A method for calculating the random slowing probability of tunnel groups in an intelligent connected environment

Abstract

This invention proposes a method for calculating the random slowdown probability of tunnel groups in an intelligent connected environment. Considering comprehensive factors such as tunnel ventilation, lighting, safety, and driver visual characteristics, as well as traffic flow characteristics of each section of the tunnel group, the method rationally divides the tunnel group into sections and determines the initial value of the random slowdown probability for each section. A cellular automaton simulation model of the tunnel group sections is constructed. By inputting survey data such as vehicle speed and density of traffic flow in each section of the tunnel group, and through an adaptive genetic algorithm, the optimized random slowdown probability values ​​for each section can be obtained, thereby improving the accuracy of heterogeneous mixed traffic flow simulation in tunnel group sections.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation technology, specifically to traffic flow simulation of tunnel group road sections under intelligent connected environment, and more particularly to a method for calculating the probability of random slowdown of vehicles in the simulation of heterogeneous mixed traffic flow composed of CAV vehicles and HV vehicles in different areas of tunnel group road sections under intelligent connected environment. Background Technology

[0002] Vehicles on traditional roads exhibit random slowdown during travel. To realistically reproduce the changing patterns of traffic flow, cellular automata models are widely used for simulating microscopic traffic flow, yielding numerous research results. However, under current simulation conditions, most studies express the random slowdown phenomenon of vehicles using random slowdown probabilities and assign them a single, fixed value, which affects the accuracy of the simulation results.

[0003] In intelligent connected vehicle environments, traffic flow will long consist of a heterogeneous mix of connected and automated vehicles (CAVs) and human-driving vehicles (HVs). Due to the complexity of traffic flow in tunnel complexes under intelligent connected vehicle environments, with vehicles entering and exiting tunnels multiple times, accurately simulating the random slowdown probability of heterogeneous mixed traffic flow in different areas of tunnel complexes and precisely revealing the evolution patterns of traffic flow in tunnel complexes under intelligent connected vehicle environments is of significant theoretical research value and practical application significance. Summary of the Invention

[0004] To accurately reveal the evolution patterns of heterogeneous mixed traffic flow in different areas of tunnel groups under intelligent connected vehicle environments, this invention proposes a method for calculating the random slowdown probability in the simulation of heterogeneous mixed traffic flow in tunnel groups under intelligent connected vehicle environments. Considering comprehensive factors such as tunnel ventilation, lighting, safety, and driver visual characteristics, as well as the traffic flow characteristics of each section of the tunnel group, the tunnel group is rationally divided into zones, and initial values ​​for the random slowdown probability of each zone are determined. A cellular automata simulation model of the tunnel group is constructed, and survey data such as vehicle speed and density of traffic flow in each section of the tunnel group are input. Through an adaptive genetic algorithm, optimized random slowdown probability values ​​for each section can be obtained, thereby improving the accuracy of the simulation of heterogeneous mixed traffic flow in tunnel groups.

[0005] The specific technical solution adopted is as follows:

[0006] A method for calculating the random slowdown probability of tunnel groups in an intelligent connected environment considers comprehensive factors including tunnel ventilation, lighting, safety, and driver visual characteristics, as well as traffic flow characteristics of each section of the tunnel group. The tunnel group is divided into sections, and the initial value of the random slowdown probability of each section is determined. Then, a cellular automaton simulation model of the tunnel group is constructed. Data including traffic flow speed and density of each section of the tunnel group are input, and an adaptive genetic algorithm is used to obtain the optimized random slowdown probability values ​​of each section.

[0007] Furthermore, the specific steps include:

[0008] Step 1: Considering comprehensive factors including tunnel ventilation, lighting, safety, and driver visual characteristics, as well as the traffic flow characteristics of each section of the tunnel group, the tunnel group road sections are divided into basic road sections before the tunnel entrance. The affected section at the tunnel entrance Tunnel entrance transition section Tunnel basic section Tunnel exit transition section Spacing between two adjacent tunnels Tunnel Exit Affected Section Basic road section after tunnel exit n is the number of tunnels in the tunnel group;

[0009] Step 2: Define the initial value of P as the random slowdown probability of traditional vehicles in each section of the tunnel group. i n , i = 1, 2, ... 6, where: P1 1 For the random slowdown probability of the basic road section before the tunnel group, For the random slowdown probability of the affected section before the tunnel entrance, P3 n P4 represents the random slowing probability of the transition section at the nth tunnel entrance. n For the random slowdown probability of the nth basic tunnel segment, P5 n For the random slowdown probability of the nth tunnel exit transition section, P6 n-1 The probability of random slowing down the tunnel spacing before the nth tunnel. For the random slowdown probability of the affected section at the tunnel exit, Let be the probability of random slowdown on the basic road section after the tunnel exit; define the probability of random slowdown for intelligent connected vehicles in each section of the tunnel group as . d is the distance between the front ends of the vehicles, L is a constant, and a is a constant between 7 and 9;

[0010] Step 3: Collect traffic flow density and vehicle speed parameters for each section under different tunnel group combinations, select classic tunnel group road sections to construct cellular automata simulation models, and conduct traffic flow simulations for each scenario to analyze the traffic flow variation patterns in each section of the tunnel group.

[0011] Step 4: Compare and analyze the field survey data and simulation analysis data, and calculate the fitness function E(x) in the adaptive genetic algorithm. k i To conduct a field investigation of the density of the i-th segment of the tunnel group, v represents the density of the i-th segment obtained through the tunnel swarm cellular automata model; i Let i be the average speed of the i-th segment in the field survey. The velocity of the i-th segment is obtained through the tunnel swarm cellular automata model;

[0012] Step 5: Use a genetic algorithm to perform crossover, mutation, and population update on the traditional vehicle random slowdown probability. The random slowdown probability value obtained when the fitness function reaches its minimum value is the optimal solution.

[0013] Furthermore, it is stipulated that vehicles are not allowed to change lanes inside the tunnels of the tunnel section.

[0014] Furthermore, in Step 1, the length of the tunnel group section is not fixed, and the length of each tunnel group section is different depending on the characteristics of the different tunnel groups.

[0015] Furthermore, the traffic flow consists of: traditional manual vehicles (HV), adaptive cruise control vehicles (ACC), and coordinated adaptive cruise control vehicles (CACC) combined in adjustable proportions; the adaptive genetic algorithm calculates the random slowdown probability of traffic flow in tunnel sections for HV vehicles, while the random slowdown probability of CAV vehicles is obtained based on the headway of the preceding vehicle; after the tunnel group sections are divided into regions, the random slowdown probability within each section is only related to the heterogeneous mixed traffic flow characteristics of that section.

[0016] Furthermore, in Step 2, the random slowdown probability of intelligent connected vehicles, including Adaptive Cruise Control (ACC) and Coordinated Adaptive Cruise Control (CACC), is defined to be independent of and does not interfere with the random slowdown probability of traditional driving vehicles; the random slowdown probability of intelligent connected vehicles is adjusted and determined based on the distance to the vehicle in front.

[0017] Furthermore, in Step 3, vehicle location speed and vehicle density are statistically analyzed every 20m; tunnel characteristics of the tunnel section are collected, including: tunnel length, tunnel speed limit, tunnel clearance height and number of lanes; simulation scenarios are established based on the tunnel length, tunnel combination and tunnel speed limit obtained from the survey; tunnel partitions, vehicle location speed and vehicle density of each scenario are obtained from the survey and substituted into the model to establish a cellular automata simulation model.

[0018] Furthermore, in Step 4, the survey data and simulation analysis data are compared and analyzed to calculate the fitness value of each vehicle; the density of each vehicle is defined as the distance between the current vehicle and the vehicle in front.

[0019] Furthermore, the specific process of the adaptive genetic algorithm in Step 5 is as follows:

[0020] (1) After evaluating the random slowing probability dataset X(i) of the initial tunnel swarm model, an elite individual retention selection method is used to select some individuals for the next step, resulting in a new population XS(i).

[0021] (2) Crossover: A single-point crossover is performed between two randomly selected individuals, giving a new set XC(i), and the crossover probability (pc) is obtained from the evaluated individuals as shown in the following formula;

[0022]

[0023] fmax refers to the maximum fitness value in the population, favg refers to the average fitness value in the population, and f′ refers to the higher fitness value among the two individuals to be crossovered.

[0024] (3) Mutation: Apply the mutation operator to all chromosomes from XC(i) and obtain the probability (pm) from the evaluated individuals. The value of pm is shown in the following formula, thereby obtaining a brand new population XM(i) of the same size.

[0025]

[0026] (4) Population renewal: Finally, a new population X(i+1) is selected from the original population X(i) and XM(i) populations using the elite individual retention method.

[0027] Repeat the above steps multiple times. The random slowing probability of each segment when the fitness of the new population is at its minimum is the random slowing probability of each segment of the tunnel group.

[0028] Compared with existing technologies, the present invention and its preferred embodiments can reduce the inaccuracy of simulation results caused by a single random slowing probability determination value, making the simulation results closer to reality. Attached Figure Description

[0029] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0030] Figure 1 This is a schematic diagram of the calculation process for the random slowdown probability of highway tunnel group sections according to an embodiment of the present invention;

[0031] Figure 2 This is a schematic diagram of highway tunnel zoning according to an embodiment of the present invention;

[0032] Figure 3 This is a schematic diagram of a tunnel group section in an embodiment of the present invention, which consists of two adjacent tunnels. Detailed Implementation

[0033] To make the features and advantages of this patent more apparent and understandable, specific embodiments are provided below, along with accompanying drawings, for detailed explanation:

[0034] according to Figure 3 As shown, when considering a tunnel group segment consisting of two adjacent tunnels, this embodiment of the invention proposes a method for calculating the random slowdown probability in tunnel group traffic flow simulation under an intelligent connected environment, such as... Figure 1 As shown, the steps are as follows:

[0035] Step 1: Considering comprehensive factors such as tunnel ventilation, lighting, safety, and driver visual characteristics, as well as the heterogeneous and mixed traffic flow characteristics of each section of the tunnel group, the tunnel group road sections are divided into basic road sections before the tunnel entrance. The affected section at the tunnel entrance Tunnel entrance transition section Tunnel basic section Tunnel exit transition section Spacing between two adjacent tunnels Tunnel Exit Affected Section Basic road section after tunnel exit n is the number of tunnels in the tunnel group, and its basic structure is as follows: Figure 2 As shown.

[0036] Step 2: Define the initial value of P as the random slowdown probability of traditional vehicles in each section of the tunnel group. i n (i = 1, 2, ..., 6, n = 1, 2, ..., n) (initial values ​​are random values ​​between 0 and 1), where: P1 1 For the random slowdown probability of the basic road section before the tunnel group, For the random slowdown probability of the affected section before the tunnel entrance, P3 n P4 represents the random slowing probability of the transition section at the nth tunnel entrance. n For the random slowdown probability of the nth basic tunnel segment, P5 n For the random slowdown probability of the nth tunnel exit transition section, P6 n-1 The probability of random slowing down the tunnel spacing before the nth tunnel. For the random slowdown probability of the affected section at the tunnel exit, Let be the probability of random slowdown on the basic road section after the tunnel exit; define the probability of random slowdown for intelligent connected vehicles in each section of the tunnel group as . d is the distance between the front of the vehicle, L is a constant, and a is a constant between 7 and 9.

[0037] Step 3: Collect parameters such as traffic flow density and vehicle speed in various sections under different tunnel group combinations, including extra-long tunnels, long tunnels, and medium-short tunnels. Select classic tunnel group road sections to construct cellular automata simulation models, and conduct traffic flow simulations for each scenario to analyze the traffic flow change patterns in each section of the tunnel group.

[0038] Step 4: Compare and analyze the field survey data and simulation analysis data, and calculate the fitness function E(x) in the adaptive genetic algorithm. k i It is a field survey of the density of the i-th vehicle in the tunnel group. The density of the i-th vehicle is obtained through a tunnel swarm cellular automata model. i This refers to the average speed of the i-th vehicle in the field survey. The speed of the i-th vehicle is obtained through the tunnel swarm cellular automata model.

[0039] Step 5: Use a genetic algorithm to perform crossover, mutation, and population update on the traditional vehicle random slowdown probability to obtain the fitness value. After multiple iterations, the random slowdown probability value obtained when the fitness function reaches its minimum value is the optimal solution.

[0040] In the specific test example of this embodiment, according to the first step above, the specific values ​​for each segment are as follows: This is the basic section of the highway before the tunnel entrance. This is the basic section of the highway after the tunnel exit, with a length of 300m.

[0041] The section affected by the tunnel entrance For the affected section at the tunnel exit, the reference identification sight distance values ​​are shown in the table below:

[0042] Values ​​of the affected sections at tunnel entrances and exits

[0043] 120 460 100 380 80 300 60 240

[0044] This is the tunnel entrance transition section, tunnel entrance transition section D. th The calculation formula is as follows:

[0045]

[0046] In the formula: D th D is the length of the inlet segment. s is the sight distance for parking with lighting; h is the clearance height inside the tunnel; V is the speed limit in the tunnel.

[0047] This is the basic section of the tunnel, and its value is the total length of the tunnel minus the lengths of the entrance transition section and the exit section.

[0048] This is the transition section at the tunnel exit. The length of the transition section at the exit of a typical highway tunnel is 60m.

[0049] Spacing between two adjacent tunnels The maximum spacing varies depending on the actual value and the speed limit, as shown in the table below.

[0050] Maximum spacing between tunnel groups

[0051] 120 400 100 350 80 300

[0052] Based on the second step above, define P1 1 For the random slowdown probability of the basic road section before the tunnel group, For the random slowdown probability of the affected section before the tunnel entrance, P3 1 and P3 2 For the random slowing probability of the tunnel entrance transition section, P4 1 and P4 2 For the random slowdown probability of the basic tunnel segment, P5 1 and P5 2 For the random slowdown probability of the gradual transition section at the tunnel exit, P6 1 For the random slowdown probability of tunnel spacing, For the random slowdown probability of the affected section at the tunnel exit, This represents the probability of random slowdown of the basic road segment after the tunnel exit.

[0053] Based on step three above, traffic parameter monitoring devices should be installed every 20 meters along tunnel sections to obtain location speed and density data. Investigate tunnel characteristic data for each tunnel type: tunnel length, tunnel speed limit, tunnel combination, internal clearance height, number of lanes, and tunnel spacing, etc. Establish a preliminary CA model based on the survey data.

[0054] Based on the fourth step above, integrate the location speed data and density data of each vehicle obtained from the survey, and compare them with the location data and density data of each vehicle obtained from the simulation, and calculate the sum of the fitness functions of all vehicles.

[0055] Based on step 5 above, it is necessary to specify the population value and the number of iterations. The population XS(i) and the new population X(i+1) are both set to 50. The probabilities of crossover and mutation vary according to fitness. The smaller the fitness, the more necessary it is to retain the random slowing probability value, and the lower the probability of crossover and mutation. Repeat the genetic process 12,500 times, and select the random slowing probability of the minimum fitness value as the random slowing probability of each segment.

[0056] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0057] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0058] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0059] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0060] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

[0062] This patent is not limited to the above-described preferred embodiment. Anyone can derive other forms of methods for calculating the random slowing probability of tunnel groups in an intelligent connected environment based on the inspiration of this patent. All equivalent changes and modifications made within the scope of this patent application shall fall within the scope of this patent.

Claims

1. A method for calculating the random slowdown probability of tunnel groups in an intelligent connected environment, characterized in that: Taking into account comprehensive factors including tunnel ventilation, lighting, safety, and driver visual characteristics, as well as traffic flow characteristics of each section of the tunnel group, the tunnel group is divided into sections and the initial value of random slowdown probability is determined for each section. Furthermore, a cellular automata simulation model of the tunnel group road segments is constructed. Data on traffic flow speed and density of each road segment of the tunnel group are input, and an adaptive genetic algorithm is used to obtain the optimized random slowdown probability values ​​for each segment. Specifically, the following steps are included: Step 1: Considering comprehensive factors including tunnel ventilation, lighting, safety, and driver visual characteristics, as well as the traffic flow characteristics of each section of the tunnel group, the tunnel group road sections are divided into basic road sections before the tunnel entrance. Section affected by tunnel entrance Tunnel entrance transition section Tunnel basic section Tunnel exit transition section The distance between two adjacent tunnels Section affected by tunnel exit Basic road section after tunnel exit , The number of tunnels in the tunnel group; Step 2: Define the initial value of the random slowdown probability of traditional vehicles in each section of the tunnel group as follows: , ,in: For the random slowdown probability of the basic road section before the tunnel group, For the random slowdown probability of the affected section before the tunnel entrance, Let m be the random slowdown probability of the transition section at the tunnel entrance. Let m be the random slowdown probability of the m-th basic tunnel segment. Let m be the probability of random slowing down of the gradual transition section at the tunnel exit. The probability of random slowing down the tunnel spacing before the m-th tunnel. For the random slowdown probability of the affected section at the tunnel exit, Let be the probability of random slowdown on the basic road section after the tunnel exit; define the probability of random slowdown for intelligent connected vehicles in each section of the tunnel group as . , For the distance between the front of the car, It is a constant. A constant between 7 and 9; Step 3: Collect traffic flow density and vehicle speed parameters for each section under different tunnel group combinations, select classic tunnel group road sections to construct cellular automata simulation models, and conduct traffic flow simulations for each scenario to analyze the traffic flow variation patterns in each section of the tunnel group. Step 4: Compare and analyze the field survey data and simulation analysis data, and calculate the fitness function in the adaptive genetic algorithm. , , For the first time to conduct on-site investigation of the tunnel group Segment density, The first cellular automaton model obtained by tunneling swarm is... Density of the section; The first field investigation Average speed of the section The first cellular automaton model obtained by tunneling swarm is... The speed of the section; Step 5: Use a genetic algorithm to perform crossover, mutation, and population update on the traditional vehicle random slowdown probability. When the fitness function reaches its minimum value, the obtained random slowdown probability value is the optimal solution. Traffic flow consists of: traditional manual vehicles (HV), adaptive cruise control vehicles (ACC), and coordinated adaptive cruise control vehicles (CACC) combined in adjustable proportions; the adaptive genetic algorithm calculates the random slowdown probability of traffic flow in tunnel sections, applicable to HV vehicles, while the random slowdown probability of CAV vehicles is obtained based on the headway of the preceding vehicle; after the tunnel group sections are divided into regions, the random slowdown probability within each section is only related to the heterogeneous mixed traffic flow characteristics of that section; In Step 2, the random slowdown probability of intelligent connected vehicles, including Adaptive Cruise Control (ACC) and Coordinated Adaptive Cruise Control (CACC), is defined to be independent of the random slowdown probability of traditional driving vehicles and does not interfere with each other; the random slowdown probability of intelligent connected vehicles is adjusted and determined based on the distance to the vehicle in front.

2. The method for calculating the random slowdown probability of tunnel groups in an intelligent connected environment according to claim 1, characterized in that: Vehicles are prohibited from changing lanes inside tunnels along the designated tunnel section of the road.

3. The method for calculating the random slowdown probability of tunnel groups in an intelligent connected environment according to claim 1, characterized in that: In Step 1, the length of the tunnel group section division is not fixed, and each tunnel group is divided according to its different characteristics. The partition lengths can be different.

4. The method for calculating the random slowdown probability of tunnel groups in an intelligent connected environment according to claim 1, characterized in that: In Step 3, the vehicle speed and vehicle density are recorded at each 20m interval; The tunnel characteristics of the tunnel section are collected, including: tunnel length, tunnel speed limit, tunnel clearance height and number of lanes; the simulation scene is established based on the tunnel length, tunnel combination and tunnel speed limit obtained from the survey; the tunnel zoning, vehicle location speed and vehicle density of each scene are obtained from the survey and substituted into the model to establish the cellular automata simulation model.

5. The method for calculating the random slowdown probability of tunnel groups in an intelligent connected environment according to claim 1, characterized in that: In Step 4, the survey data and simulation analysis data are compared and analyzed to calculate the fitness value of each vehicle; the density of each vehicle is defined as the distance between the current vehicle and the vehicle in front.

6. The method for calculating the random slowdown probability of tunnel groups in an intelligent connected environment according to claim 1, characterized in that: The specific process of the adaptive genetic algorithm in Step 5 is as follows: (1) On the random slowing probability dataset of the preliminary tunnel group model After evaluation, an elite individual retention selection method is used to select some individuals for the next step, resulting in a new population. ; (2) Crossover: A single-point crossover is performed between two randomly selected individuals to give a new set. And obtain the crossover probability from the evaluated individuals. As shown in the following formula; It refers to the maximum fitness value in the population. This refers to the average fitness of the population; This refers to the higher fitness value among the two individuals to be crossed. (3) Mutation: for those from Mutation operators are applied to all chromosomes, and probabilities are obtained from the evaluated individuals. , The values ​​are shown in the following formula, thus obtaining a completely new population of the same size. ; (4) Population renewal: Finally, the elite individual preservation method is used to renew the population from the original population. and Select a new population from the existing population. ; Repeat the above steps multiple times. The random slowing probability of each segment when the fitness of the new population is at its minimum is the random slowing probability of each segment of the tunnel group.

Images