Ramp metering mixed traffic simulation method and signal plan evaluation method

Abstract

The application discloses an entrance ramp mixed traffic simulation method and a signal scheme evaluation method, which comprises the following steps: establishing an entrance ramp traffic flow simulation based on a cellular automaton model; adding competition and cooperation strategies to the model; adding observation points to investigate the degree of vehicle fleet dispersion phenomenon; setting road condition simulation parameters and signal lamp initial conditions; simulating traffic flow driving process under a given traffic light control scheme; analyzing traffic flow state and the influence of large vehicles on ramp traffic capacity, and generating charts; and iteratively evaluating traffic light control schemes multiple times and performing scoring and sorting. The application studies lane changing and following rules of mixed traffic flow of autonomous vehicles and manually driven vehicles at an entrance ramp bottleneck, simulates possible traffic phenomena and corresponding traffic flow parameters, tests comprehensive performance of a signal lamp scheme under specific road conditions, and obtains real-time evaluation results of the signal lamp scheme through a scoring system.

Description

Technical Field

[0001] This invention relates to the field of traffic control technology, and more specifically, to a method for simulating mixed traffic at entrance ramps and a method for evaluating signal schemes. Background Technology

[0002] Autonomous vehicles will become a core component of future smart city transportation, forming a complex transportation network alongside human-driven vehicles. To address this challenge, research needs to focus on mixed traffic simulation at entrance ramps and signaling scheme evaluation. Currently, the following two key technologies are primarily employed:

[0003] First, it is necessary to study heterogeneous traffic flow models for autonomous vehicles and manually driven vehicles. Since both autonomous and manually driven vehicles will jointly constitute the traffic flow, studying this mixed traffic flow is crucial. Due to limited measured data, microscopic simulations are needed to study the passage status of autonomous vehicles at ramp bottlenecks.

[0004] Secondly, it is necessary to establish lane-changing and following rules for autonomous vehicles on ramps that can simulate real-world road scenarios. Real-world ramp bottlenecks involve complex scenarios such as lane changing and following. By studying the characteristics of mixed traffic flow at ramp bottlenecks, it is necessary to analyze vehicle density ranges, ramp vehicle ratios, and the impact of ramp facilities on urban road traffic efficiency.

[0005] In addition, innovative hybrid traffic simulation methods are needed to combine the characteristics of autonomous vehicles and human-driven vehicles to simulate ramp traffic in real-world scenarios. In particular, the relationship between the proportion of large vehicles and ramp capacity should be considered, and the combination of traffic light durations should be evaluated to calculate the optimal signal scheme.

[0006] Existing research has not adequately considered the above-mentioned technical requirements. Summary of the Invention

[0007] To address these issues, this invention proposes a method for simulating mixed traffic at entrance ramps and a method for evaluating signal schemes.

[0008] According to one aspect of the present invention, a method for simulating mixed traffic at entrance ramps is proposed, comprising: S1, establishing a traffic flow simulation of entrance ramps based on a cellular automata model, including: discretizing the driving states of autonomous and manually driven vehicles on entrance ramps and main roads by constructing a cellular automata model to simulate vehicle driving dynamics under various traffic conditions; S2, setting simulation parameters and initial conditions, including: setting road condition simulation parameters and initial conditions for traffic flow according to actual road conditions and traffic demand; S3, simulating the traffic flow process, including: simulating the driving process of vehicles on entrance ramps and main roads using a cellular automata model, considering the interaction between vehicles to reflect the actual traffic flow situation; S4, analyzing traffic flow characteristics, including: analyzing the driving characteristics of traffic flow on entrance ramps and main roads based on simulation results, as well as the impact of the proportion of large vehicles on traffic flow, and drawing real-time visualization charts to show the driving state of traffic flow under different conditions.

[0009] According to another aspect of the present invention, an evaluation method for entrance ramp signal schemes is proposed, comprising: first, defining an evaluation index to measure the performance of different combinations of signal light durations; this index consists of two parameters with adjustable weights, the parameters being vehicle throughput efficiency and ramp area queue length, denoted by E; determining the search range for signal light durations, and then traversing all possible combinations of green light duration G and red light duration R; for each combination of signal light durations (G, R), performing the following operations: performing simulation with the given durations (G, R) and calculating the evaluation index E; during the traversal, recording the signal light duration combination with the highest score (Gbest, Rbest); each time the evaluation index E is calculated, checking whether the score of the current combination (G, R) is higher than the score of the current best combination (Gbest, Rbest); if the score of the current combination is higher, then updating the best combination to the current combination; after the traversal, the best signal light duration combination (Gbest, Rbest) is the duration combination with the highest score.

[0010] This invention studies lane-changing and following rules for mixed traffic flow of autonomous and manually driven vehicles at entrance ramp bottlenecks. It simulates potential traffic phenomena and corresponding traffic flow parameters, and tests the comprehensive performance of traffic light schemes under specific road conditions. A scoring system is used to obtain real-time evaluation results of the traffic light schemes. This invention can simulate the driving states of autonomous and manually driven vehicles on entrance ramps and main roads, discretizing the characteristics of the two driving modes to more accurately reflect real-world traffic conditions. This invention simulates the competitive and cooperative behaviors of vehicles on main roads in the entrance ramp area, such as accelerating to overtake and decelerating to yield, to more realistically simulate the behavioral differences between different drivers. Through multiple simulation iterations, this invention traverses various traffic light control schemes and uses adjustable weights for multi-factor comprehensive scoring and ranking, thereby providing the optimal traffic light control scheme for specific traffic flow conditions. Attached Figure Description

[0011] Figure 1 This is a flowchart of the method for simulating mixed traffic and evaluating signal schemes at entrance ramps proposed in an embodiment of the present invention. Detailed Implementation

[0012] The present invention will be further described below with reference to specific embodiments.

[0013] This invention proposes a method for simulating mixed traffic and evaluating signal schemes at entrance ramps. The process of this method is as follows: Figure 1 As shown, the process includes: establishing an entrance ramp traffic flow simulation based on a cellular automata model; incorporating competition and cooperation strategies, as well as observation points for vehicle dispersion; setting road condition simulation parameters and initial traffic light conditions to simulate the traffic flow process; analyzing traffic flow status and the impact of large vehicles on ramp capacity, generating real-time charts; iterating the traffic light control scheme multiple times and scoring and ranking the traffic lights. In this embodiment, "mixed traffic" does not specifically refer to a mixed traffic flow containing both manually driven and autonomous vehicles, but also includes traffic flows containing only manually driven or only autonomous vehicles, i.e., where one vehicle type accounts for 0% and the other for 1%. In some embodiments, the ramp cellular automata model for mixed traffic flow can be established based on the Nagel-Schreckenberg model. Of course, those skilled in the art will understand that if other models are used, the same model can be established according to the technical ideas in this embodiment. This embodiment only uses this model as an example to elaborate on the specific implementation principles of this embodiment.

[0014] The following two examples illustrate the implementation process of the entrance ramp mixed traffic simulation method and the entrance ramp signal scheme evaluation method.

[0015] Example 1

[0016] This embodiment proposes a method for simulating mixed traffic at entrance ramps. The main steps of this method include S1 to S4:

[0017] S1. Establish a traffic flow simulation for entrance ramps based on a cellular automata model. Specifically, by constructing a cellular automata model, the driving states of autonomous and manually driven vehicles on entrance ramps and main roads are discretized to simulate vehicle driving dynamics under various traffic conditions.

[0018] S2. Set simulation parameters and initial conditions. Specifically, based on actual road conditions and traffic demand, set road condition simulation parameters and initial traffic flow conditions. These include parameters such as the number of lanes, lane length, ramp length, weaving zone length, percentage of automated vehicles, maximum vehicle speed, acceleration, deceleration, initial vehicle density, initial vehicle speed entering the system, and initial traffic light conditions.

[0019] S3. Simulate Traffic Flow. Using a cellular automata model, simulate the movement of vehicles on entrance ramps and main roads, considering inter-vehicle interactions such as safety distance and lane-changing motivations to reflect actual traffic flow. Specifically, under a given traffic light control scheme, obtain the speed and position of each vehicle at time t through real-time monitoring equipment or iterative calculations during the simulation process. Update the speed and position of each vehicle within each time step, along with lane-changing rules based on inter-vehicle distance, speed, and acceleration. Through iterative updates, simulate the movement of vehicles on entrance ramps and main roads, reflecting the actual traffic flow. Furthermore, the traffic light control scheme can be iterated multiple times, and traffic light scoring and ranking can be performed.

[0020] S4. Analyze Traffic Flow Characteristics. Based on the simulation results, analyze the traffic flow characteristics on entrance ramps and main roads, such as vehicle density, average speed, flow rate, and the impact of the proportion of large vehicles on traffic flow. Real-time visualization charts are generated to show the traffic flow status under different conditions. Flow rate is the number of vehicles passing through a certain road segment, calculated by dividing the number of vehicles passing the observation point by the simulation time; average speed is the average speed of all vehicles passing through the observation point, calculated by dividing the sum of the vehicle speeds at each observation point by the number of vehicles passing through that point; vehicle density refers to the number of vehicles per unit length of road, calculated by dividing the flow rate by the average speed; correction factors are also included. f HV This method is used to represent the impact of large vehicles on ramp capacity. By observing the simulation results under different proportions of large vehicles, the traffic flow changes under different proportions of large vehicles are calculated, and the correction coefficient is fitted by the linear regression method.

[0021] Specifically, speed and position updates refer to obtaining the speed and position of each vehicle at time t through real-time monitoring equipment or iterative calculations during the simulation process, and updating the speed and position of each vehicle within each time step; through iterative updates, the driving process of vehicles on entrance ramps and main roads is simulated, reflecting the actual driving situation of traffic flow.

[0022] For each manually driven vehicle, the driving speed is updated according to the following formula:

[0023]

[0024] in, It is the first i A number of manually driven vehicles at all times t driving speed, a Indicates acceleration. The maximum speed allowed on the current road. Indicates the first i A number of manually driven vehicles at all times t Location, Indicates the time of the vehicle in front. t Location, d Indicates a safe following distance;

[0025] For each autonomous vehicle, the driving speed is updated according to the following formula:

[0026]

[0027] in, It is the first i A self-driving car at all times t driving speed, C This indicates the communication capability constraints of autonomous vehicles; when the first i When the vehicle is manually driven, C =0.

[0028] In particular, vehicle speeds may be affected by special weather conditions, such as rain or snow. In such cases, the maximum speed, acceleration, and deceleration parameters in the model can be adjusted to adapt to different weather conditions. Thus, this embodiment can better adapt to special situations and provide targeted solutions for traffic management and planning.

[0029] If the vehicle is within a lane-changing zone, a lane-changing rule is validated at each time step. This rule is based on the distance, speed, and acceleration between vehicles. The merits of a lane-changing operation are evaluated by calculating the distance, speed, and acceleration between vehicles; if a lane change results in higher speed and / or a safer distance, the lane change operation is executed. Specific steps include: first, calculating the speed difference between the vehicle ahead and the current vehicle in the current lane. and distance difference Based on speed difference, distance difference, and safety time interval T To calculate the safe distance between the current vehicle and the vehicle in front. s The formula is as follows:

[0030]

[0031] in, self.a and self.b These represent the current vehicle's acceleration and deceleration, respectively; then, the lane-changing cost is assessed based on whether the lane-changing operation can increase the current vehicle's speed. cost Define two functions: lane change cost and get lane change cost These are used to calculate lane-changing costs and compare the priorities of two candidate lanes, respectively; among them, Lane change cost The function calculates lane-changing costs based on the following conditions: if changing lanes increases speed, the cost is low; if changing lanes doesn't change speed but increases the safe distance from the vehicle ahead, the cost is low; if changing lanes decreases speed, the cost is high; if the cost is low, the lane-changing operation is performed. When there are two candidate lanes... Get lane change cost The function can compare the priority of two candidate lanes based on the lane-changing cost. The lane with the lower lane-changing cost has higher priority, and the lane with higher priority is selected to perform the lane-changing operation.

[0032] Furthermore, competitive and cooperative strategies can be incorporated into the main road traffic, simulating two types of vehicles: those accelerating to overtake and those slowing down to yield in the entrance ramp area. Specifically, if a vehicle is in an adjacent lane in the ramp merging area, the system will consider whether to adopt a cooperative or competitive strategy. If the vehicle does not adopt any strategy, it will proceed according to normal rules. If the vehicle adopts a competitive strategy, it will attempt to accelerate through the gap ahead.

[0033] Competition and cooperation strategies are achieved by adjusting vehicle speeds. First, the following symbols are defined:

[0034] a The acceleration of a vehicle under a competitive strategy;

[0035] b Vehicle deceleration under a cooperative strategy;

[0036] d The distance between the vehicle and the vehicle in front;

[0037] v The vehicle's current speed;

[0038] The maximum speed allowed for the vehicle;

[0039] s The safe distance between a vehicle and the vehicle in front;

[0040] strategy A parameter used to switch between competition and cooperation strategies, with a value range of {'none', 'competition', 'cooperation'}.

[0041] Based on these symbols, the design of competition and cooperation strategies can be described as follows:

[0042] if strategy If it is 'none', the vehicle will not adopt competitive or cooperative strategies and will drive according to normal rules;

[0043] if strategy If the condition is 'competition', the vehicle adopts a competitive strategy, under which the vehicle attempts to accelerate through the gap ahead; specifically, if d > s The vehicle will accelerate until it reaches its maximum speed. ;

[0044] if strategy If the setting is 'cooperation', the vehicle adopts a cooperative strategy. Under this strategy, the vehicle slows down to increase the clearance with the vehicle in front; specifically, if d < s The vehicle will slow down.

[0045] During the simulation, at each time step, the vehicle adjusts its speed according to the current strategy; by controlling the strategy parameters, the competition and cooperation strategies can be switched on and off during the simulation.

[0046] To examine the impact of traffic signal schemes on vehicle movement in greater depth, observation points were added at specific locations ahead of the stop line to investigate the dispersion of vehicle platoons. The investigation methods included calculating the average distance between all vehicles in the platoon and using this average to calculate the deviation of vehicle distances within the platoon at the intersection, thereby assessing the dispersion of vehicle platoons under different traffic parameters. Specific investigation methods included:

[0047] First, calculate the average distance between all vehicles in the convoy. μ :

[0048]

[0049] in, n This refers to the number of vehicles in the convoy. x i Indicates the first iThe location of the vehicle;

[0050] Next, the deviation of the distance between vehicles within the convoy is calculated. σ :

[0051]

[0052] The dispersion of the team is also used This can be represented as follows: during simulation, the dispersion of the vehicle fleet can be calculated at each time step, and its changes can be observed; by comparing different vehicle fleets... It can assess the dispersion of vehicle fleets under different traffic parameters.

[0053] Example 2

[0054] Based on the simulation method and traffic light settings provided in Example 1 above, a method for evaluating entrance ramp signal schemes is provided. The specific steps of this method are as follows:

[0055] First, we define an evaluation metric to measure the performance of different combinations of traffic light durations. This metric consists of two parameters with adjustable weights: vehicle throughput efficiency and ramp queue length, denoted by E.

[0056] Determine the search range for traffic light durations. For example, the minimum green light duration can be set to 5 seconds and the maximum to 60 seconds; the minimum red light duration can be set to 5 seconds and the maximum to 60 seconds.

[0057] Iterate through all possible combinations of green light duration G and red light duration R. In this example, G ranges from [5, 60], and R ranges from [5, 60]. The method will iterate through each combination of G and R (G, R).

[0058] For each traffic light duration combination (G, R), perform the following operations:

[0059] Simulate using a given duration (G, R) and calculate the evaluation index E. During the iteration, record the highest-scoring traffic light duration combination (Gbest, Rbest). Each time the evaluation index E is calculated, check if the score of the current combination (G, R) is higher than the score of the current best combination (Gbest, Rbest). If the current combination has a higher score, update the best combination to the current combination.

[0060] After the traversal is complete, the optimal traffic light duration combination (Gbest, Rbest) is the duration combination with the highest score.

[0061] As can be seen, the entrance ramp mixed traffic simulation method provided in the foregoing embodiments of the present invention can simulate the driving states of autonomous and manually driven vehicles on entrance ramps and main roads. It discretizes the characteristics of the two driving modes to more accurately reflect real traffic conditions. It simulates the competitive and cooperative behaviors of vehicles on main roads in the entrance ramp area, such as accelerating to overtake and decelerating to yield, to more realistically simulate the behavioral differences between different drivers. Through multiple simulation iterations, various traffic light control schemes are traversed, and multi-factor comprehensive scoring and ranking are performed using adjustable weights, thereby providing the optimal traffic light control scheme for specific traffic flow states. The simulation uses cellular automata, which is discretized, easily scalable, easy to compute, and highly efficient.

[0062] Considering individual driving behavior, this study simulates the behavioral differences among drivers through competition and cooperation strategies. By iterating multiple times and using adjustable weights for comprehensive scoring and ranking, it provides the optimal traffic light control scheme for specific traffic flow conditions, helping to reduce operating costs.

[0063] The above description, in conjunction with specific embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various equivalent substitutions or obvious modifications can be made without departing from the concept of the present invention, and all such modifications should be considered within the scope of protection of the present invention.

Claims

1. A method for simulating mixed traffic at entrance ramps, characterized in that, include: S1. Establish traffic flow simulation of entrance ramps based on cellular automata models, including: discretizing the driving states of autonomous and manual vehicles on entrance ramps and main roads by constructing cellular automata models, and simulating vehicle driving dynamics under various traffic conditions. S2. Set simulation parameters and initial conditions, including: set road condition simulation parameters and initial conditions for traffic flow based on actual road conditions and traffic demand; S3. Simulate the traffic flow process, including: based on the cellular automata model, obtain the real-time speed and position of each vehicle through real-time monitoring equipment or iterative calculations during the simulation process, update the speed and position of each vehicle in each time step, and lane-changing rules based on the distance, speed and acceleration between vehicles; through iterative updates, simulate the driving process of vehicles on the entrance ramps and main roads to reflect the actual traffic flow situation. S4. Analyze traffic flow characteristics, including: based on the simulation results, analyze the driving characteristics of traffic flow on entrance ramps and main roads, as well as the impact of the proportion of large vehicles on traffic flow, and draw visualization charts in real time to show the driving status of traffic flow under different conditions. The lane-changing rules are determined based on the distance, speed, and acceleration between vehicles. The merits of a lane-changing operation are evaluated by calculating the distance, speed, and acceleration between vehicles. If a lane-changing operation can result in higher speed and / or a safer distance, then the lane-changing operation is performed. The steps for the lane-changing rule are as follows: First, calculate the speed difference between the vehicle ahead and the current vehicle in the current lane. and distance difference Based on speed difference, distance difference, and safety time interval T To calculate the safe distance between the current vehicle and the vehicle in front. s The formula is as follows: in, self.a and self.b These represent the current acceleration and deceleration of the vehicle, respectively. Next, the cost of lane changing is assessed based on whether the lane-changing operation can increase the speed of the current vehicle. cost Define two functions: lane change cost and get lane change cost These are used to calculate lane-changing costs and compare the priorities of two candidate lanes, respectively. in, Lane change cost The function calculates lane-changing costs based on the following conditions: if changing lanes increases speed, the lane-changing cost is low; if changing lanes does not change speed but increases the safe distance from the vehicle in front, the lane-changing cost is low; if changing lanes decreases speed, the lane-changing cost is high; when the lane-changing cost is low, the lane-changing operation is performed. Get lane change cost The function is used to compare the priority of two candidate lanes based on lane-changing cost. The lane with lower lane-changing cost has higher priority, and the lane with higher priority is selected for lane changing. This also includes: adding observation points at predetermined positions ahead of the traffic light stop line to examine the dispersion of the convoy; the methods for examination include: First, calculate the average distance between all vehicles in the convoy. μ : in, n This refers to the number of vehicles in the convoy. x i Indicates the first i The location of the vehicle; Next, the deviation of the distance between vehicles within the convoy is calculated. σ : The dispersion of the team is also used This can be represented as follows: during simulation, the dispersion of the vehicle fleet can be calculated at each time step, and its changes can be observed; by comparing different vehicle fleets... It can assess the dispersion of vehicle fleets under different traffic parameters.

2. The entrance ramp mixed traffic simulation method as described in claim 1, characterized in that, For each manually driven vehicle, the driving speed is updated according to the following formula: in, It is the first i A number of manually driven vehicles at all times t driving speed, a Indicates acceleration. The maximum speed allowed on the current road. Indicates the first i A number of manually driven vehicles at all times t Location, Indicates the time of the vehicle in front. t Location, d Indicates a safe following distance; For each autonomous vehicle, the driving speed is updated according to the following formula: in, It is the first i A self-driving car at all times t driving speed, C This indicates the communication capability constraints of autonomous vehicles; when the first i When the vehicle is manually driven, C =0.

3. The entrance ramp mixed traffic simulation method as described in claim 1, characterized in that, It also includes: adding competition and cooperation strategies to vehicles on the main road, simulating two types of vehicles that accelerate to overtake and decelerate to yield in the entrance ramp area; The steps of a competition and cooperation strategy are as follows: Competition and cooperation strategies are achieved by adjusting vehicle speeds. First, the following symbols are defined: a The acceleration of a vehicle under a competitive strategy; b Vehicle deceleration under a cooperative strategy; d The distance between the vehicle and the vehicle in front; v The vehicle's current speed; The maximum speed allowed for the vehicle; s The safe distance between a vehicle and the vehicle in front; strategy A parameter used to switch between competition and cooperation strategies, with a value range of {'none', 'competition', 'cooperation'}. Based on these symbols, the design of competition and cooperation strategies can be described as follows: if strategy If it is 'none', the vehicle will not adopt competitive or cooperative strategies and will drive according to normal rules; if strategy If the condition is 'competition', the vehicle adopts a competitive strategy, under which the vehicle attempts to accelerate through the gap ahead; specifically, if d > s The vehicle will accelerate until it reaches its maximum speed. ; if strategy In the 'cooperation' mode, the vehicle adopts a cooperative strategy; under this strategy, the vehicle decelerates to increase the clearance with the vehicle in front; specifically, if d < s The vehicle will slow down. During the simulation, at each time step, the vehicle adjusts its speed according to the current strategy; by controlling the strategy parameters, the competition and cooperation strategies can be switched on and off during the simulation.

4. The entrance ramp mixed traffic simulation method as described in claim 1, characterized in that, Step S4 specifically includes: Based on the simulation results, the driving characteristics of traffic flow on entrance ramps and main roads are analyzed, including the impact of vehicle density, average speed, flow rate and the proportion of large vehicles on traffic flow. Wherein, flow rate is the number of vehicles passing through a certain section of road, obtained by dividing the number of vehicles passing through the observation point by the simulation time; average speed is the average speed of all vehicles passing through the observation point, obtained by dividing the sum of the vehicle speeds passing through the observation point by the number of vehicles passing through the observation point; vehicle density refers to the number of vehicles per unit length of road, obtained by dividing flow rate by average speed; and a correction factor is used. f HV This represents the impact of large vehicles on ramp capacity. By observing the simulation results under different proportions of large vehicles, the traffic flow changes under different proportions of large vehicles are calculated, and the correction coefficient is fitted by the linear regression method.

5. The entrance ramp mixed traffic simulation method as described in claim 1, characterized in that, Also includes: Set the initial conditions for the traffic lights, and in step S2, simulate the traffic flow process under the given traffic light control scheme; The traffic light control scheme was iterated multiple times and scored and ranked.

6. The entrance ramp mixed traffic simulation method as described in claim 5, characterized in that, The traffic light control scheme was iterated multiple times and scored and ranked, including: First, we define an evaluation index to measure the performance of different combinations of traffic light durations. This index consists of two parameters with adjustable weights: vehicle throughput efficiency and ramp queue length, denoted by E. Determine the search range for the traffic light duration, and then iterate through all possible combinations of green light duration G and red light duration R; For each traffic light duration combination (G, R), perform the following operations: Simulate using a given duration (G, R) and calculate the evaluation index E; during the traversal, record the traffic light duration combination with the highest score (Gbest, Rbest); each time the evaluation index E is calculated, check whether the score of the current combination (G, R) is higher than the score of the current best combination (Gbest, Rbest); if the score of the current combination is higher, then update the best combination to the current combination; After the traversal is complete, the optimal traffic light duration combination (Gbest, Rbest) is the duration combination with the highest score.

Images