Intelligent network connection mixed flow fusion control method for expressway in foggy days

Abstract

The invention discloses an intelligent network connection mixed flow fusion control method for highways in foggy days, and belongs to the technical field of traffic control. The method comprises the following steps: firstly, establishing a highway basic information database and a foggy day management and control sample library; dividing the expressway into a plurality of subsystems, and dividing each subsystem into a plurality of road sections; establishing a foggy day management and control knowledge base of each road section, wherein the foggy day management and control knowledge base comprises an HVs speed limit upper limit, a CAVs speed limit upper limit and a maximum allowable flow under different visibility and permeability; and finally, predicting the visibility of foggy days which are about to appear on the expressway in the next few days, making a control plan, and issuing traffic guidance information in advance. For foggy days appearing on the expressway, the demand quantity, flow and visibility of CAVs and HVs are collected in real time, and the road section is refined according to the visibility; and determining the HVs and CAVs parking amount of each entrance ramp and the HVs and CAVs speed upper limits of each road section, and carrying out online real-time control. Visibility and mutual influence of HVs and CAVs are fully considered, and the traffic capacity is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of traffic management and control in transportation engineering, and in particular relates to a method for intelligent networked mixed flow fusion control on foggy highways. Background Art

[0002] Connected Automatic Vehicles (CAVs) use onboard sensing devices like radar and video, along with roadside equipment, to quickly and accurately collect and perceive road and environmental information. They then employ various collaborative algorithms to control vehicle operation. This approach essentially addresses the reduced visibility experienced by human vehicles (HVs) in foggy conditions, minimizing the negative impacts of fog. However, full CAV deployment will likely remain for a long time to come, resulting in heterogeneous traffic flows mixed with HVs. The challenges associated with HVs in foggy conditions persist, with CAVs experiencing reduced speeds and queue sizes, leading to mode degradation. Traditionally, highway management agencies have implemented various measures in foggy conditions, such as mainline speed limits, ramp flow restrictions (including extreme closures), and integrated control, to maximize traffic safety and improve road capacity. However, these measures are designed for all HVs. Under the hybrid flow of intelligent connected vehicles, if control is implemented based on CAVs, safety cannot be guaranteed; if the policy of "allowing CAVs to pass while prohibiting HVs from entering" is adopted, social fairness between CAVs and HVs cannot be achieved; if the management technical standards are determined based on the existing mainline and ramp integrated control method, the technical advantages of CAVs cannot be utilized, and the technical standards will also vary under different penetration rates. Therefore, the implementation of refined integrated control measures based on the different visibility and penetration rates of each road section, the implementation of quasi-all-weather highway access while ensuring traffic safety, maximizing social fairness, and improving actual highway capacity have become one of the technical issues that relevant government departments, highway management departments, and travelers urgently need to address.

[0003] Existing research technologies are divided into three aspects:

[0004] 1. Traditional (all HVs) foggy highway mainline and ramp integrated control

[0005] For example, Wang Min (Wang Min. Research on Intelligent Guidance Strategy for Driving in Normal Fog Zones on Highways [D]. Chongqing: Chongqing Jiaotong University, 2015.) analyzed the vehicle operation and accident characteristics in foggy areas on highways, established an adaptive neural network model for speed and distance control in foggy areas on highways, and obtained an intelligent guidance strategy for driving in foggy areas; Gong Lingyan et al. (Gong Lingyan, Wang Keke, Mao Xuejun, et al. Traffic control method for highways under foggy conditions [J]. Transportation Science and Engineering, 2017, 31(1): 85-90.) established an improved foggy highway traffic flow model based on METANET, and constructed a mainline and ramp coordinated prediction control model with the coordination of mainline operation efficiency, ramp queue length constraint and driving safety as the goal; Feng Fengjiang et al. (Feng Fengjiang, Cheng Xinping, Liu Huiyang, et al. Coordinated control method for highways in foggy conditions based on cellular automaton model: China, 112396834 [P]. 2021-09-07.) established a cellular automaton model for each section of highway in foggy conditions (Cellular Automaton A coordinated control strategy for maximum speed limits and optimal ramp flow rates on highway entrances and exits is established based on the CAM (Computational Automated Vehicle Model) for different visibility conditions on each road section. This strategy is implemented based on information collected by the detection system, ensuring safety while determining coordinated control strategies for maximum speed limits and optimal ramp flow rates on each highway section. However, these approaches are limited to the integrated control of mainline and ramp traffic in foggy conditions involving HVs and do not address the integrated control of highway traffic in foggy conditions involving mixed traffic flows of CAVs and HVs.

[0006] 2. Integrated control of highways under intelligent networking

[0007] For example, Baskar et al. (Baskar LD, Schutter BD, Hellendoorn H, Traffic management for automated highway systems using model-based predictive control [J]. IEEE Transactions ITS, 2012, 13(2): 838–847.) proposed an integrated traffic management and hierarchical traffic control method for AHS. Based on the model predictive control idea, the queue speed and trajectory, as well as the time to enter AHS were determined to optimize the system performance, and dynamic speed limit, lane allocation and ramp flow were used as control measure variables; Rubin et al. (Rubin I, Andrea B, Yu l S, et al. Traffic management and networking for autonomous vehicular highway systems [J]. Ad Hoc Networks, 2019, 83: 125-148.) proposed an integrated coordinated control system based on comprehensive management of vehicle queues for multi-lane and multi-section AHS systems; Perraki et al. (Perraki G, Roncoli C, Papamichail I, et al. Evaluation of a model predictive control framework for motorway traffic involving conventional and automated A basic framework for fused coordinated predictive control of highways under heterogeneous traffic flows was proposed. This framework uses a multi-layered collaborative control framework to achieve dynamic feedback control. However, the paper focuses on CAVs, with some applications involving mixed traffic flows. The paper is limited to normal weather conditions and does not cover adverse weather conditions. It does not involve fused control based on CAM highway segmented micro-simulation.

[0008] 3. Intelligent connected highway control in foggy weather

[0009] For example, Zhang Cunbao et al. (Zhang Cunbao, Lü Changping, Zhang Shan, et al. Highway safety speed guidance system and method for intelligent connected vehicles in foggy conditions: China, 106251666 [P]. 2016-12-21.) provide a highway safety speed guidance system for CAVs in foggy conditions, which sends the real-time visibility and vehicle status information to the roadside equipment and calculates the optimal vehicle speed through relevant modules; Zhao et al. (Zhao XH, Xu WX, Ma JM, et al. Effects of connected vehicle-based variable speed limit under different foggy conditions based on simulated driving [J]. Accident Analysis and Prevention, 2019, 128: 206-216.) establish a connected vehicle test platform based on a driving simulator, analyze the speed adjustment characteristics of drivers after receiving warnings for different visibility levels, and determine the variable speed limit value for connected vehicles in foggy conditions; Gong et al. (Gong BW, Wei RX, Wu D Y, et al. Fleet Management for HDVs and CAVs ... On Highway in Dense Fog Environment[J]. Journal of Advanced Transportation, 2020, 2020(2):1-21.) Based on a distributed model predictive control algorithm, combined with CAM and considering the following behavior of drivers of high-density vehicles, the spatial state equations for CAVs and HVs on foggy highways are established, and a mixed traffic flow platoon control framework and distributed platoon control method are proposed. Disadvantages: This method focuses on speed limits and platoon control on the main line of highways, does not involve on-ramp control, and does not involve integrated control of the main line and ramp coordination; and partially applies to mixed traffic flow environments.

[0010] Therefore, based on the existing traffic flow nonlinearity, traffic control technology, and smart transportation, a method for the integrated control of mixed flow main lines and multiple ramps on foggy highways is developed. This method is aimed at intelligent connected mixed traffic flow, determines the upper speed limit of HVs and CAVs under different visibility and penetration rates of each section, and the maximum allowable flow (maximum allowable traffic demand, referred to as maximum allowable flow). On this basis, based on the information collected by the perception system, the maximum speed limit of HVs and CAVs on each section and the coordinated release volume of CAVs and HVs on each entrance ramp are quickly determined, so that vehicles traveling on each section are within the maximum speed limit range, and vehicles coming from upstream of the section do not exceed the maximum allowable flow, ensuring traffic safety and achieving social fairness, and maximizing its flow." This method has become an urgent problem to be solved in traffic control and management. Summary of the Invention

[0011] To address the shortcomings of existing technologies, the present invention aims to provide a method for integrated control of mainline and ramp traffic on foggy highways using intelligently connected hybrid flows. This method establishes a CAM model for each section based on the characteristics of the highway, determining the upper speed limits and maximum allowable flow rates for HVs and CAVs under different visibility and permeability conditions. During implementation, the method determines the upper speed limits for HVs and CAVs on each section and the number of CAVs and HVs allowed on each entry ramp based on road information collected by the perception system. This allows for refined integrated control of highways while ensuring traffic safety and social equity, meeting quasi-all-weather traffic requirements, improving traffic efficiency, and reducing unnecessary ramp queues and time.

[0012] The present invention solves the technical problem by adopting the following technical solutions:

[0013] A method for controlling intelligent networked mixed flows on foggy highways, comprising the following steps:

[0014] Step 1: Establish a basic highway information database;

[0015] Step 2: Establish a fog control sample library;

[0016] Step 3: Divide the highway into several subsystems, and each subsystem into multiple sections. Build a fog management knowledge base for each section, including upper speed limits for HVs and CAVs, and maximum allowable flow rates under different visibility and penetration rates.

[0017] Step 4: Forecast visibility in foggy days on highways over the next few days, formulate emergency control plans, and release traffic guidance information in advance;

[0018] Step 5: In foggy conditions on highways, determine the number of HVs and CAVs allowed on each on-ramp and the upper speed limits for HVs and CAVs on each road section, and implement online real-time control.

[0019] ① Collect CAVs and HVs demand, traffic volume, and visibility in real time, and refine road sections based on visibility;

[0020] ② Let the subsystems affected by fog be i = i1, i1 + 1, ..., n, where i1 is the first subsystem affected by fog and n is the most downstream subsystem affected by fog. Let i = n. For subsystem i, if the upstream contains an interchange with other highways, turn to step 9; otherwise, turn to step 3.

[0021] ③Calculate the upstream permeability P of the main line A1 , entrance ramp permeability P A2 , comprehensive permeability of the main line upstream and entrance ramp P A3 ;

[0022] ④Based on P A3 and visibility, query the fog control knowledge base of each road section of subsystem i, and obtain the maximum allowable flow of subsystem i in, represents the maximum allowable flow rate of the jth road section of subsystem i, n i represents the number of road sections contained in subsystem i, Q Di+1 Indicates that subsystem i+1 needs to reduce flow;

[0023] ⑤If Turn to ⑥; otherwise, let Q Di =0, according to P A3 and visibility, query the fog control knowledge base of each road section of subsystem i, determine the upper speed limit of HVs and CAVs of each road section of subsystem i, and turn to ⑩; where q ui represents the total upstream demand of the main line, r di represents the total demand for the entrance ramp, r i 、r i C 、r i H represents the total number of vehicles released at the entrance ramp of subsystem i, and the number of CAVs and HVs released, represents the demand for CAVs and HVs on the entrance ramp, Q Di Indicates that subsystem i needs to reduce flow;

[0024] ⑥ If Turn to ⑧; if and Turn to ⑦; otherwise, let Q Di=0, according to P A3 and visibility, query the fog control knowledge base of each road section of subsystem i, determine the upper speed limit of HVs and CAVs of each road section of subsystem i, and turn to ⑩;

[0025] ⑦ Determine the number of CAVs and HVs allowed on the entrance ramp when the demand for HVs on the entrance ramp is low, as well as the upper speed limits for HVs and CAVs on each road section;

[0026] a. Order r i C =r i H ·P A2 / (1-P A2 ), r i =r i C +r i H , Q i =q ui +r i ;

[0027] b. If Turn to c, otherwise turn to f;

[0028] c. Let r i C =r i C +1, updated r i and Q i ;

[0029] d. Order

[0030] e. According to P A4 and visibility, query the fog control knowledge base of each road section of subsystem i, and update Turn to b until Turn to f;

[0031] f. Let r i =r i C +r i H , Q i =q ui +r i , Q Di =0, according to P A4 and visibility, query the fog control knowledge base of each road section of subsystem i, determine the upper speed limit of HVs and CAVs of each road section of subsystem i, and turn to ⑩;

[0032] ⑧Confirm The number of CAVs released at that time and the upper speed limit of HVs and CAVs on each road section;

[0033] a. Let r i C =0, r i H = 0, initialize the set R of CAVs cases that can be placed in the entrance ramp to an empty set;

[0034] b. Let r i C =r i C +1, r i =r i C , Q i =q ui +r i ,calculate According to P A5 and visibility, query the fog control knowledge base of each road section of subsystem i, and determine

[0035] c. If Add the case to set R, otherwise do not add it and go to d;

[0036] d. If Turn to b; otherwise, turn to e;

[0037] e. If the set R is an empty set, go to g; otherwise, go to f;

[0038] f. The corresponding maximum value in set R In the case of i C ; Let Q Di =0, according to P A5 and visibility, query the fog control knowledge base of each road section of subsystem i, determine the upper speed limit of HVs and CAVs of each road section of subsystem i, and turn to ⑩;

[0039] g. Let r i =r i C =r i H =0, according to P A1 and visibility, query the fog control knowledge base of each road section of subsystem i, and determine the upper speed limit of HVs and CAVs on each road section of subsystem i;

[0040] h. If i≠i1, let Turn to ⑩; otherwise, let Turn to ⑩; where s is the number of closed entrance ramps upstream of subsystem i, is the flow rate of the exit ramp of subsystem i1-k, is the total upstream demand and maximum allowable flow of the main line of subsystem i1, is the total number of vehicles released from the entrance ramps of subsystems i1-1, i1-2, i1-s, and i1-s-1;

[0041] ⑨ For subsystem i, if the upstream contains an "interchange bridge that intersects with other highways", then:

[0042] a. Calculate the comprehensive penetration rate after the main line upstream vehicles intersect with other high-speed vehicles According to P A6 and visibility, query the fog control knowledge base of each road section of subsystem i, and obtain the maximum allowable flow of subsystem i in, and are the demands of CAVs and HVs of other high-speed merging subsystem i, respectively;

[0043] b. If Let Q Di =0, according to P A6 and visibility, query the fog control knowledge base of each road section of subsystem i, determine the upper limit of the speed limit of HVs and CAVs on each road section, and turn to ⑩; otherwise, turn to c; where q 2ui represents the total demand of CAVs and HVs from other high-speed merging subsystem i;

[0044] c. Computing subsystem i needs to reduce traffic According to P A6 and visibility, query the fog control knowledge base of each road section of subsystem i, determine the upper speed limit of HVs and CAVs, and turn to ⑩; where β i is the ratio of highway vehicles to upstream vehicles of subsystem i;

[0045] ⑩If i=i1, go to ①, otherwise let i=i-1, go to ②;

[0046] Step 6: Obtain the newly added fog control samples, return to steps 2 and 3, and modify the fog control sample library and fog control knowledge base.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] (1) Different from the traditional fusion control method for foggy highways, the present invention fully considers the mutual influence of HVs and CAVs and the differences caused by different penetration rates when constructing the CAM model as the basis for determining the speed limit and traffic capacity of different visibility sections. In terms of fusion control, when the demand for HVs is lower than the number of HVs released, the road flow can be increased by increasing the number of CAVs released, which is in line with the goal of improving safety and traffic capacity through the development of smart highways.

[0049] (2) Different from the existing coordinated control methods for highways under intelligent networking, the present invention not only controls the main line traffic flow, but also provides integrated control of the main line and ramps of the highway in foggy environments, which meets the actual needs of highway management and is suitable for large-scale promotion and application.

[0050] (3) In terms of the specific control measures, a refined control method is adopted that takes into account the principle of social fairness. Different speed limit upper limits are adopted for each section of the highway, and the number of CAVs is increased only when the demand for HVs is lower than the number of HVs available. This satisfies the principle of fairness and refined traffic flow regulation, thus ensuring traffic safety to the greatest extent possible, improving road traffic efficiency, and reducing the negative impact of fog.

[0051] (4) In terms of applicable objects, this method is applicable to highways where CAVs and HVs coexist for a long period of time, fog lasts for a long time (e.g., more than 4 hours), visibility distribution is uneven due to the environment (e.g., passing through mountainous areas, lakes, rivers, etc.), and the road closure standards for HVs are not met. This makes this method adaptable to the development needs of smart highways, with low control costs and good results. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is the overall flow chart of the present invention;

[0053] Figure 2 A schematic diagram of a highway composed of several subsystems of the present invention;

[0054] Figure 3 This is a schematic diagram of a highway subsystem (composed of several sections) according to the present invention; the section between two adjacent upstream and downstream entrance ramps, excluding an interchange with another highway, is divided into one subsystem (a); if it does, the section is divided into two subsystems (b) with the interchange as the dividing point.

[0055] Figure 4 A flow chart for determining the technical standards for fog control on various road sections according to the present invention;

[0056] Figure 5 Flowchart for establishing CAM models for each road section in the present invention;

[0057] Figure 6Flowchart of online real-time control for the present invention;

[0058] Figure 7 This is a control flow chart of a subsystem of the present invention that does not include an interchange bridge intersecting with other highways upstream;

[0059] Figure 8 This is a control flow chart of the present invention when the total upstream demand of the main line is not greater than the maximum allowable flow rate and the demand of the HVs on the entrance ramp is relatively small;

[0060] Figure 9 This is a control flow chart for the present invention when the total upstream demand of the main line is greater than the maximum allowable flow rate;

[0061] Figure 10 The upstream of the subsystem of the present invention contains a control flow chart of an interchange bridge intersecting with other highways;

[0062] Figure 11 This is a partial map of the Cangyu Expressway (from Fuping East to Baoding West) in the embodiment of the present invention;

[0063] Figure 12 Schematic diagram of subsystem division in an embodiment of the present invention. DETAILED DESCRIPTION

[0064] Specific embodiments are given below in conjunction with the accompanying drawings. The specific embodiments are only used to introduce the technical solutions of the present invention in detail and are not intended to limit the scope of protection of the present application.

[0065] The present invention provides a method for controlling the fusion of intelligent network mixed flows on foggy highways (hereinafter referred to as the method, see Figures 1 to 12 ), with multiple entrance and exit ramps, main lines, interchanges with other highways, various perception systems, HVs and CAVs, etc. as the research object, the basic principle is: establish a basic information database of highways, collect data such as traffic flow, speed, accident location, cause, etc. under different visibility and penetration rates in foggy days, and establish a foggy day control sample library; divide the highway into several subsystems, each subsystem consists of several sections, establish the CAM model of each section, and determine the safety of each section through model verification, parameter calibration, simulation experiments, etc. A fog control knowledge base is developed for road sections that "combine the maximum allowable flow and speed limit of the main line under different visibility and permeability"; based on meteorological information, visibility is predicted for foggy days on highways in the next few days, various emergency plans are formulated, and guidance information is released in advance; when foggy days occur on highways, the number of HVs and CAVs allowed on each entrance ramp and the speed limit of HVs and CAVs on each road section are determined, and real-time online control is performed; newly added fog control samples are obtained, and the fog control sample library and the fog control knowledge base of the corresponding road section are revised. The steps of this method are as follows:

[0066] Step 1: Establish a basic highway information database: GIS including road geometry, historical traffic flow data, meteorological data, and road information;

[0067] Step 2: Establish a fog control sample library that includes data on highway traffic volume, speed, accident locations, and causes under different fog conditions (visibility levels) and penetration rates.

[0068] Step 3: Divide the highway into several subsystems, and divide each subsystem into multiple sections (see Figure 2 and 3 ) ; Use the fog control sample library to establish the CAM model of each road section, and use the CAM model simulation to obtain the upper limit of the HVs speed limit of each road section under different visibility and penetration rates Upper speed limit for CAVs and the maximum allowable flow rate Q max , and obtain the fog control knowledge base of each road section (see Figure 4 );

[0069] ① The expressway is divided into several subsystems based on whether there is an interchange with other expressways between the two adjacent upstream and downstream entrance ramps. If there is no interchange with other expressways between the two adjacent upstream and downstream entrance ramps, the portion between the adjacent upstream and downstream entrance ramps of the expressway is divided into one subsystem. If there is, the portion between the adjacent upstream and downstream entrance ramps of the expressway is divided into two subsystems with the interchange as the dividing point. For details, see Figure 2 ;

[0070] ② Each subsystem is divided into several sections according to the road geometry (such as straight road, curve, downhill slope, uphill slope, uphill curve, downhill curve, etc.). Figure 3 ;

[0071] ③ Using the foggy weather control sample library, a CAM (cellular automation) model is established for each road section with mixed traffic consisting of HVs and CAVs in foggy weather. Based on the CAM model, the evolution rules of HVs and CAVs, including longitudinal following, lateral lane changing, acceleration and deceleration, random slowing, position update, etc., are determined. The accident discrimination rules, boundary conditions, accident probability, traffic flow state variables, etc. are also determined. The specific process is shown in [1]. Figure 5, see "Simulation of Mixed Traffic Flow of Artificial and Intelligent Connected Vehicles on Highways in Fog" (Feng Fengjiang, Cheng Xinping, Sun Xiaoning. Simulation of Mixed Traffic Flow of Artificial and Intelligent Connected Vehicles on Highways in Fog [J]. Journal of China Safety Science. 2022, 32(07): 158-164.), "CAM of Possible Accidents on Downhill Curves on a Mountain Highway" (Pang Mingbao, Cai Zhanghui. CAM of Possible Accidents on Downhill Curves on a Mountain Highway [J]. Journal of System Simulation, 2018, 30(4): 1414-1422.), and Zhang et al. (Zhang YT, Hu MB, Chen YZ, et al. Cooperative platoonforming strategy for connected autonomous vehicles in mixed traffic flow [J]. Physica A: Statistical Mechanics and its Applications, 2023, 623: 128828.) CAVs platooning strategy in mixed traffic flow;

[0072] ④CAM model verification and parameter calibration: Through simulation experiments (taking the average of at least 40 experiments), driving simulation experiments, and traffic data and accident probabilities obtained from the basic data sample library, the data and accident probabilities of the three methods are compared and analyzed to verify the CAM model, and the model parameters are corrected to obtain the corrected CAM model;

[0073] ⑤ Using the modified CAM model, the simulation experiments were conducted to determine the road sections with different visibility and permeability. and Q max For each road section, under different visibility and permeability, by inputting different upstream flow rates and setting different speed limit values, simulation experiments (average value of at least 40 experiments) are carried out to obtain the maximum allowable flow rate, accident probability and other indicators under different visibility and permeability. Through comprehensive evaluation and comparison, the safety and efficiency of different visibility and permeability are determined. and Q max ;

[0074] ⑥ The road sections under different visibility and permeability and Q max As the technical standard for fog control of the road section, a fog control knowledge base of the road section is obtained;

[0075] Step 4: Based on meteorological data provided by the meteorological station, the visibility of foggy days on the highway in the next few days will be predicted, and a control plan will be formulated. Traffic guidance information will be issued in advance to reduce traffic demand on the highway.

[0076] Step 5: When fog occurs on highways, determine the number of HVs and CAVs allowed on each entrance ramp and the upper speed limit for HVs and CAVs on each road section, and perform online real-time control (see the detailed process for details). Figure 6 );

[0077] ① The highway sensing system collects CAVs and HVs demand, traffic flow, and visibility in real time, and refines the road sections divided in step 3 based on visibility;

[0078] a. Based on visibility, the road segments divided in step 3 are further refined. If the visibility at each monitoring point on a road segment is the same, no further refinement is performed. If the visibility is different, the road segment is further refined into multiple sections based on the visibility differences.

[0079] b. Based on the fog control technical standards obtained in step 3, determine the maximum allowable flow rate, upper speed limit for HVs and upper speed limit for CAVs for each road section at different visibility and permeability; and represents the maximum allowable flow rate, upper speed limit of HVs and upper speed limit of CAVs on the j-th road segment of subsystem i;

[0080] c. The subsystems affected by fog are denoted as i = i1, i1 + 1, ..., n, where i1 is the first subsystem affected by fog where the on-ramp can implement flow restrictions, and n is the most downstream subsystem affected by fog.

[0081] ② Let i = n. For subsystem i, if the upstream contains an "interchange with other highways", turn to 9, otherwise turn to 3. The specific process is shown in Figure 7 ;

[0082] ③Calculate the upstream permeability of the main line On-ramp penetration Combined penetration rate of mainline upstream and entrance ramps in, and The demand for CAVs and HVs upstream of the main line, and the total demand for upstream of the main line and The demand for CAVs and HVs on the entrance ramp, and the total demand for the entrance ramp are

[0083] ④Based on P A3 and visibility, query the fog control technical standards to obtain P A3 The maximum allowable flow of each section is obtained, and the maximum allowable flow of subsystem i is where Q Di+1The traffic flow of subsystem i+1 needs to be reduced due to the fact that “even if the entrance ramp of downstream subsystem i+1 is closed due to low visibility, it is necessary to reduce the number of vehicles coming from the upstream of the main line to meet the safety requirements”. When i=n, ​​let Q Dn+1 = 0 indicates that the most downstream subsystem n is only limited by the maximum allowable flow of each section of its own road; i represents the number of road sections included in subsystem i;

[0084] ⑤If Turn to ⑥; otherwise, Let r i =r i C +r i H , Q Di =0, according to P A3 and visibility, query the fog control technical standards in the third step to determine the upper speed limit of HVs and CAVs on each road section of subsystem i, where the upper speed limit of HVs and CAVs on the jth road section in subsystem i is recorded as r i 、r i C 、r i H represents the total number of vehicles released at the entrance ramp of subsystem i, the number of CAVs and HVs released, Q Di Indicates that subsystem i needs to reduce flow. This condition indicates that when the sum of the total demand of the mainline upstream and the on-ramp is less than the maximum allowable flow of the subsystem, there is no need to limit the flow of the on-ramp. All vehicles on the on-ramp can be released to the highway. At the same time, there is no need to reduce the number of vehicles on the upstream subsystem to meet the safety and flow maximization requirements. Turn to ⑩;

[0085] ⑥ If Turn to ⑧; if and Turn to ⑦; otherwise, let Q Di =0, according to P A3 Based on the visibility, the fog control technical standards in step 3 are queried to determine the upper speed limits for HVs and CAVs on each road section of subsystem i. This condition indicates that the total demand for the subsystem exceeds the maximum allowable flow rate, and the proportion of HVs on the on-ramp demand is large. "Reducing on-ramp release" can be used to meet safety and maximize flow. However, to ensure fairness, the release rates of CAVs and HVs are determined based on their respective on-ramp demand ratios. This also eliminates the need to reduce incoming traffic from the upstream subsystem. Turn to (10);

[0086] ⑦ Determine the vehicle release capacity and speed limit of the entrance ramp when the demand for HVs on the entrance ramp is small (see the detailed process for details). Figure 8 );

[0087] a. Order r i C =r i H ·P A2 / (1-P A2 ), r i =r i C +r i H , Q i =q ui +r i ;

[0088] b. If Turn to c, otherwise turn to f;

[0089] c. Let r i C =r i C +1, updated r i and Q i ;

[0090] d. Order

[0091] e. According to P A4 and visibility, query the fog control technical standards of each section of subsystem i, and update Turn to b until Turn to f;

[0092] f. Let r i =r i C +r i H , Q i =q ui +r i , Q Di =0, by P A4 and visibility to determine the upper speed limit of HVs and CAVs in each road section of subsystem i, turning to ⑩;

[0093] This condition indicates that when the sum of the total demand on the mainline upstream and on-ramp exceeds the maximum allowable flow of the subsystem, and the demand for HVs on the on-ramp is small, all HVs on the on-ramp can be accommodated on the highway. Safety and flow maximization requirements, as well as the principle of fairness, can be met by "appropriately increasing the number of CAVs accommodated on the on-ramp to increase the maximum allowable capacity of the mainline" without reducing the number of vehicles entering the upstream subsystem.

[0094] ⑧Confirm The number of CAVs released and the upper limit of speed limit (see the detailed process for details) Figure 9 ), which indicates that the upstream demand on the main line is greater than the maximum allowable flow of the subsystem (because the maximum allowable capacities of the upstream and downstream subsystems may be different). In this case, the approach of "increasing the number of CAVs released on the entrance ramp, thereby increasing the maximum allowable capacity of the main line" can be tried to meet the requirements of safety, maximum flow, and fairness. Otherwise, the entrance ramp must be closed and the number of vehicles entering the upstream subsystem must be reduced.

[0095] a. Let r i C =0, r i H = 0, initialize the set R of CAVs cases that can be placed in the entrance ramp to an empty set;

[0096] b. Let r i C =r i C +1, r i =r i C , Q i =q ui +r i ,calculate According to P A5 and visibility, query the control technical standards of each road section of subsystem i, and determine

[0097] c. If Add the case to the initial set R, otherwise do not add it and go to d;

[0098] d. If Turn to b; otherwise, turn to e;

[0099] e. If the set R is an empty set, go to g; otherwise, go to f;

[0100] f. The corresponding maximum value in set R In the case of i C , Q Di = 0, and P A5 and visibility to determine the upper speed limit of HVs and CAVs in each road section of subsystem i, turning to ⑩;

[0101] g. Let r i =r i C =r i H =0, by P A1 and visibility to determine the upper speed limit for HVs and CAVs on each road segment of subsystem i;

[0102] h. If i≠i1, let Turn to ⑩; otherwise, let Where s is the number of closed entrance ramps upstream of subsystem i, is the flow rate of the exit ramp of subsystem i1-k, is the total upstream demand and maximum allowable flow of the main line of subsystem i1, is the total number of vehicles released from the entrance ramps of subsystems i1-1, i1-2, i1-s, and i1-s-1. This condition indicates that the upstream traffic flow of the main line of the first subsystem affected by fog is greater than its maximum allowable flow rate, causing the entrance ramp of this subsystem to be closed. In addition, the upstream traffic flow of the main line needs to be reduced. This is achieved by adjusting the entry of the upstream subsystem entrance ramp. Among them, the upstream s entrance ramps are closed, and the entrance ramp of subsystem i1-s-1 is adjusted, turning to ⑩.

[0103] ⑨ For subsystem i, if there is an "interchange bridge connecting with other highways" upstream, the specific process is shown in Figure 10 ;

[0104] a. Calculate the comprehensive penetration rate after the main line upstream vehicles intersect with other high-speed vehicles in and are the demands of CAVs and HVs of other high-speed inflow subsystem i, according to P A6 and visibility, query the fog control technical standards, and obtain the maximum allowable flow of each section of subsystem i, and then obtain the maximum allowable flow of subsystem i as

[0105] b. If Let Q Di =0, according to P A6 The upper speed limit for HVs and CAVs on each road section is determined based on the visibility. This condition indicates that when the sum of the upstream demand of the main line and the incoming demand from other highways is less than the maximum allowable flow of subsystem i, there is no need to reduce the upstream traffic of this subsystem to meet the safety and flow maximization requirements, and turn to ⑩; otherwise, turn to c; where, represents the total demand of CAVs and HVs of other high-speed merging subsystem i,

[0106] c. Computing subsystem i needs to reduce traffic According to P A6 The upper speed limit for HVs and CAVs is determined based on the visibility of each road section. This condition indicates that when the sum of the upstream demand of the main line and the demand from "other highways" is greater than the maximum allowable flow of subsystem i, the upstream vehicles must be reduced in proportion to meet the safety and flow maximization requirements, turning to ⑩; where βi is the ratio of highway vehicles to upstream vehicles of subsystem i, 1-β i is the ratio of vehicles coming from other highways (through interchanges) to vehicles coming from the upstream of subsystem i;

[0107] ⑩If i=i1, go to ①, otherwise let i=i-1, go to ②;

[0108] Step 6: Obtain the newly added fog control samples, return to steps 2 and 3, and revise the fog control sample library; use the revised fog control sample library to retest and revise the CAM model parameters, and then revise the fog control technical standards and fog control knowledge base of the corresponding road section.

[0109] In the present invention, GIS data including road geometry and historical traffic flow data are obtained through highway operating companies and management departments. At the same time, traffic flow status information data such as visibility and permeability can be obtained in real time by highway companies and management departments through various perception systems on intelligent connected highways. Meteorological data and predicted visibility in foggy days are obtained through meteorological stations, meteorological management departments, and highway perception systems.

[0110] In the present invention, the regulation (opening or closing) and release of vehicles on the entrance ramp are achieved through the toll gate before the existing highway entrance ramp, and the ramp release flow rate is controlled through traffic control in foggy weather.

[0111] Example

[0112] The test site of this embodiment is selected from the Fuping East to Baoding West section of the Cangyu Expressway in Hebei Province. Figure 11 and 12 As shown in the figure, P represents the entrance ramp, R represents the exit ramp, I represents the interchange entrance, and E represents the interchange exit. The numerical sequence represents the subsystem number. This section of expressway includes six pairs of entrance and exit ramps (one for each exit ramp) and two interchanges connecting to other expressways. The main line is a four-lane, two-way highway with a length of 106 kilometers. The entire road is divided into three sections: the mountainous area from Fuping East to Tang County, the hilly area from Tang County to the Beijing-Kunming Expressway, and the plain area from the Beijing-Kunming Expressway to Baoding West. The hilly and mountainous areas, especially the mountainous areas, have more curves and longitudinal slopes. The hilly areas have the lowest visibility in fog due to the presence of rivers, while the mountainous areas have slightly higher visibility in fog due to strong winds. According to surveys, the proportion of trucks on the road is approximately 35%.

[0113] According to the third step of the present invention, the expressway is divided into 7 subsystems, the specific scopes of which are shown in Table 1 and Figure 12Each subsystem is further divided into several sections based on road geometry. The characteristics of the subsystems 4 and 5 are shown in Tables 2 and 3. Taking into account road geometry and safety requirements, fixed maximum speed limits are applied under normal traffic conditions on sunny days (non-congested, non-accidental, road repair, etc.). The upper speed limits for cars and trucks in subsystem 4 are 100 km / h and 70 km / h, respectively. The upper speed limits for cars and trucks in subsystem 5 (hilly terrain) and subsystems 6 and 7 (plain terrain) are both 120 km / h and 90 km / h.

[0114] Table 1 Example Test Location - Subsystem of Cangyu Expressway Fuping East to Baoding West Section

[0115]

[0116] Table 2 Basic characteristics of each section of subsystem 4

[0117]

[0118] Table 3 Basic characteristics of each section of subsystem 5

[0119]

[0120]

[0121] A CAM model was established for each road section of each subsystem. The model was then calibrated and its parameters modified based on various statistical data and relevant standards. Simulation experiments were conducted using the modified CAM model to establish technical standards for fog control for each road section. Each experiment consisted of 9,000 simulation steps with a 1-second step length. To reduce initial vehicle state interference and eliminate the influence of randomness on the experiment, the results were averaged over 40 experiments with a final simulation step length of 7,200 (under the same experimental conditions).

[0122] Table 4 shows the technical standards for fog management for Section 4 in Subsystem 4, when the permeability is 50%. Other sections are similar. As can be seen, as visibility decreases, the upper speed limit for HVs decreases significantly. While CAVs experience degradation due to the impact of HVs, the upper speed limit decreases relatively little. The maximum permissible flow rate on this section also decreases with decreasing visibility. In extreme cases (visibility below 30 meters), HVs are not permitted on this section. However, CAVs are not affected by HVs, and their upper speed limit returns to normal. The maximum permissible flow rate on the section reaches 2850 veh / h, at which point the permeability reaches 100%, allowing more CAVs to enter.

[0123] Table 4 Technical standards for fog control on road section 4 in subsystem 4 (penetration rate 50%)

[0124]

[0125] An online implementation example was selected, using a day with low visibility but not yet reaching the point of requiring a road closure for HVs. The fog lasted for seven hours, from 5:00 AM to 12:00 AM. Visibility varied across subsystems and subsystem sections, with visibility ranges of 150m-110m, 110m-90m, 90m-80m, 80m-50m, 50m-30m, 50m-30m, and 50m-30m, respectively. Based on the method presented in this invention, visibility, traffic flow status, and penetration rate were collected for each section to determine the maximum allowable flow rate and upper speed limits for CAVs and HVs. The maximum allowable flow rate and upper speed limits for CAVs and HVs for each section of Subsystem 4 are shown in Table 5 (assuming a penetration rate of 60%). Similar values ​​were used for the other subsystems. The method presented in this invention determined the control variables for the maximum allowable flow rate, ramp capacity, and other highway capacity for each subsystem, as shown in Table 6. The calculated control effects are shown in Table 7. The table lists the speed limits under normal weather conditions that are not controlled, as compared with the existing methods. For example, the speed limits of subsystem 4 section 4 are 100 km / h and 70 km / h for large vehicles and small vehicles respectively; Feng fusion control, see Feng Fengjiang et al. (Feng Fengjiang, Cheng Xinping, Liu Huiyang, et al. Coordinated control method for foggy highways based on cellular automaton model: China, 112396834[P]. 2021-09-07.), all controlled in accordance with HVs standards; Perraki fusion control, see Perraki et al. (Perraki G, Roncoli C, Papamichail I, et al. Evaluation of a model predictive control framework for motorway traffic involving conventional and automated vehicles[J]. Transportation Research Part C, 2018, 92(8): 456-471.) predictive control method; Gong mainline predictive control, see Gong et al. (Gong BW, Wei RX, Wu DY, et al. Fleet Management for HDVsand CAVs on Highway in Dense Fog Environment[J].Journal of AdvancedTransportation,2020,2020(2):1-21.), is limited to mainline speed limits and fleet management.

[0126] Table 5. Speed ​​limits and maximum allowable flow rates for CAVs and HVs on each road section of subsystem 4 of the online implementation case

[0127]

[0128] Table 6 Online implementation case subsystem coordinated flow control values ​​(veh / h)

[0129]

[0130]

[0131] Table 7 Comparison of the effects of various control methods in online implementation cases

[0132]

[0133] From Table 7 we can see that:

[0134] (1) If not controlled, the probability of accidents is very high. At the same time, due to the congestion caused by accidents, the average and maximum lengths of ramp queues are very large, and the traffic volume of the highway will also drop significantly.

[0135] (2) The Feng Fusion control method significantly reduces the probability of accidents and significantly reduces the traffic volume on the highway. However, due to the traditional control method, it does not take advantage of the fact that CAVs are relatively less affected by fog, which prevents the traffic volume from being further improved. As a result, the average and maximum ramp queue lengths remain relatively large.

[0136] (3) Although all indicators have been improved by the Perraki fusion control method, the indicators have not yet achieved the optimal effect due to the problems that "the different road geometric shapes, visibility, and permeability of multiple sections are not involved, the micro-traffic flow behavior of specific sections cannot be mapped, and predictive control needs to be based on accurate prediction."

[0137] (4) The Gong mainline predictive control method is used. Although all indicators are better than those without control, the flow rate and accident rate are not as good as the Perraki fusion control and Feng speed limit and flow coordination fusion control methods. This is because the control is limited to the mainline control and does not fully utilize the advantages of ramp control to improve road flow and reduce accident rates. The average and maximum values ​​of the ramp queue length are low, which is caused by the lack of ramp control.

[0138] (5) Compared with the uncontrolled method, Feng fusion control method, Perraki fusion control method, and Gong mainline predictive control method, the accident probability of the method of the present invention is reduced by 0.74‰, 0.14‰, 0.12‰, and 0.23‰, respectively. The flow rate is increased by 353 veh / h, 187 veh / h, 149 veh / h, and 262 veh / h, respectively. The average queue length of the ramp is reduced by 63 veh, 41 veh, 27 veh, and 15 veh, respectively. The maximum queue length of the ramp is reduced by 108 veh, 56 veh, 45 veh, and 24 veh, respectively. The control effect is obvious.

[0139] Any matters not described in the present invention are applicable to the prior art.

Claims

1. A method for controlling the fusion of intelligent network mixed flows on foggy highways, characterized in that: The following steps are involved: Step 1: Establish a basic highway information database; Step 2: Establish a fog control sample library; Step 3: Divide the highway into several subsystems, and each subsystem into multiple sections. Build a fog management knowledge base for each section, including upper speed limits for HVs and CAVs, and maximum allowable flow rates under different visibility and penetration rates. Step 4: Forecast visibility in foggy days on highways over the next few days, formulate emergency control plans, and release traffic guidance information in advance; Step 5: In foggy conditions on highways, determine the number of HVs and CAVs allowed on each on-ramp and the upper speed limits for HVs and CAVs on each road section, and implement online real-time control. ① Collect CAVs and HVs demand, traffic volume, and visibility in real time, and refine road sections based on visibility; ② Let the subsystems affected by fog be i = i1, i1 + 1, ..., n, where i1 is the first subsystem affected by fog and n is the most downstream subsystem affected by fog. Let i = n. For subsystem i, if the upstream contains an interchange with other highways, turn to step 9; otherwise, turn to step 3. ③Calculate the upstream permeability P of the main line A1 , entrance ramp permeability P A2 , comprehensive permeability of the main line upstream and entrance ramp P A3 ; ④Based on P A3 and visibility, query the fog control knowledge base of each road section of subsystem i, and obtain the maximum allowable flow of subsystem i in, represents the maximum allowable flow rate of the jth section of subsystem i, n i represents the number of road sections contained in subsystem i, Q Di+1 Indicates that subsystem i+1 needs to reduce flow; ⑤If Turn to ⑥; otherwise, let r i =r i C +r i H , Q Di =0, according to P A3 and visibility, query the fog control knowledge base of each road section of subsystem i, determine the upper speed limit of HVs and CAVs of each road section of subsystem i, and turn to ⑩; where q ui represents the total upstream demand of the main line, r di represents the total demand for the entrance ramp, r i 、r i C 、r i H represents the total number of vehicles released at the entrance ramp of subsystem i, and the number of CAVs and HVs released, represents the demand for CAVs and HVs on the entrance ramp, Q Di Indicates that subsystem i needs to reduce flow; ⑥ If Turn to ⑧; if and Turn to ⑦; otherwise, let Q Di =0, according to P A3 and visibility, query the fog control knowledge base of each road section of subsystem i, determine the upper speed limit of HVs and CAVs of each road section of subsystem i, and turn to ⑩; ⑦ Determine the number of CAVs and HVs allowed on the entrance ramp when the demand for HVs on the entrance ramp is low, as well as the upper speed limits for HVs and CAVs on each road section; a. Let r i C = r i H · P A2 / (1 - P A2 ), r i = r i C + r i H ,Q i = q ui + r i ; b. If Turn to c, otherwise turn to f; c. Let r i C =r i C +1, updated r i and Q i ; d. Order e. According to P A4 and visibility, query the fog control knowledge base of each road section of subsystem i, and update Turn to b until Turn to f; f. Let r i =r i C +r i H , Q i =q ui +r i , Q Di =0, according to P A4 and visibility, query the fog control knowledge base of each road section of subsystem i, determine the upper speed limit of HVs and CAVs of each road section of subsystem i, and turn to ⑩; ⑧Confirm The number of CAVs released at that time and the upper speed limit of HVs and CAVs on each road section; a. Let r i C =0, r i H = 0, initialize the set R of CAVs cases that can be placed in the entrance ramp to an empty set; b. Let r i C =r i C +1, r i =r i C , Q i =q ui +r i ,calculate According to P A5 and visibility, query the fog control knowledge base of each road section of subsystem i, and determine c. If Add the case to set R, otherwise do not add it and go to d; d. If Turn to b; otherwise, turn to e; e. If the set R is an empty set, go to g; otherwise, go to f; f. The corresponding maximum value in set R In the case of i C ; Let Q Di =0, according to P A5 and visibility, query the fog control knowledge base of each road section of subsystem i, determine the upper speed limit of HVs and CAVs of each road section of subsystem i, and turn to ⑩; g. Let r i =r i C =r i H =0, according to P A1 and visibility, query the fog control knowledge base of each road section of subsystem i, and determine the upper speed limit of HVs and CAVs on each road section of subsystem i; h. If i≠i1, let Turn to ⑩; otherwise, let Turn to ⑩; where s is the number of closed entrance ramps upstream of subsystem i, is the flow rate of the exit ramp of subsystem i1-k, is the total upstream demand and maximum allowable flow of the main line of subsystem i1, is the total number of vehicles released from the entrance ramps of subsystems i1-1, i1-2, i1-s, and i1-s-1; ⑨ For subsystem i, if the upstream contains an "interchange bridge that intersects with other highways", then: a. Calculate the comprehensive penetration rate after the main line upstream vehicles intersect with other high-speed vehicles According to P A6 and visibility, query the fog control knowledge base of each road section of subsystem i, and obtain the maximum allowable flow of subsystem i in, and are the demands of CAVs and HVs of other high-speed merging subsystem i, respectively; b. If Let Q Di =0, according to P A6 and visibility, query the fog control knowledge base of each road section of subsystem i, determine the upper limit of the speed limit of HVs and CAVs on each road section, and turn to ⑩; otherwise, turn to c; where q 2ui represents the total demand of CAVs and HVs from other high-speed merging subsystem i; c. Computing subsystem i needs to reduce traffic According to P A6 and visibility, query the fog control knowledge base of each road section of subsystem i, determine the upper speed limit of HVs and CAVs, and turn to ⑩; where β i is the ratio of highway vehicles to upstream vehicles of subsystem i; ⑩If i=i1, go to step ①, otherwise let i=i-1 and go to step ②; Step 6: Obtain the newly added fog control samples, return to steps 2 and 3, and modify the fog control sample library and fog control knowledge base.

2. The method for intelligent network mixed flow fusion control on foggy highways according to claim 1 is characterized in that: In the third step, the cellular automaton model of each road section is established using the fog control sample library, and the parameters of the cellular automaton model are modified to obtain the modified cellular automaton model; the modified cellular automaton model is used to simulate and determine the maximum allowable flow rate, HVs and CAVs speed limit upper limits under different visibility and permeability.

Images