machine mixed driving scenario.

[0005] In a first aspect, the embodiments of the present application provide a method for intersection group cooperative optimization in a human-machine mixed driving scenario, comprising: obtaining regional comprehensive information of a to-be-optimized urban area, and dividing the to-be-optimized urban area into a plurality of traffic zones; obtaining vehicle type purchase willingness data of the traffic zones, and predicting the number of automatic driving vehicles and human-driven vehicles in the traffic zones according to the vehicle type purchase willingness data and a pre-constructed multiple linear regression model; calculating the human-driven travel ratio and the automatic driving travel ratio of the traffic zones respectively according to a pre-constructed binary Logit model, calculating the number of human-driven vehicle trips and automatic driving vehicle trips in the traffic zones according to the travel ratio and the number of the corresponding vehicle types, and obtaining a preset attraction amount of the traffic zones; calculating the traffic demand of human-driven vehicles and automatic driving vehicles between the start and end points of the traffic zones according to the number of human-driven vehicle trips and automatic driving vehicle trips and the attraction amount; According to the regional comprehensive information, a road network of the traffic zone is obtained, and a traffic distribution model is constructed according to the road network, the traffic distribution model being used for distributing traffic demands between all origins and destinations in the traffic zone to the road network; According to the traffic distribution model, a coordinated optimization model of the intersection group is constructed, and the coordinated optimization model of the intersection group is solved to obtain a final signal optimization scheme.

[0006] In an embodiment, the vehicle type purchase willingness data includes age of the population in the traffic zone, type of car ownership, understanding degree of new technology, and willingness value of purchasing the next car; and the method further includes: The age of the population in the traffic zone, the type of car ownership, the understanding degree of new technology, and the willingness value of purchasing the next car are input into an automatic driving vehicle purchase willingness calculation formula in the multiple linear regression model to obtain the automatic driving vehicle ownership; The age of the population in the traffic zone, the type of car ownership, the understanding degree of new technology, and the willingness value of purchasing the next car are input into a human driving vehicle purchase willingness calculation formula in the multiple linear regression model to obtain the human driving vehicle ownership.

[0007] In an embodiment, the method further includes: according to the pre-constructed binomial Logit model, calculating a human driving travel ratio and an automatic driving travel ratio of the traffic zone respectively, and calculating a human driving vehicle and an automatic driving vehicle travel occurrence amount of the traffic zone according to the travel ratio and the ownership of the corresponding vehicle type, and setting an attraction amount of the traffic zone to be equal to the travel occurrence amount. The personnel information of the population in the traffic zone is obtained, and the personnel information is input into the binomial Logit model to obtain the human driving travel ratio and the automatic driving travel ratio of the traffic zone; The human driving travel ratio is multiplied by the human driving vehicle ownership to calculate the human driving vehicle travel occurrence amount, and the human driving vehicle attraction amount of the traffic zone is set to be equal to the human driving vehicle travel occurrence amount; The automatic driving travel ratio is multiplied by the automatic driving vehicle ownership to calculate the automatic driving vehicle travel occurrence amount, and the automatic driving vehicle attraction amount of the traffic zone is set to be equal to the automatic driving vehicle travel occurrence amount.

[0008] In an embodiment, the traffic distribution model constructed according to the road network includes: obtain a road network of the traffic zone from the regional comprehensive information, the road network comprising road segments and intersections; construct a travel cost function, and calculate a comprehensive travel cost of a vehicle path according to the travel cost function; wherein the travel cost function comprises a BPR function for estimating road segment travel time and a Webster formula for estimating intersection travel time; combine the comprehensive travel cost of the vehicle path, and distribute traffic demand of the human-driven vehicles and the autonomous vehicles by using an improved user equilibrium model to obtain the traffic distribution model.

[0009] In an embodiment, the construction of the intersection group cooperative optimization model comprises: obtain a set of vehicle traffic conditions and signal control parameters of each road segment and intersection in the traffic zone; construct a calculation formula of total travel time of a path according to the set of vehicle traffic conditions and signal control parameters; combine the calculation formula of total travel time of the path and the traffic distribution model to obtain the intersection group cooperative optimization model.

[0010] In an embodiment, the intersection group cooperative optimization model is solved according to the traffic distribution model to obtain a final signal optimization scheme, comprising: determine an initial signal optimization scheme according to the current traffic condition of the intersection, the initial signal optimization scheme comprising signal cycle length and green time of the intersection; solve the traffic distribution model according to the initial signal optimization scheme to obtain a traffic distribution result, the traffic distribution result comprising road segment traffic of human driving and autonomous driving, and flow direction at each intersection; calculate total travel time of a path according to the traffic distribution result and the current signal optimization scheme; determine whether the change of the total travel time of the path is within a preset tolerance range, if the determination result is yes, it is considered that the optimization has converged, and the current signal optimization scheme is output as the final optimization scheme; otherwise, the signal cycle length and the green time of the intersection are adjusted based on selection, crossover and mutation operations of a genetic algorithm, and the current signal optimization scheme is updated for continuous iteration.

[0011] In an embodiment, the multiple linear regression model is:

[0012] wherein, is age, is commuting time, is family population, is annual family income, a predicted human-driven vehicle fleet size, a predicted autonomous vehicle fleet size; , , , , regression coefficients for predicting the human-driven vehicle fleet size, , , , , regression coefficients for predicting the autonomous vehicle fleet size.

[0013] In an embodiment, the binomial Logit model is:

[0014] wherein, is the human-driven trip rate in a traffic zone, is the autonomous-driven trip rate in a traffic zone, is the observed utility of choosing to travel using a human-driven vehicle that one owns, is the observed utility of choosing to travel using an autonomous vehicle that one owns, and similarly, is the observed utility of choosing to travel using public transportation, is age, is gender, is trip length, , , , , , , , , , , , are regression coefficients.

[0015] In an embodiment, the flow allocation model is:

[0016] wherein OBJ is the integral of the cumulative travel time of traffic flow on a link, denotes the value of OBJ that minimizes , , are the human-driven and autonomous-driven traffic demands between r and s for any origin-destination pair, respectively, is the human-driven traffic flow on link i, is the autonomous-driven traffic flow on link i, , respectively represent human-driven and autonomous traffic flow on a link a; represents whether the ith path in the origin-destination pair r-s passes through the approach of the intersection numbered b, a represents a link number, A represents a link set, l represents an intersection approach number, and L represents an intersection approach set.

[0017] In an embodiment, the intersection group collaborative optimization model is:

[0018] wherein TTT represents a total travel time of a path, is a free-flow travel time of a link ; is a traffic flow of a link ; is a capacity of a link ; is an average queue length of vehicles at an approach of an intersection ; is a traffic volume of an approach of an intersection ; is an average speed of vehicles at an approach of an intersection ; is a saturation flow of an approach of an intersection ; is a current signal optimization scheme of the intersection b, including a signal cycle length of each intersection and a green time of each phase.

[0019] In a second aspect, an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the intersection group collaborative optimization method in the human-machine mixed driving scenario when executing the computer program.

[0020] In a third aspect, an embodiment of the present application provides a computer readable storage medium, which stores a computer program executable by a processor, and the processor implements the intersection group collaborative optimization method in the human-machine mixed driving scenario when executing the computer program.

[0021] The intersection group collaborative optimization method, device and storage medium in the human-machine mixed driving scenario provided by the embodiments of the present application at least have the following technical effects: The application extends the research perspective from a single intersection to the entire road network system, and performs spatio-temporal coordinated optimization design on intersection groups in the region, so that the traffic flow and intersection signal control strategy are adapted and coordinated with each other, thereby achieving the optimal traffic system operation state.

[0022] Details of one or more embodiments of the application are set forth in the accompanying drawings and description below to make the other features, objects, and advantages of the application more apparent. BRIEF DESCRIPTION OF DRAWINGS

[0023] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings: Figure 1 is a flow chart of the intersection group coordinated optimization method in the human-machine mixed driving scenario in an embodiment of the application; Figure 2 is an example diagram of traffic cell division in an embodiment of the application; Figure 3 is an example diagram of the traffic cell flow distribution model in the application; Figure 2 Figure 4 is a flow chart of solving the intersection group coordinated optimization model in the intersection group coordinated optimization method in the human-machine mixed driving scenario of the application; Figure 5 is a structural block diagram of an electronic device in an embodiment of the application. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical scheme and advantages of the application more clear, the application is described and explained below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the application and do not limit the application. Based on the embodiments provided in the application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of the application.

[0025] Obviously, the drawings in the following description are only some examples or embodiments of the application, and for those of ordinary skill in the art, the application can be applied to other similar scenarios without creative labor on the basis of these drawings. In addition, it can be understood that although the efforts made in this development process can be complex and lengthy, for those of ordinary skill in the art related to the content disclosed in the application, some design, manufacture or production changes based on the technical content disclosed in the application are only routine technical means, and should not be understood as insufficient disclosure of the content disclosed in the application.

[0026] ​Reference to an "embodiment" in this application means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily mutually exclusive of other embodiments. It is expressly understood that the embodiments described in this application are open to be combined with each other in their various permutations and combinations.

[0027] Unless otherwise defined, technical terms and scientific terms used in the present application shall have the ordinary meanings as understood by one of ordinary skill in the art to which this application pertains. The terms "a", "an", "one", "this", and the like, as used in the present application, do not denote a limitation of quantity and can mean one or more. The terms "comprising", "containing", "having", and "including" and their variations, as used in the present application, are intended to cover the respective recited elements or steps in their open-ended and close-ended senses, respectively. For example, a process, method, system, product, or apparatus that comprises a list of steps or elements is not necessarily limited to only those steps or elements, but can include additional steps or elements not expressly listed or inherent to such process, method, system, product, or apparatus. The terms "connected", "coupled", and "linked", as used in the present application, are not limited to direct connections, but can include indirect connections unless otherwise specifically stated. The term "plurality" means two or more. The term "and / or" describes associated objects in association relationships, which means that there can be three relationships, for example, "A and / or B" can mean that A exists alone, A and B exist together, and B exists alone. The character " / " generally means that the associated objects are in an "or" relationship. The terms "first", "second", "third", and the like, merely distinguish similar objects, and do not represent a specific order.

[0028] To illustrate the intersection group cooperative optimization method in the mixed driving scenario of man and machine in this application in detail, this embodiment selects an intelligent network connection core area in a region, and based on the cloud control platform and field roadside facilities, uses the intersection group cooperative optimization method in the mixed driving scenario of man and machine in this application to analyze and research the adaptability of the optimization design of the key signal control intersection, to verify the effectiveness of the proposed time-space cooperative control method of intersection network connection mixed traffic, and on the basis of the signal control of the existing intersection, the time-space optimization design of the intersection signal of the whole area is carried out, and the specific implementation process is as shown in Figure 1 .

[0029] S1, obtaining the regional comprehensive information of a to-be-optimized urban area, and dividing the to-be-optimized urban area into a plurality of traffic zones.

[0030] Specifically, the comprehensive regional information in this embodiment includes land feature data, population feature data, and vehicle feature data. Land feature data includes road network data and land use data (land use type, administrative division, land use area, etc.). Population feature data includes age, gender, economic status, travel purpose, car purchase demand factors (efficiency, convenience, economy, etc.), and vehicle purchase intention data. Vehicle feature data includes vehicle ownership information (car ownership type, number of electric bicycles, etc.). A preset division rule is obtained, and based on the road conditions in the land feature data and the preset division rule, the urban area to be optimized is divided into multiple traffic zones. In this embodiment, traffic zones can be divided based on land use, administrative division, or natural terrain. Here, it specifically refers to dividing zones using land use type. Considering that the method for dividing traffic zones is relatively mature, it will not be described in detail here. The specific division rule can be set according to actual needs.

[0031] In this embodiment, demographic and vehicle characteristic data can be obtained through questionnaires. For example, questionnaires can be designed and distributed geographically, covering questions such as age, vehicle ownership, understanding of new technologies, economic status, and willingness to purchase human-driven and autonomous vehicles. The collected questionnaire data is then cleaned and organized, handling missing and outlier values. Based on the questionnaire content, relevant factors (such as age, efficiency, convenience, and economy) are defined as independent variables, and the number of human-driven and autonomous vehicles are defined as dependent variables. A multiple linear regression method is then used, with the preprocessed data input, to calculate regression coefficients to fit the model and determine the influence of each independent variable on the dependent variable. Finally, statistical tests are used to evaluate the goodness of fit and significance of the regression model, ensuring its reliability and predictive ability.

[0032] In a preferred embodiment, this embodiment refers to the result of dividing the area into traffic zones. Figure 2 , Figure 2 The numbers in the text are the traffic zone numbers defined in this embodiment, that is, in this embodiment, the traffic zones are... Figure 2 The area is divided into 55 traffic zones.

[0033] S2. Obtain vehicle purchase intention data for the traffic community, and predict the number of autonomous vehicles and human-driven vehicles in the traffic community based on the vehicle purchase intention data and a pre-built multiple linear regression model.

[0034] Specifically, the vehicle model purchase willingness data including the age of the population in the traffic zone, the type of owning a car, the understanding degree of new technology, and the willingness value of purchasing the next car are obtained separately or from the regional comprehensive information. Then, the age of the population in the traffic zone, the type of owning a car, the understanding degree of new technology, and the willingness value of purchasing the next car are input into the automatic driving vehicle purchase willingness calculation formula in the multiple linear regression model to obtain the automatic driving vehicle ownership; the age of the population in the traffic zone, the type of owning a car, the understanding degree of new technology, and the willingness value of purchasing the next car are input into the human driving vehicle purchase willingness calculation formula in the multiple linear regression model to obtain the human driving vehicle ownership.

[0035] Specifically, the embodiment adopts a multiple linear regression model to predict the human driving vehicle and automatic driving vehicle ownership of each zone, and the multiple linear regression model includes a human driving vehicle purchase willingness calculation formula and an automatic driving vehicle purchase willingness calculation formula, and the mathematical expressions are respectively:

[0036] wherein, is the age, is the commuting time, is the family population, is the family annual income, is the predicted human driving vehicle ownership, is the predicted automatic driving vehicle ownership. is the regression coefficient of the predicted human driving vehicle ownership, is the regression coefficient of the predicted automatic driving vehicle ownership.

[0037] In the embodiment, the coefficient values of the above formula are obtained by regression analysis fitting after obtaining a large amount of data, such as using existing multiple linear regression analysis software, using least square method, etc. Based on this, the fitting process is not further described herein. In the data acquisition process, the efficiency, convenience, economy, and purchase willingness of the two types of cars can be quantitatively processed. In the actual operation process, the method of the application can score according to 0-5 points for data that cannot be quantified or quantified with great difficulty. For data that can be quantified or has a large numerical value, the minimum and maximum values can be mapped to the corresponding 0-5 point interval in the processing process, thereby realizing the unity of units.​​​​​​​​

[0038] In a specific embodiment, the prediction models of the number of human-driven vehicles and the number of autonomous vehicles are obtained by questionnaire survey and regression analysis as follows:

[0039] S3, according to the pre-constructed binomial Logit model, the human-driven travel ratio and the autonomous travel ratio of the traffic zone are calculated respectively, the travel occurrence of the human-driven vehicles and the autonomous vehicles in the traffic zone is calculated according to the travel ratio and the number of vehicles of the corresponding vehicle type, and the attraction of the traffic zone is obtained, in this embodiment, the attraction of the traffic zone can be set to be equal to the travel occurrence.

[0040] In a preferred embodiment, the calculation of travel demand is realized according to the following steps: S301, the personnel information of the crowd in the traffic zone is obtained, and the personnel information is input into the binomial Logit model to obtain the human-driven travel ratio and the autonomous travel ratio of the traffic zone. Specifically, the internal resident travel is studied, the population characteristic data and the land characteristic data are obtained from the regional comprehensive information, the population characteristic data of the traffic zone is analyzed, and the resident travel is divided into four travel types according to different travel destinations, i.e. work travel from home, school travel from home, flexible travel from home and travel from non-home; S302, the human-driven travel ratio is multiplied by the number of human-driven vehicles to calculate the travel occurrence of the human-driven vehicles, and the attraction of the human-driven vehicles in the traffic zone is set to be equal to the travel occurrence of the human-driven vehicles; S303, the autonomous travel ratio is multiplied by the number of autonomous vehicles to calculate the travel occurrence of the autonomous vehicles, and the attraction of the autonomous vehicles in the traffic zone is set to be equal to the travel occurrence of the autonomous vehicles.

[0041] The binomial Logit model includes an autonomous travel utility calculation function and a human-driven travel utility calculation function. In this embodiment, when the travel ratio is calculated, the age, the type of owning a car, the understanding degree of new technology and the willingness value of purchasing the next car of the crowd in the traffic zone are input into the autonomous travel utility calculation function to obtain the autonomous travel utility, the age, the type of owning a car, the understanding degree of new technology and the willingness value of purchasing the next car of the crowd in the traffic zone are input into the human-driven travel utility calculation function to obtain the human-driven travel utility, and the travel ratio is calculated according to the autonomous travel utility and the human-driven travel utility.

[0042] Preferably, the present embodiment assumes that the user can only choose to travel by human-driven vehicles and autonomous vehicles, assuming that the choice of autonomous vehicle travel is alternative i, and the choice of human-driven vehicle travel is alternative j. The present application can predict the travel ratio P of autonomous vehicles and human-driven vehicles in each cell according to the binomial Logit model of travel choice intention n (i), and the specific mathematical expression is:

[0043] wherein, is the ratio of human-driven vehicle owners in the cell who choose to travel by their human-driven vehicles, is the ratio of autonomous vehicle owners in the cell who choose to travel by their autonomous vehicles, is the observed utility of choosing to travel by the human-driven vehicle owned by the user, is the observed utility of choosing to travel by the autonomous vehicle owned by the user, and similarly, is the observed utility of traveling by public transportation, is the age, is the gender, is the travel duration, , , , , , , , , , , , are regression coefficients. In the present embodiment, the calculation formula of the observed utility of travel can be obtained by referring to the questionnaire provided in step S1 and combining the regression analysis model fitting. That is, the numerical values of the above formula are also obtained by fitting the existing regression analysis algorithm or software after obtaining a large amount of data, and the above data are quantified in the form of scoring during the data acquisition process.

[0044] In a specific embodiment, the utility formula of human-driven vehicle owners choosing their own small cars, autonomous vehicle owners choosing their own small cars, and choosing to travel by bus obtained by questionnaire survey and regression analysis is as follows:

[0045] S4, according to the trip generation and attraction of the human-driven vehicles and the autonomous vehicles in the region to be optimized, calculate the traffic demand between the origin-destination (OD) of the traffic zone for the human-driven vehicles and the autonomous vehicles, and obtain the OD matrix of the human-driven trips and the autonomous trips.

[0046] After obtaining the trip demand, according to the trip demand and the traffic distribution prediction model of the region to be optimized, the traffic demand between the OD of the traffic zone for the human-driven vehicles and the autonomous vehicles can be predicted, that is, two OD matrices of the human-driven trips and the autonomous trips are obtained. In this embodiment, in addition to the gravity model, other models such as the growth coefficient, the intervention opportunity model, and the maximum entropy model are also applicable.

[0047] S5, obtain the road network of the traffic zone according to the regional comprehensive information, and construct a traffic assignment model according to the road network, wherein the traffic assignment model is used to distribute the traffic demand between all the origin-destinations in the traffic zone to the road network.

[0048] Specifically, the implementation process of the traffic assignment model is as follows: S501, obtain the road network of the traffic zone according to the regional comprehensive information, wherein the road network includes road segments and intersections; S502, construct a trip cost function, and calculate the comprehensive trip cost of the vehicle path according to the trip cost function; In this embodiment, in order to consider the influence of the optimization of the intersection on the road network traffic distribution, the trip cost function defined contains both the trip cost of the road segment and the trip cost of the intersection; wherein the trip cost function includes the BPR function for estimating the road segment travel time and the Webster formula for estimating the intersection travel time; S503, combine the comprehensive trip cost of the vehicle path, and distribute the traffic demand of the human-driven vehicles and the autonomous vehicles by using the improved user equilibrium model, and obtain the traffic assignment model.

[0049] In this embodiment, the expression of the BPR function is as follows:

[0050] The expression of the Webster formula is as follows:

[0051] wherein, is the free flow travel time of the road segment is the free flow travel time of the road segment is the free flow travel time of the road segment traffic flow of the road segment; the capacity of the road segment; the capacity of the road segment; the average queue length of vehicles at the intersection; the average queue length of vehicles at the intersection; the traffic flow of the approach of the intersection; the traffic flow of the approach of the intersection; the traffic flow of the approach of the intersection; the average speed of vehicles at the intersection; the average speed of vehicles at the intersection; the average speed of vehicles at the intersection; the saturation flow of the approach of the intersection. the saturation flow of the approach of the intersection. the saturation flow of the approach of the intersection. the saturation flow of the approach of the intersection.

[0052] The traffic flow is allocated by improving the user equilibrium model, and the traffic flow allocation model is obtained. Specifically, the traditional user equilibrium model (UE) is improved to allocate traffic flow, and the specific mathematical expression is as follows:

[0053] wherein, OBJ is the integral of the cumulative travel time of traffic flow on the road segment, is the value when OBJ is minimum, , , are the travel demands of any OD pair in the r-s interval for human driving and autonomous driving, is the human driving traffic flow of path i, is the autonomous driving traffic flow of path i, , are the human driving and autonomous driving traffic flows of the a road segment; represents whether the ith path in the OD pair r-s passes through the approach of the intersection numbered b, and is 1 if yes, and 0 if no, a is the road segment number, A is the road segment set, l (letter) is the intersection approach number, and L is the intersection approach set. R and S represent the origin set and the destination set in the OD pair, that is, R is the set of all departure locations, S is the set of all destination locations, r and s represent the specific departure location and the specific destination (i.e., the origin and the destination) in the OD pair, that is, r is a specific departure location, and s is a specific destination location. The traffic flow allocation prediction result is shown in Figure 3 , Figure 3 The color change of the road segment in

[0054] S6, acquire a pre-constructed and trained intersection group collaborative optimization model, solve the intersection group collaborative optimization model according to the flow distribution model, and obtain a final signal optimization scheme.

[0055] Specifically, the construction of the intersection group collaborative optimization model comprises: acquiring a vehicle traffic condition and signal control parameter set of each road section and intersection in the traffic zone; constructing a calculation formula of path total travel time according to the vehicle traffic condition and signal control parameter set; combining the calculation formula of path total travel time and the flow distribution model to obtain the intersection group collaborative optimization model.

[0056] In the embodiment, an intersection group collaborative optimization model is established to collaboratively optimize traffic flow distribution and intersection signal control, and the specific mathematical expression is as follows:

[0057] wherein, TTT represents path total travel time of a vehicle in a road network, is a target function value of the collaborative optimization model constructed in the application, is a free flow travel time of a road section is a flow of a road section is a traffic capacity of a road section is an average queue length of a vehicle at an approach of an intersection is a flow of a vehicle at an approach of an intersection is an average speed of a vehicle at an intersection is a saturation flow of a vehicle at an approach of an intersection is a current signal optimization scheme of an intersection b, comprising a signal cycle length of each intersection and a green time of each phase.

[0058] ​​​​​​​​​​​In the solving process, an initial signal optimization scheme is determined according to the current traffic condition of the intersection, the initial signal optimization scheme including signal cycle length and green time of the intersection; the traffic distribution model is solved according to the initial signal optimization scheme, to obtain a traffic distribution result, the traffic distribution result including road segment traffic of human driving and automatic driving and flow direction at each intersection; the path total travel time is calculated according to the traffic distribution result and the current signal optimization scheme; it is judged whether the change of the path total travel time is within a preset tolerance range, if the result of the judgment is yes, it is considered that the optimization has converged, and the current signal optimization scheme is output as a final optimization scheme; otherwise, the signal cycle length and the green time of the intersection are adjusted based on selection, crossover and mutation operations of the genetic algorithm, and the current signal optimization scheme is updated to continue iteration.

[0059] In a preferred embodiment, the iterative solving process of the intersection group cooperative optimization model is as shown in Figure 4 , which includes the following steps: S601, initialization scheme: according to the current situation of the intersection, an initial optimization scheme is set, including signal cycle length of each intersection and green time of each phase; S602, traffic distribution model solving: using the initial signal optimization scheme, the traffic distribution model is solved to obtain road segment traffic of human driving and intelligent networked traffic flow and flow direction at each intersection; S603, intersection optimization: according to the result of the traffic distribution model and the current signal optimization scheme, including signal cycle length of each intersection and green time of each phase, then the signal control parameters of each intersection are calculated, such as queue length , based on selection, crossover and mutation operations of the genetic algorithm, the signal control parameters of the intersection are adjusted according to the specific optimization method. Among them, the signal cycle length of each intersection and the green time of each phase are included.

[0060] S604, double-layer optimization iteration: the upper layer adjusts the traffic flow distribution, and the road segment traffic is recalculated to ensure reasonable distribution of road network traffic, and the lower layer optimizes the signal control strategy and signal control parameters of each intersection based on the distribution result of the upper layer, to calculate new ​​​​​​​​S605, convergence determination: after each iteration, check the change of the optimization target (total travel time TTT), if the change of the objective function value is within the preset tolerance range, it is considered that the optimization has converged, and the final optimization scheme and result are output, otherwise, continue iteration; S606, output optimization result: after the optimization converges, output the final optimization scheme of each intersection , queue length , flow , average speed and the travel time of the overall traffic system .

[0061] The embodiment selects a suitable model for optimization according to the characteristics of the intersection, compares the average delay of vehicles before and after optimization, and the results are shown in Tables 1-3: Table 1 optimization result when predicting CAV penetration rate of 12.9% in 2025

[0062] Table 2 optimization result when predicting CAV penetration rate of 39.1% in 2030

[0063] Table 3 optimization result when predicting CAV penetration rate of 60.1% in 2035

[0064] In the embodiment, the CAV penetration rate can be estimated by the number of autonomous and human-driven vehicles, and according to the changes in population, technology and purchase willingness, the number of vehicles in different years can be estimated, and based on this, the CAV penetration rate can be further estimated, and further, the road traffic distribution results and the corresponding optimization scheme in different years can be predicted according to the above method, therefore, the intersection group cooperative optimization method provided by the application can predict the road traffic conditions in the same region in different years and provide the corresponding optimization scheme.

[0065] To sum up, the intersection group cooperative optimization method in the mixed driving scenario provided by the embodiment of the application extends the research perspective from a single intersection to the entire road network system, and performs spatio-temporal cooperative optimization design on the intersection group in the region, so that the traffic flow and the intersection signal control strategy adapt to and coordinate with each other, thereby achieving the optimal traffic system operation state.

[0066] In a second aspect, an electronic device is provided according to an example embodiment, Figure 5 is a block diagram of an electronic device according to an example embodiment. As shown in Figure 5As shown, the electronic device can include a processor 11 and a memory 12 having stored computer program instructions.

[0067] In particular, the processor 11 described above can include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement one or more embodiments of the application.

[0068] The memory 12 can include a mass storage for data or instructions. By way of example and not limitation, the memory 12 can include a Hard Disk Drive (HDD), floppy disk drive, a Solid State Drive (SSD), flash memory, a Compact Disc Read Only Memory (CDROM), a Digital Versatile Disk (DVD), a magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. The memory 12 can be removable and / or non-removable (or fixed) as appropriate. The memory 12 can be internal or external as appropriate. In particular embodiments, the memory 12 is a Non-Volatile memory. In particular embodiments, the memory 12 includes a Read-Only Memory (ROM) and a Random-Access Memory (RAM). The ROM can be a mask-programmed ROM, a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically EPROM (EEPROM), an Electrically Alterable ROM (EAROM), or a FLASH, or a combination of two or more of these, as appropriate. The RAM can be a Static Random-Access Memory (SRAM) or a Dynamic Random-Access Memory (DRAM), which can be a Fast Page Mode Dynamic Random-Access Memory (FPMDRAM), an Extended Data Out Dynamic Random-Access Memory (EDODRAM), a Synchronous Dynamic Random-Access Memory (SDRAM), or the like, as appropriate.

[0069] The memory 12 can be used to store or buffer various data files required for processing and / or communication, and possible computer program instructions executed by the processor 11.

[0070] The processor 11 realizes the intersection group cooperative optimization method in any one of the above embodiments by reading and executing the computer program instructions stored in the memory 12.

[0071] In an embodiment, the electronic device can further include a communication interface 13 and a bus 10. As shown, the processor 11, the memory 12, and the communication interface 13 are connected through the bus 10 and complete communication among each other. Figure 5

[0072] The communication interface 13 is used to realize communication among various modules, devices, units, and / or equipment in the embodiments of the present application. The communication interface 13 can also realize data communication with other components, such as external devices, image / data acquisition devices, databases, external storage, and image / data processing workstations, etc.

[0073] ​Bus 10 includes hardware, software, or both, to couple components of electronic device to each other. Bus 10 includes, but is not limited to, at least one of a data bus, an address bus, a control bus, an expansion bus, a local bus, and the like. By way of example and not limitation, bus 10 can include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand (IB) interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or the like, or a combination of two or more of these. Where appropriate, bus 10 can include one or more buses. Although the present application is described and illustrated with a particular bus, it is contemplated that any appropriate bus or interconnect can be used.

[0074] In a third aspect, the embodiments of the present application provide a computer readable storage medium, having a program stored thereon, when the program is executed by a processor, the method for intersection group cooperative optimization in a man-machine mixed driving scene is realized.

[0075] In which, the more specific readable storage medium can include but not limited to: portable disk, hard disk, random access memory, read only memory, erasable programmable read only memory, optical storage device, magnetic storage device or any suitable combination of the above.

[0076] In possible implementation manners, the present application can also be implemented in the form of a program product, which comprises program codes for causing terminal equipment to execute the steps of the intersection group cooperative optimization method in a man-machine mixed driving scene provided by the first aspect when the program product is run on the terminal equipment.

[0077] The program codes for executing the present application can be written in any combination of one or more programming languages, and can be executed entirely on the user equipment, partially on the user equipment, as a stand-alone software package, partially on the user equipment and partially on a remote device, or entirely on a remote device.

[0078] The technical features of the above-described embodiments can be combined in any manner. To make the description concise, all possible combinations of the technical features in the above-described embodiments are not described, but it should be considered that any combination of the technical features is within the scope of the present disclosure, as long as the combination does not result in contradictions.

[0079] The above-described embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be noted that, for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A collaborative optimization method for intersection groups in a human-machine hybrid driving scenario, characterized in that, include: Obtain comprehensive regional information of the urban area to be optimized, and divide the urban area to be optimized into multiple traffic zones; Obtain vehicle purchase intention data for the traffic community, and predict the number of autonomous vehicles and human-driven vehicles in the traffic community based on the vehicle purchase intention data and a pre-built multiple linear regression model. Based on the pre-built binary Logit model, the ratio of human driving trips and the ratio of autonomous driving trips in the traffic zone are calculated respectively. Based on the trip ratio and ownership of the corresponding vehicle types, the trip volume of human driving vehicles and autonomous driving vehicles in the traffic zone is calculated, and the preset attraction volume of the traffic zone is obtained. Based on the number of trips generated and attracted by human-driven vehicles and autonomous vehicles, the traffic demand between the origin and destination of the traffic zone is calculated for both human-driven and autonomous vehicles. The road network of the traffic zone is obtained based on the comprehensive regional information, and a traffic allocation model is constructed based on the road network. The traffic allocation model is used to allocate the traffic demand between all origins and destinations in the traffic zone to the road network. An intersection group collaborative optimization model is constructed based on the traffic allocation model, and the intersection group collaborative optimization model is solved to obtain the final signal optimization scheme.

2. The method according to claim 1, characterized in that, The vehicle purchase intention data includes the age of the population within the traffic zone, the type of car they own, their understanding of new technologies, and their willingness to purchase a new car. The step of predicting the number of autonomous vehicles and human-driven vehicles in each traffic zone of the urban area to be optimized, based on the vehicle purchase intention data and a pre-built multiple linear regression model, includes: The age of the people in the traffic zone, the type of car they own, their understanding of new technologies, and their willingness to buy the next car are input into the autonomous vehicle purchase intention calculation formula in the multiple linear regression model to obtain the number of autonomous vehicles in operation. The age of the people in the traffic zone, the type of car they own, their understanding of new technologies, and their willingness to buy the next car are input into the human driving vehicle purchase intention calculation formula in the multiple linear regression model to obtain the number of human driving vehicles.

3. The method according to claim 1, characterized in that, The process involves calculating the human driving trip ratio and the autonomous driving trip ratio of the traffic zone based on a pre-built binomial Logit model. The trip volume of human-driven vehicles and autonomous vehicles in the traffic zone is calculated based on the trip ratio and vehicle ownership of the corresponding vehicle types. The attraction of the traffic zone is set to equal the trip volume, including: Obtain the personnel information of the population within the traffic zone, input the personnel information into the binary Logit model, and obtain the human driving trip ratio and autonomous driving trip ratio of the traffic zone; The number of trips by human driving is calculated by multiplying the ratio of human driving trips by the number of human driving vehicles, and the attraction of human driving vehicles in the traffic zone is set to be equal to the number of trips by human driving vehicles. The autonomous driving trip ratio is multiplied by the number of autonomous vehicles to calculate the number of trips taken by autonomous vehicles, and the attraction of autonomous vehicles in the traffic zone is set to be equal to the number of trips taken by autonomous vehicles.

4. The method according to claim 1, characterized in that, The traffic allocation model constructed based on the road network includes: The road network of the traffic zone is obtained from the comprehensive regional information, and the road network includes road segments and intersections; Construct a travel cost function and calculate the comprehensive travel cost of the vehicle route based on the travel cost function; wherein, the travel cost function includes the BPR function for estimating the travel time of road segments and the Webster formula for estimating the travel time of intersections; By combining the comprehensive travel costs of the vehicle routes, the traffic demand of human-driven vehicles and autonomous vehicles is allocated through an improved user equilibrium model, resulting in the traffic allocation model.

5. The method according to claim 1, characterized in that, The construction of the intersection group collaborative optimization model includes: Obtain the vehicle traffic conditions and signal control parameters of each road segment and intersection in the traffic zone; A formula for calculating the total travel time of the route is constructed based on the aforementioned vehicle traffic conditions and signal control parameters. The intersection group collaborative optimization model is obtained by combining the formula for calculating the total travel time of the path with the traffic allocation model.

6. The method according to claim 5, characterized in that, The step of solving the intersection group collaborative optimization model based on the traffic allocation model to obtain the final signal optimization scheme includes: An initial signal optimization scheme is determined based on the current traffic conditions at the intersection. The initial signal optimization scheme includes the signal cycle length and green light time of the intersection. The traffic allocation model is solved according to the initial signal optimization scheme to obtain the traffic allocation result, which includes the traffic flow of human driving and autonomous driving, as well as the flow direction at each intersection. The total travel time of the path is calculated based on the traffic allocation results and the current signal optimization scheme. Determine whether the change in the total travel time of the path is within a preset tolerance range. If the determination result is yes, it is considered that the optimization has converged, and the current signal optimization scheme is output as the final optimization scheme; otherwise, based on the selection, crossover and mutation operations of the genetic algorithm, the signal cycle length and green light time of the intersection are adjusted, and the current signal optimization scheme is updated to continue iterating.

7. The method according to claim 3, characterized in that, The multiple linear regression model is as follows: in, For age, For commuting time, For the number of family members, For the family's annual income, This represents the projected number of human-driven vehicles. The projected number of autonomous vehicles. , , , , To predict the regression coefficients for the number of vehicles driven by humans. , , , , , Regression coefficients for predicting the number of autonomous vehicles.

8. The method according to claim 1, characterized in that, The binomial Logit model is as follows: in, The percentage of human driving trips within a traffic zone. The percentage of autonomous driving trips within a traffic-intensive community. It is the observed utility of choosing to use one's own human-driven vehicles for travel. The observed utility is the choice to use their own autonomous vehicles for travel; similarly, The observed utility is the use of public transportation. For age, For gender, For travel duration, , , , , , , , , , , , is the regression coefficient.

9. The method according to claim 4, characterized in that, The traffic allocation model is as follows: Wherein, OBJ is the integral of the cumulative travel time of traffic flow on the road segment. This refers to taking the smallest value of OBJ. Values, , These represent the traffic demand for human driving and autonomous driving between any origin and destination pair between rs. Let i be the human driving traffic flow along path i. Let i be the traffic flow for autonomous driving. , The traffic flow for human driving and autonomous driving is shown for road segment a, respectively. This indicates whether the i-th path in the origin-endpoint pair rs passes through the intersection entrance lane numbered b, where a is the road segment number, A is the set of road segments, l is the intersection entrance lane number, and L is the set of intersection entrance lanes.

10. The method according to claim 5, characterized in that, The intersection group collaborative optimization model is as follows: Where TTT represents the total travel time of the route. For road section Free-flow travel time; For road section Traffic; For road section Traffic capacity; It is an intersection Import channel The average queue length of vehicles; It is an intersection Import channel Traffic flow; intersection Import channel The average speed of vehicles at the intersection; intersection Import channel saturation flow rate The current signal optimization scheme for intersection b includes the signal cycle length of each intersection and the green light time for each phase.

11. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the intersection group collaborative optimization method in the human-machine hybrid driving scenario as described in any one of claims 1 to 10.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the intersection group collaborative optimization method in the human-machine hybrid driving scenario as described in any one of claims 1 to 10.

Images