Vehicle-road collaborative lane changing decision and control method based on potential conflict analysis

Abstract

The invention provides a vehicle-road collaborative lane changing decision and control method based on potential conflict analysis. The method specifically comprises the following steps: S1, collecting and processing vehicle and surrounding environment information; s2, starting lane changing decision calculation; s3, calculating a lane changing influence range; s4, lane changing conflict vehicles are screened; analyzing the relationship among the vehicles in the lane changing influence range of each lane changing vehicle, and judging whether potential conflicts exist among the lane changing vehicles or not; s5, calculating the lane changing priority; and S6, lane changing execution. According to the invention, the potential lane changing conflict risk can be effectively reduced, and the vehicle lane changing efficiency is improved; by considering the arrival randomness and space-time evolution characteristics of the lane changing vehicle, the dynamic change rule of the lane changing process of the vehicle can be reflected more objectively to adapt to different lane changing scenes and lane changing requirements, and the method has important theoretical significance and practical application value for promoting the application of a vehicle-road cooperation technology and improving traffic safety.

Description

Technical Field

[0001] The present invention relates to the field of intelligent traffic control technology, and in particular to a vehicle-road cooperative lane changing decision and control method based on potential conflict analysis. Background Art

[0002] The existing research on vehicle lane-changing decision models mainly includes two types: traditional rule-making models and predictive control models based on sensor data. Among them, the rule-based vehicle-road cooperative lane-changing model usually includes three stages: perception, judgment and execution. The lane-changing decision rules are mainly established through the lane-changing safety distance or based on game theory; the decision-making based on sensor data mainly focuses on the perception and analysis of the surrounding environment information, obtains the information of surrounding vehicles through sensors, and uses machine learning algorithms to analyze these data to make more accurate lane-changing decisions.

[0003] From the perspective of the research object, the existing research on vehicle-road cooperative lane-changing models mostly focuses on the lane-changing process under fixed lane-changing scenarios. By designing typical scenarios, such as lane-changing behaviors between three vehicles in two lanes and eight vehicles in three lanes, it is assumed that there are one or several lane-changing vehicles, and then the specific motion state and lane-changing trajectory of the lane-changing vehicle and surrounding vehicles are planned. Since the lane-changing scenario is relatively idealized, the proposed model is only applicable to certain specific lane-changing scenarios and is not compatible with other lane-changing scenarios and processes. Therefore, how to improve the existing vehicle-road cooperative lane-changing model to adapt to more lane-changing scenarios and processes is the focus of this research method.

[0004] From the perspective of model application, current lane-changing research includes single-vehicle lane-changing and multi-vehicle lane-changing, but most of the research on multi-vehicle coordinated lane-changing models focuses on single lane-changing processes in fixed scenarios, without considering the randomness and spatiotemporal evolution characteristics of the arrival of lane-changing vehicles. In real traffic environments, lane-changing scenarios and lane-changing processes are changing all the time. How to define the beginning and end of the lane-changing process based on spatiotemporal evolution, as well as the interaction between vehicles during the lane-changing process, especially the dynamic interaction between the lane-changing vehicle and other lane-changing vehicles around it, is relatively rare. Therefore, how to effectively define the beginning and end of a vehicle's lane change, as well as the mutual influence between the lane-changing vehicle and other lane-changing vehicles around it, is one of the key and difficult points of this research method. Summary of the invention

[0005] The purpose of the present invention is to propose a vehicle-road cooperative lane changing decision and control method based on potential conflict analysis, which can effectively reduce the potential lane changing conflict risk and improve the vehicle lane changing efficiency; by considering the arrival randomness and spatiotemporal evolution characteristics of lane changing vehicles, it can more objectively reflect the dynamic change law of the vehicle lane changing process to adapt to different lane changing scenarios and lane changing requirements, which has important theoretical significance and practical application value for promoting the application of vehicle-road cooperative technology and improving traffic safety.

[0006] To achieve the above object, the technical solution of the present invention is: a vehicle-road cooperative lane change decision and control method based on potential conflict analysis, which specifically includes the following steps:

[0007] S1. Collect and process vehicle and surrounding environment information, including vehicle position, vehicle spacing, lane change requirements, intersection signal light status, distance between the vehicle and the intersection stop line, and upstream lane change area; and determine whether the vehicle is in the lane change area. If the vehicle has entered the solid line segment of the lane divider, lane change is prohibited.

[0008] S2. Start lane-changing decision calculation; obtain the lane-changing requirements of the vehicle in real time through detection and calculation of the vehicle-road cooperative system, and record it in the dynamic lane-changing array; design a lane-changing decision trigger model to determine when to start and stop the lane-changing calculation, and calculate the feasibility of lane-changing vehicles in the dynamic array according to the spatiotemporal evolution;

[0009] S3, calculating the lane change impact range: when the vehicle has a lane change demand, the lane change impact range is calculated based on the received driving speed, vehicle spacing and safety distance of surrounding vehicles;

[0010] S4. Screening of vehicles with lane-changing conflicts: Analyze the relationship between vehicles within the lane-changing influence range of each lane-changing vehicle to determine whether there is a potential conflict between the lane-changing vehicles;

[0011] S5. Calculate lane-changing priority; quantify various influencing factors in the lane-changing process of vehicles, so that vehicles with less disturbance to the entrance lane have higher priority, and determine the final lane-changing array and lane-changing order;

[0012] S6, lane change execution; make the best lane change decision based on the comprehensive calculation results, and execute the lane change according to the decision results while ensuring safety; if the distance between the lane-changing vehicle and the target lane vehicle meets the minimum safety distance required for lane change, directly execute the lane change; otherwise, it is necessary to plan the coordination mode of surrounding vehicles based on the surrounding real-time environmental information and vehicle motion status to cooperate with the lane-changing vehicle to complete the lane change task; when all vehicles in the final lane change array have completed the lane change, restart a new round of lane change decision calculation.

[0013] Preferably, the lane change decision triggering model performs the following steps:

[0014] (1) When a vehicle is detected to have a lane change requirement, its vehicle number, position, speed, distance from the stop line, and target lane are stored in a dynamic lane change array;

[0015] (2) When the lane-changing array is not zero, the lane-changing decision calculation switch is activated; the lane-changing vehicles are sorted in descending order according to the distance from the stop line, the lane-changing influence range of each lane-changing vehicle is calculated, and the interaction relationship between the vehicles is determined to determine the final lane-changing array for subsequent lane-changing planning:

[0016] ① If there are no other lane-changing vehicles within the affected range, the lane change is performed after the safety conditions are met, and the lane-changing vehicle number and total number N1 in this case are recorded;

[0017] ② If there are other lane-changing vehicles within the impact range but the lane-changing trajectories between the vehicles do not conflict, each vehicle will change lanes after the safety conditions are met, and the lane-changing vehicle numbers and total number N2 in this case will be recorded;

[0018] ③ When there are conflicting vehicles with intersecting trajectories within the lane change influence range, the number of lane change conflicts between them needs to be counted: when the number of conflicts is greater than or equal to 2, the lane change conflict is large and the vehicles will temporarily keep following; when the number of conflicts between vehicles is equal to 1, the lane change priority needs to be calculated, and the vehicle with the highest priority is selected to perform the lane change, and the lane change vehicle number and total number N3 in this case are recorded;

[0019] Then the total number of lane-changing vehicles in the final lane-changing array is N=N1+N2+N3; the final non-lane-changing vehicles include: vehicles with a conflict number greater than or equal to 2 and vehicles with a conflict number of 1 but a smaller priority; the final non-lane-changing vehicles will rejoin the lane-changing array after the current round of lane changing is completed;

[0020] (3) Declare a temporary variable cl_flag to control the start and end of the lane change calculation; set the initial value of cl_flag to 0. When the final lane change array is not zero, set the value of cl_flag to N. When cl_flag is not 0, execute the vehicle lane change operation; when each vehicle completes the lane change, reduce the value of cl_flag by 1. This process will continue to loop until cl_flag is equal to 0 again, that is, all vehicles in the final lane change array have completed the lane change, and then a new round of lane change calculation can be started.

[0021] Preferably, the lane change impact range is a rectangular area with the lane change impact distance as the length and the total lane width as the width. The lane change impact distance is calculated as follows:

[0022] The expected distance between the vehicle and the target vehicle in front during the lane change process is obtained by the following formula:

[0023]

[0024] The vehicle lane change affects the distance:

[0025]

[0026] Where: T is the safety time interval; V i，j is the speed of the jth vehicle in the i-th lane; D safe is the safety space distance that needs to be met; ΔV (i+1,j)-(i,j) is the speed difference between the lane-changing vehicle and the vehicle in front of the target lane; amax is the maximum acceleration of the lane-changing vehicle; b is the comfortable deceleration of the lane-changing vehicle; W is the required lateral safety distance;

[0027] When the speed of the vehicle in front of the target lane is greater than the expected lane-changing speed, the calculation of the vehicle lane-changing impact distance is changed from equation (2) to equation (3):

[0028]

[0029] Where: L exp ΔV is the lane-changing impact distance of the vehicle in front of the target lane at the expected speed; exp-(i,j) is the difference between the lane-changing vehicle speed and the expected speed of the preceding vehicle in the target lane.

[0030] Preferably, the required lateral safety distance is the lane width.

[0031] Preferably, the lane-changing conflict vehicle screening is specifically as follows: according to the vehicle lane-changing influence range, determine whether there are conflicting cross-lane-changing vehicles; if there are no other lane-changing vehicles within the influence range, directly execute the lane change after determining that they meet the safety conditions; if there are other lane-changing vehicles but there is no conflict between the vehicles, determine that they meet the safety conditions and then each executes the lane change; if there are lane-changing conflict vehicles with intersecting trajectories, it is necessary to count their respective lane-changing conflict numbers: ① When the conflict numbers are all greater than or equal to 2, the lane change has a greater interference on the entrance lane and the vehicle does not execute the lane change temporarily; ② When the conflict numbers between vehicles are all equal to 1, it is necessary to calculate the lane-changing priority, and the one with the larger priority executes the lane change.

[0032] Preferably, the lane changing priority comprehensively considers the influence of the number of lanes that the vehicle needs to cross when changing lanes, the traffic flow density and the distance from the stop line on the lane changing priority.

[0033] Preferably, the lane change priority is calculated as follows:

[0034]

[0035] Where P r is the lane-changing priority; λ, μ are weight coefficients; P ln is the influencing factor of the number of lanes; P kl is the influencing factor of traffic flow density and distance to the stop line;

[0036]

[0037] In the formula, Δn is the number of lanes that the vehicle needs to cross; n is the total number of lanes where the vehicle is currently located;

[0038]

[0039] In the formula, K, Kj is the traffic density and blocking density; α is the influence coefficient of traffic density; β is the influence coefficient of the distance to the stop line; γ is the process parameter; L is the length of the entrance lane; ΔL is the distance between the lane-changing vehicle and the stop line.

[0040] Preferably, the traffic flow density influence coefficient α is set to 3.597, the stop line distance influence coefficient β is set to 0.104, and the process parameter γ is set to 1.

[0041] Preferably, if the distance between the lane-changing vehicle and the vehicle in the target lane meets the minimum safety distance required for lane changing, the lane changing is performed directly; otherwise, it is necessary to plan the coordination mode of surrounding vehicles based on the surrounding real-time environmental information and vehicle motion status to cooperate with the lane-changing vehicle to complete the lane changing task; specifically:

[0042] The distance between the lane-changing vehicle and the vehicle in front of it must always satisfy Δd (i，j)-(i+1，j+1) ≥D safe ; The lane-changing vehicle and the vehicle in front of the target lane need to maintain a minimum diagonal safety distance D f :

[0043]

[0044] Set the time when the distance between the vehicle behind the target lane and the lane-changing vehicle exceeds D tg When the lane-changing vehicle is 0, there is no need to implement a cooperative strategy. The lane-changing vehicle changes lanes directly without considering the impact of the lane-changing vehicle on the following vehicle.

[0045] D tg =V max T c +D safe (17)

[0046] Where: D tg No lane change cooperation interval is required; V max is the road speed limit; T c The time for vehicles to change lanes;

[0047] If the distance between the current position of the lane-changing vehicle and the vehicle behind the target lane does not meet the lane-changing requirements, the vehicle behind the target lane will actively respond to the lane-changing request by slowing down and widening the gap after receiving the lane-changing signal. When the lane-changing vehicle determines that the gap can meet the requirements for safe lane changing, it starts to execute the lane-changing operation.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] The present invention analyzes multiple factors (the current position and speed of the lane-changing vehicle, the speed difference with the vehicle in front of the target lane, the distance between vehicles, etc.) and other lane-changing participants, designs a lane-changing decision trigger model based on the arrival randomness and spatiotemporal evolution characteristics of the lane-changing vehicles, calculates the influence range of each lane-changing vehicle, and resolves potential conflicts in the lane-changing process by adjusting the lane-changing sequence of the vehicles, and finally determines the lane-changing plan with the least overall disturbance to the entrance lane. This model can effectively reduce the potential risk of lane-changing conflicts and improve the efficiency of vehicle lane-changing; by considering the arrival randomness and spatiotemporal evolution characteristics of the lane-changing vehicles, it can more objectively reflect the dynamic change law of the vehicle lane-changing process to adapt to different lane-changing scenarios and lane-changing needs, and has relatively important theoretical significance and practical application value for promoting the application of vehicle-road cooperative technology and improving traffic safety.

[0050] The relationship between lane-changing vehicles in the vehicle-road cooperative environment is analyzed. The vehicle lane-changing influence range is defined based on information such as traffic flow density, the position and speed of the front vehicle in the target lane, and the corresponding calculation method is given, which effectively describes the dynamic interaction relationship between lane-changing vehicles in the local area; based on the lane-changing influence range, it is determined whether there is a potential conflict between lane-changing vehicles, and the various influencing factors in the conflict lane-changing scenario are quantified. Combined with the real-time status of the lane-changing vehicles, a cooperative lane-changing rule based on lane-changing priority is proposed. By adjusting the order of vehicle lane changes, vehicles are guided to change lanes in an orderly manner, and the impact of lane changes on the stability of traffic flow at intersections is reduced. In the process of vehicle lane changes, the global information of the road network is fully considered, which provides an important reference for the application of future vehicle-road cooperative technology in vehicle lane-changing decision-making and control. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A schematic diagram of the positional relationship between vehicles in multiple lanes in one embodiment of the present invention;

[0052] Figure 2 A schematic diagram of a typical potential conflict lane-changing scenario in an embodiment of the present invention;

[0053] Figure 3 A detailed flow chart of lane change decision in one embodiment of the present invention;

[0054] Figure 4 A flow chart of a lane change decision triggering model in one embodiment of the present invention;

[0055] Figure 5 is the vehicle lane-changing impact distance in one embodiment of the present invention;

[0056] Figure 6 The lane change influence range calculation process in one embodiment of the present invention;

[0057] Figure 7 A lane-changing conflict scenario in an embodiment of the present invention;

[0058] Figure 8 The lane change priority calculation process in one embodiment of the present invention;

[0059] Fig. 9 Schematic diagram of a critical collision between a lane-changing vehicle and a rear vehicle in a target lane in one embodiment of the present invention. DETAILED DESCRIPTION

[0060] The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings.

[0061] The present invention proposes a vehicle-road cooperative lane changing decision and control method based on potential conflict analysis.

[0062] In order to express the vehicle lane-changing decision process in the form of a mathematical model, the following assumptions need to be made:

[0063] ① All vehicles are intelligent connected vehicles, and information exchange between vehicles and between vehicles and roadside facilities is smooth, without considering communication delay issues.

[0064] ② When planning the vehicle lane-changing trajectory, the longitudinal and lateral movements of the vehicle are decoupled, that is, the longitudinal and lateral movements of the vehicle are relatively independent, and the vehicle can drive completely according to the control instructions after receiving the decision results.

[0065] ③ In order to facilitate the establishment of a safe distance model for lane changing, the influence of vehicle appearance on lane changing is ignored. It is assumed that the vehicles involved in the model are all standard cars and other models are not considered.

[0066] In actual road conditions, multiple lanes and multiple vehicles are in dynamic driving, and the interaction between vehicles is obvious. Taking three lanes as an example, Figure 1 Vehicle C i，j The position at time t is taken as the reference point, the traffic flow direction is the X-axis, and the direction perpendicular to the traffic flow is the Y-axis. The position relationship between vehicles is described as follows:

[0067] ①i represents the lane number, j represents the jth vehicle in the i-th lane, and the vehicle number decreases with the direction of traffic flow.

[0068] ②Lane i Represents the i-th lane. The closer the lane is to the central dividing strip, the larger the value.

[0069] ③X i，j (t), Y i，j (t), V i，j (t) represents the longitudinal distance, lateral distance and speed of the jth vehicle in the i-th lane at time t, Vx i，j (t), Vy i，j (t) represents the longitudinal speed and the lateral speed respectively; Δd (i，j)-(i+1，j+1)(t) represents the longitudinal distance between the jth vehicle in lane i and the j+1th vehicle in lane i+1 at time t.

[0070] ④L0 is the length of the solid line of the lane divider. Vehicles are prohibited from changing lanes within the solid line segment. L is the length of the entrance lane. At time t, the distance between the vehicle and the intersection stop line is ΔL i，j (t).

[0071] Figure 1 In the figure, yellow vehicles represent non-lane-changing vehicles, and blue vehicles represent vehicles that need to change lanes. From the relationship between the vehicles in the figure, we can see that: ① Vehicle C i+1，j-1 Change lane from lane i+1 to lane i. At this time, there is also a lane-changing car C in lane i+2. i+2，j , but the lane-changing trajectories of the two vehicles do not intersect; ② Vehicle C i+2，j+1 Changing from lane i+2 to lane i+1, there are two other lane-changing vehicles in lane i and lane i+1, among which C i，j The lane change trajectory will not be the same as C i+2，j+1 Produce crossover, and C i+1，j+1 The lane-changing trajectory and C i+2，j+1 Produce crossover.

[0072] According to the above analysis, the process of multi-vehicle coordinated lane changing involves many scenarios and the lane changing situations are complex, and their distribution patterns cannot be directly described by mathematical models. This paper analyzes and classifies a large number of possible coordinated situations in light of actual traffic conditions. By analyzing the characteristics of vehicle lane changing, the initial position and target position of lane changing, two potential conflicting lane changing scenarios that have a greater impact on traffic flow and can reflect information interaction are finally summarized. Figure 2 .

[0073] C in sub-figure (a) 1，2 , C 3，2 The initial position is one lane away, and the target lane is the same lane after the lane change intention is generated; in sub-figure (b), C 2，2 , C 3，2 The initial position is in the adjacent lane, and the lane change target position is the other party's initial lane. If the two vehicles change lanes in a short time, the lane change trajectories will inevitably intersect, and there is a risk of conflict.

[0074] In reality, these two potentially conflicting lane-changing situations manifest themselves as follows: if one vehicle changes lanes first, the other vehicle may give up changing lanes due to insufficient required safety distance, thus failing to meet the lane-changing requirement.

[0075] Through the analysis of traffic flow characteristics and the spatiotemporal characteristics of vehicle lane-changing behavior, it can be seen that when the traffic flow density is low, the vehicle speed is fast and the vehicle spacing is large, and the vehicle driving state is good, the vehicle lane-changing behavior has basically no effect on the traffic flow. When the traffic flow density is high, the vehicle lane-changing behavior will affect the normal operation of the traffic flow to a certain extent. Because the traffic flow is in an unstable state at this time, the vehicles behind the original lane and the target lane will be disturbed by the vehicle lane-changing behavior and reduce their speed. This slowing effect will spread upstream with the traffic flow during the traffic operation. When the traffic flow density continues to rise and causes traffic congestion, the vehicle spacing decreases and the speed drops to a lower level. At this time, whether the vehicle starts to change lanes just after entering the upstream of the intersection or changes lanes before the stop line, the impact on the traffic flow is the same, which will aggravate the degree of traffic congestion. Therefore, this paper mainly focuses on the mandatory lane-changing demand under the high-density conditions of signal-controlled intersections, and studies the vehicle-road cooperative lane-changing decision model based on potential conflict analysis.

[0076] The first step is to collect and process vehicle and surrounding environment information. In the vehicle-road collaborative environment, vehicles and roadside equipment can share information in real time, including vehicle location, vehicle spacing, lane change requirements, intersection signal light status, distance between the vehicle and the intersection stop line, and upstream lane change area. At the same time, it is determined whether the vehicle is in the lane change area. If the vehicle has entered the solid line segment of the lane divider, lane change is prohibited.

[0077] Step 2: Start the lane change decision calculation. Through the detection and calculation of the vehicle-road cooperative system, the lane change requirements of the vehicle can be obtained in real time and recorded in the dynamic lane change array. To ensure the safety and smoothness of lane changes, it is necessary to design a lane change decision trigger model. The core of this model is to determine when to start and stop the lane change calculation, and perform a series of lane change feasibility calculations on the lane change vehicles in the dynamic array according to the evolution of time and space.

[0078] Step 3: Calculate the lane change impact range. During the lane change process, the vehicle interacts with the surrounding vehicles. Different road conditions and driving speeds will affect the interaction between vehicles. Therefore, the interaction area between vehicles is constantly changing. This dynamic area is defined as the vehicle's lane change impact range. When the vehicle has a lane change demand, the lane change impact range is calculated based on the received driving speed, vehicle spacing and safety distance of the surrounding vehicles.

[0079] Step 4: Filtering vehicles with lane-changing conflicts. Analyze the relationship between vehicles within the lane-changing influence range of each lane-changing vehicle to determine whether there is a potential conflict between lane-changing vehicles. There are three main types of conflicting vehicles: ① Lane-changing vehicles with intertwined lane-changing trajectories; ② Lane-changing vehicles with a conflict number greater than or equal to 2; ③ Lane-changing vehicles with a conflict number equal to 1 and a lower lane-changing priority.

[0080] Step 5: Calculate lane-changing priority. Quantify the influencing factors in the lane-changing process, so that vehicles with less disturbance to the entrance lane have higher priority, and determine the final lane-changing array and lane-changing sequence. By adjusting the lane-changing sequence of vehicles, the mutual interference of lane-changing vehicles is reduced, and the impact of lane-changing on the stability of traffic flow at the intersection is reduced.

[0081] Step 6: Lane change execution. The optimal lane change decision is made based on the comprehensive calculation results, and the lane change is executed according to the decision results while ensuring safety. If the distance between the lane-changing vehicle and the target lane vehicle meets the minimum safety distance required for lane change, the lane change is executed directly; otherwise, it is necessary to plan the coordination of surrounding vehicles based on the surrounding real-time environmental information and vehicle movement status to cooperate with the lane-changing vehicle to complete the lane change task.

[0082] When all vehicles in the final lane-changing array have completed their lane-changing, a new round of lane-changing decision calculation is restarted. Figure 3 .

[0083] In this embodiment, the lane change decision triggering model is specifically:

[0084] In actual traffic operation, vehicles arrive randomly, and the lane-changing scenarios and processes are changing all the time. If a vehicle intends to change lanes, it needs to turn on the left or right turn signal to send a lane-changing signal, and then wait for the conditions to be met before changing lanes. However, with the support of algorithms under the vehicle-road cooperative system, the lane-changing needs of vehicles can be obtained in real time. When a vehicle is detected to have a lane-changing need, the system stores the vehicle information in a dynamic dictionary called a "lane-changing array." The dynamic dictionary records information such as the location, speed, and target lane of each lane-changing vehicle. Although traffic flow is constantly changing, the "lane-changing array" can be continuously updated over time to keep it up to date.

[0085] In order to ensure the safety and smoothness of lane changing, it is necessary to calculate and control the lane changing process according to the characteristics of time and space evolution, and design an effective lane changing decision triggering model, that is, to start the "lane changing decision calculation switch". This switch will trigger a series of lane changing feasibility algorithms. The implementation of this process requires real-time acquisition of vehicle status information and perception information of the surrounding environment, such as vehicle speed, vehicle spacing, and driving status of surrounding vehicles. By comprehensively analyzing the calculation results and based on the vehicle's driving route and goals, the optimal lane changing decision can be made.

[0086] The core of the lane change decision trigger model is to determine when the lane change calculation operation is needed and when it needs to be stopped. In practical applications, certain conditions must be met to start the "lane change decision calculation switch", otherwise it may affect traffic flow and stability. At the same time, the lane change decision trigger model also needs to be real-time and accurate, and be able to quickly respond to road traffic changes to ensure safe and smooth traffic. In addition, the efficiency of system resource utilization must also be considered to ensure that the calculation process does not have too much impact on the operation of the entire intelligent transportation system. For the detailed process of the lane change decision trigger model, see Figure 4 .

[0087] (1) When a vehicle is detected to have a lane change requirement, its vehicle number, position, speed, distance from the stop line, and target lane are stored in a dynamic “lane change array”.

[0088] (2) When the "lane change array" is not zero, start the "lane change decision calculation switch". Sort the lane-changing vehicles in descending order of distance from the stop line, calculate the lane-changing impact range of each lane-changing vehicle, and determine the interaction relationship between vehicles to determine the "final lane-changing array" for subsequent lane-changing planning:

[0089] ① If there are no other lane-changing vehicles within the impact range, the lane change can be executed directly after the safety conditions are met, and the lane-changing vehicle numbers and total number N1 in this situation are recorded.

[0090] ② If there are other lane-changing vehicles within the impact range but the lane-changing trajectories between the vehicles do not conflict, they will execute the lane-changing operation separately after the safety conditions are met, and the lane-changing vehicle numbers and total number N2 in this situation will be recorded.

[0091] ③ When there are lane-changing conflicting vehicles with intersecting trajectories within the lane-changing influence range, it is necessary to count the number of lane-changing conflicts for each of them: when the number of conflicts is greater than or equal to 2, the lane-changing conflict is large and the vehicles temporarily keep following; when the number of conflicts between vehicles is equal to 1, it is necessary to calculate the lane-changing priority, select the one with the highest priority to execute the lane-changing, and record the lane-changing vehicle number and total number N3 in this situation.

[0092] The total number of lane-changing vehicles in the "final lane-changing array" is N = N1 + N2 + N3; the final non-lane-changing vehicles include: vehicles with a conflict number greater than or equal to 2 and vehicles with a conflict number of 1 but a smaller priority. The final non-lane-changing vehicles will rejoin the "lane-changing array" after the current round of lane changing is completed.

[0093] (3) Declare a temporary variable "cl_flag" to control the start and end of the lane change calculation. Set the initial value of "cl_flag" to 0. When the "final lane change array" is not zero, set the value of "cl_flag" to N. When "cl_flag" is not 0, execute the vehicle lane change operation. When each vehicle completes the lane change, the value of "cl_flag" is reduced by 1. This process will continue to loop until "cl_flag" is equal to 0 again, that is, all vehicles in the "final lane change array" have completed the lane change, and then a new round of lane change calculation can be started.

[0094] In summary, the "final lane change array" and "lane change decision calculation switch" are important components of the lane change decision model. Combining the real-time vehicle status information and the perception information of the surrounding environment, an efficient and accurate lane change decision trigger model is designed to effectively improve the traffic safety and driving efficiency of lane change vehicles.

[0095] In this embodiment, the lane change influence range is calculated as follows:

[0096] During the lane-changing process, vehicles interact dynamically with surrounding vehicles. Different road conditions and driving speeds will affect the interaction between vehicles. Therefore, the interaction area between vehicles changes dynamically. This dynamic interaction area is called the lane-changing influence range. By consulting literature and combining actual investigations, it is found that the vehicle's lane-changing influence range is mainly determined by the relationship between driving speed, vehicle spacing and safety distance.

[0097] Since the traffic volume at signalized intersections varies greatly at different times, the IDM model, which is well applicable from free flow to fully congested flow, is used to describe the following behavior of lane-changing vehicles. The IDM model takes into account the speed difference between the lane-changing vehicle and the vehicle in front as well as the distance between the two vehicles. Figure 5 For example, vehicle C i，j As the observation vehicle, determine its lane change impact distance.

[0098] The expected distance between the vehicle and the target vehicle in front during the lane change process is obtained by the following formula:

[0099]

[0100] The vehicle lane change affects the distance:

[0101]

[0102] Where: T is the safety time interval (s), generally 1.5s; V i，j is the speed of the jth vehicle in the i-th lane; D safe is the safety space distance that needs to be met; ΔV (i+1,j)-(i,j) is the speed difference between the lane-changing vehicle and the vehicle in front of the target lane (m / s); amax is the maximum acceleration of the lane-changing vehicle (m / s 2 ), generally 4m / s 2 ; b is the comfortable deceleration of the lane-changing vehicle (m / s 2 ), generally 2m / s 2 ; W is the required lateral safety distance, and the lane width is 3.5m.

[0103] During the lane change process, the expected distance between the vehicle and the vehicle in front of the target lane mainly depends on the relative running state between the vehicle and the vehicle in front, and the relative running state is mainly determined by the running state of the vehicle in front. Therefore, it is necessary to analyze the corresponding expected distance according to the different running states of the vehicle in front. As can be seen from formula (2), the expected distance equation contains a balance term D safe +V i，j T, that is, the two vehicles need to maintain at least a minimum distance. It should be noted that when the speed difference between the current vehicle and the lane-changing vehicle exceeds the speed difference between the desired speed and the lane-changing vehicle, if calculated according to the above formula, the lane-changing impact distance planning area will be much larger than the required distance, thereby increasing unnecessary strategy operation costs. Therefore, when the speed of the vehicle in front of the target lane is greater than the desired speed for lane change, formula (2) is changed to

[0104]

[0105] Where: L exp is the lane-changing impact distance of the vehicle in front of the target lane at the expected speed (m); ΔV exp-(i,j) It is the difference between the lane-changing vehicle speed and the expected speed of the preceding vehicle in the target lane (m / s).

[0106] The lane change impact range is a rectangular area with the lane change impact distance as the length and the total lane width as the width. For detailed calculation process, see Figure 6 .

[0107] Under the vehicle-road cooperative condition, the spatial motion state of the vehicle can be updated in real time. The decision module calculates the lane change impact range of each lane-changing vehicle based on the received motion state of the surrounding vehicles and the distance between vehicles, sorted in descending order according to the distance from the stop line. i+1，j In vehicle C i，j Within the lane change influence range, vehicle C i，j Also in vehicle C i+1，j The influence range of lane change means that the influence between vehicles is mutual. If the running state of any vehicle changes, it will trigger the change of the running state of other vehicles in its lane change influence range. If the distance between two lane change vehicles is outside the lane change influence range, they are not considered in the conflict lane change range.

[0108] The lane change conflict vehicle screening is specifically as follows:

[0109] In order to ensure the safety and efficiency of the vehicle lane changing process, after calculating the lane changing influence range of the lane changing vehicle, it is necessary to analyze the relationship between the vehicles in the influence range and screen out the lane changing conflict vehicles. There are mainly three types of conflicting vehicles: ① Lane changing vehicles with intertwined lane changing trajectories; ② Lane changing vehicles with a conflict number greater than or equal to 2; ③ Lane changing vehicles with a conflict number equal to 1 and a small lane changing priority.

[0110] The lane-changing conflict scenario within the lane-changing influence range is simplified to three lanes with two lane-changing vehicles. Figure 7 . Assuming that the lane-changing vehicle is currently in lane 1, if it changes lanes to the left after the lane-changing intention is generated, it may conflict with vehicles changing from 2→1 and 3→2 (vehicles changing from 3→1 need to go to 3→2 first). Assuming that the lane-changing vehicle is currently in lane 2, if it changes lanes to the right after the lane-changing intention is generated, it may conflict with vehicles changing from 1→2; if it changes lanes to the left, it may conflict with vehicles changing from 3→2. Assuming that the lane-changing vehicle is currently in lane 3, if it changes lanes to the right after the lane-changing intention is generated, it may conflict with vehicles changing from 2→3 and 1→2 (vehicles changing from 1→3 need to go to 1→2 first).

[0111] According to the impact range of the vehicle lane change, determine whether there are conflicting cross-lane-changing vehicles. If there are no other lane-changing vehicles within the impact range, determine that they meet the safety conditions and directly execute the lane change; if there are other lane-changing vehicles but there is no conflict between the vehicles, each lane change will be executed after the safety conditions are met; if there are lane-changing conflicting vehicles with intersecting trajectories, it is necessary to count the number of lane-changing conflicts for each: ① When the number of conflicts is greater than or equal to 2, the lane change will have a greater impact on the entrance lane, and the vehicle will not execute the lane change temporarily; ② When the number of conflicts between vehicles is equal to 1, it is necessary to calculate the lane change priority. The detailed calculation process is shown in Figure 8 .

[0112] When the number of conflicts between lane-changing vehicles is 1, in order to reduce the interference of vehicle lane changing on the entrance lane, it is necessary to plan the lane-changing order to avoid conflicts. Based on this, the concept of lane-changing priority is proposed. On the basis of comprehensive consideration of the actual lane-changing behavior, the conditions for real-time information exchange of vehicle-road collaboration are fully utilized, and the influence of factors such as the number of lanes that vehicles need to cross when changing lanes, traffic flow density, and distance from the stop line on the lane-changing priority are comprehensively considered.

[0113] (1) Number of lanes to cross when changing lanes

[0114] After a vehicle has the intention to change lanes, it will send its lane change request, which includes the lane it is in and its target lane. During the vehicle's lane change process, the more lanes it needs to cross, the greater the disturbance to the traffic flow. And the traffic safety law clearly stipulates that continuous lane changes are prohibited. Therefore, vehicles crossing multiple lanes need to change to the adjacent lane first, drive a distance to ensure safety, and then plan the next lane change again, and then gradually change to the ideal lane. This type of lane-changing vehicle requires a longer lane-changing distance. If the vehicle is in the lane-changing process for a long time, it will cause the vehicle behind to slow down, which will cause great disturbance to the traffic and affect the traffic capacity of the entire entrance lane. In order to avoid causing excessive disturbance to the traffic flow, it is necessary to give it a greater priority.

[0115]

[0116] Where: P ln is the lane number influencing factor; Δn is the number of lanes the vehicle needs to cross; n is the total number of lanes at the current position of the vehicle.

[0117] (2) Combined influence of traffic flow density and distance to the stop line

[0118] Under actual road conditions, vehicle density is always in dynamic change. Generally, under low-density conditions, the distance between vehicles in the target lane is large. At this time, no matter how far the vehicle is from the stop line, the impact of lane changing on traffic flow is minimal. When the traffic flow density begins to increase, the traffic operating environment of the original lane cannot meet the driver's demand for the current speed. At this time, if the distance between vehicles in the adjacent lane meets the lane change demand, the vehicle can accelerate after the lane change, and the possibility of lane change increases, but the interference of lane changing on traffic flow also gradually increases; and the closer the vehicle is to the stop line, the greater its lane change expectation, and the increase rate increases with the shortening of the distance. When the vehicle is very close to the solid line segment of the guide lane, lane changing will have a negative impact on the traffic capacity of at least two lanes. When the traffic flow continues to increase, the interaction between vehicles is significantly enhanced, the speed of the original lane and the adjacent lane is reduced, the distance between vehicles is small, lane changing cannot improve the current driving environment, and safety conditions are also difficult to meet. At this time, no matter how far the vehicle is from the stop line, lane changing has a great interference on traffic flow.

[0119] Therefore, when the density is high, in order to meet the mandatory lane change requirements of turning vehicles, they can be guided to change lanes as early as possible to avoid the situation where the closer the vehicle is to the guide lane line, the greater the disturbance of lane change. At the same time, when the distance between vehicles in the target lane is small, the negative impact of vehicle lane change increases, so it is necessary to give it a higher priority to avoid the situation where lane change is delayed due to unsatisfactory distance (the vehicle may drive on the line for a while or even stop and queue to wait for the opportunity to change lanes), further reduce the impact on other moving vehicles, and improve traffic safety and efficiency.

[0120]

[0121] Where: P k are the factors affecting traffic flow density and the distance to the stop line; K, K j is the traffic density and congestion density (pcu / km); α is the traffic density influence coefficient; β is the distance influence coefficient from the stop line; γ is the process parameter; L is the entrance lane length; ΔL is the distance between the lane-changing vehicle and the stop line.

[0122] After determining the influence coefficient, its weight needs to be calibrated to reflect the influence of different traffic parameters on the lane change priority. Existing studies mostly use qualitative analysis methods to assign a specific value to the weight of the influence coefficient. However, qualitative analysis can only roughly represent the relationship between the influence coefficients, and has its scope of application and limitations. Therefore, the maximum likelihood estimation method is used to determine the specific weight values ​​of different influence coefficients, conduct a more comprehensive and systematic weight judgment, and more accurately evaluate the relationship between different influencing factors and lane change priority. The main steps are as follows:

[0123] ① Define the likelihood function that conforms to the data distribution:

[0124]

[0125] Where: L(θ) is the likelihood function of the sample; θ is the unknown parameter.

[0126] The maximum likelihood method is to use θ j (i=1,2,3,…k) find the estimated value of the unknown parameter that maximizes L(θ)

[0127] ②Convert the likelihood function L(θ) into the log-likelihood function and simplify it:

[0128]

[0129] ③ Calculate the partial derivative and set it to 0:

[0130]

[0131] ④ Further simplify and solve the likelihood equation to obtain the final parameters:

[0132] This research method mainly studies lane change guidance under high-density conditions. The traffic flow density changes with the flow rate, which causes the change of vehicle speed. When the traffic density is large, the speed-density relationship is calculated using the logarithmic model proposed by Greenberg:

[0133]

[0134] Where: V is the average speed of the section (km / h); Vm It is the speed when the traffic volume is the largest, that is, the critical speed (km / h).

[0135] According to the relevant theorem of normal distribution theory, if X is a normally distributed random variable, then exp(X) is logarithmically distributed. The mean and variance of the lognormal distribution are

[0136]

[0137] Based on the relevant theorem of normal distribution theory, it can be known that a linear combination of multiple independent normal distributions still satisfies the normal distribution, that is, X1 and X2 are independent of each other and X1~N(E1,σ1 2 ), X2~N(E2,σ2 2 ), αX1+βX2~N(αE1+βE2,α 2 σ1 2 +β 2 σ2 2 ). So the lane-changing priority P kl Obey the following distribution:

[0138] P kl ~N[e 17.745 α+0.207β, (e 52.66 -e 35.49 )α 2 +0.003β 2 ] (12)

[0139] P kl The probability density function of is:

[0140]

[0141] Through the above steps, the values ​​of α and β can be calculated, and the final parameter settings are shown in Table 1.

[0142] Table 1 Main parameter settings for lane change priority calculation

[0143]

[0144] The above indicators can provide analysis and judgment for lane-changing decisions. The difference in influencing factors determines the order in which vehicles change lanes, assists in determining the optimal lane-changing order, and realizes quantitative evaluation of vehicle lane-changing disturbance, operating efficiency, and safety. Through the above analysis of lane-changing priority, it can be concluded that:

[0145]

[0146] Where: P r is the lane changing priority; λ and μ are weight coefficients.

[0147] The state of the lane-changing vehicle determines the values ​​of λ and μ. In the face of different traffic environments, the values ​​are dynamically adjusted according to the lane-changing requirements while ensuring the safety of the vehicle's lane-changing. For vehicles with low traffic density and more lanes to cross, λ is larger and μ is smaller. On the contrary, when the traffic density is high and the lane-changing vehicle is closer to the stop line, λ is smaller and μ is larger.

[0148] Lane changing priority not only reflects the ability of vehicle-to-vehicle communication, but also helps to avoid conflicts and traffic accidents. When multiple vehicles need to change lanes at the same time, in order to minimize the overall disturbance to the entrance lane, it is necessary to evaluate the lane changing priority, coordinate the lane changing order of each lane changing vehicle, avoid conflicts and blockages, and maximize the use of road capacity. However, the lane changing priority evaluation method used in this chapter only represents a reasonable approximation, and a more accurate evaluation of lane changing priority can be further explored in future work.

[0149] The lane change is specifically performed as follows:

[0150] After the above series of lane change feasibility calculations, the vehicle starts the lane change operation. In the traditional vehicle lane change process, the driver mostly evaluates the speed and distance between the front and rear vehicles based on his own experience, which requires consideration of many factors, including the distance between the front and rear vehicles in the target lane and the front and rear vehicles in the original lane. This subjective evaluation method is prone to misjudgment and increases the risk of vehicle conflict. The vehicle-road collaborative technology, through the perception of the road environment and vehicle status, enables real-time information exchange between vehicles, breaking the limitations of the driver's subjective lane change decision. Therefore, the traditional lane change safety distance judgment method is no longer applicable to the vehicle-road collaborative environment. Under the condition of obtaining vehicle interaction information, the motion status of the lane-changing vehicle and the surrounding vehicles should be comprehensively considered, and the lane change strategies for different scenarios should be discussed in a classified manner.

[0151] This research method mainly refers to the vehicle lane change safety distance model in the Gipps model for discussion. In the actual lane change process, the front vehicle in the own lane and the front vehicle in the target lane are generally in a passive state, and their main function is to constrain the lane-changing vehicle. Therefore, the distance between the lane-changing vehicle and the front vehicle in the own lane needs to satisfy Δd at all times. (i，j)-(i+1，j+1) ≥D safe The minimum oblique safety distance to the vehicle in front of the target lane can be converted into

[0152]

[0153]

[0154] Where: D f is the minimum oblique safety distance (m) to be maintained with the vehicle in front in the target lane; θ is the angle between the centerline of the lane-changing vehicle and the centerline of the target lane.

[0155] Therefore, the diagonal distance between the lane-changing vehicle and the vehicle in front of the target lane only needs to be greater than or equal to D f The vehicle behind the lane is a following vehicle, and only needs to calculate the change in the next second and update it according to the displacement calculation formula in the Gipps model. Therefore, the main analysis is the interaction between the lane-changing vehicle and the vehicle behind the target lane, and other vehicles are not discussed for the time being.

[0156] (1) Direct lane change: Determine whether the vehicle can directly execute the lane change without considering whether the lane change process will affect the normal driving of surrounding vehicles. Not all lane change scenarios require lane change cooperation. In order to reduce the operating cost of the strategy, it is necessary to design a lane change interval trigger logic to reduce unnecessary action evaluation. Set the interval between the target lane rear vehicle and the lane change vehicle to exceed D. tg When the lane-changing vehicle is in the same lane as the vehicle behind, there is no need to implement a cooperative strategy. At this time, the lane-changing vehicle can change lanes directly without considering the impact of the lane-changing on the following vehicle.

[0157] D tg =V max T c +D safe (17)

[0158] Where: D tg No lane change coordination is required (m); V max is the road speed limit (m / s); T c is the lane changing time of the vehicle (s).

[0159] (2) Collaborative lane changing: When collaborative lane changing is adopted, the distance between the lane changing vehicle and surrounding vehicles is further analyzed to confirm the feasibility of the lane changing distance. If the distance between the current position and the vehicle behind the target lane does not meet the lane changing requirements, the vehicle behind the target lane should actively respond to the lane changing request by slowing down and widening the gap after receiving the lane changing signal. When the lane changing vehicle determines that the gap can meet the requirements for safe lane changing, it starts to execute the lane changing operation. After the lane changing vehicle successfully enters the target lane, it adjusts its speed in time to adapt to the driving status of the vehicle in front.

[0160] Assume that the lane-changing vehicle C i，j The lane change operation starts at time t0, T c is the lane change completion time; and the rear vehicle C in the target lane i+1，j+1 Always drive in the lane, that is, no longitudinal displacement occurs; to ensure C i，j Safe lane change, should avoid C i+1，j+1 With C i，j When a collision occurs during the lane-changing process, ε(t) is the angle between the centerline of the lane-changing vehicle and the longitudinal direction of the road at time t, then

[0161]

[0162] Assume M is C i+1，j+1 With Ci，j The critical collision point of C i+1，j+1 With C i，j Oblique collision is likely to occur before point M, and its critical state is shown in Fig. 9 .

[0163] Therefore, C i+1，j+1 With C i，j The conditions for avoiding collision are

[0164] X i+1，j+1 (t)≤X i，j (t)-L A cos[ε(t)]-Δd (i，j)-(i+1，j+1) (t)t∈(t0,t c ) (19)

[0165] Among them, L A is the vehicle length; t c Represents C i，j The time taken to travel from the lane change starting position to the critical collision point M. When t ≥ t c When cos[ε(t)] reaches its maximum value, the above formula can be simplified to

[0166] X i+1，j+1 (t)≤X i，j (t)-L A -Δd (i，j)-(i+1，j+1) (t) t∈[ c t,T c ) (20)

[0167] Let C i+1，j+1 With C i，j The distance between B (t)

[0168] D B (t) = X i，j (t)-X i+1，j+1 (t)-L A (twenty one)

[0169] Then C i，j In the lane-changing process, theoretically when t ≥ t c When D B >0, there will be no collision. However, in this case, the calculated safety distance is too small, and it does not take into account that the vehicles still need to follow each other when the lane change is completed. Therefore, it is necessary to consider the C i+1，j+1 With C i，j The safe following distance is D B >D safe

[0170] D B (t) = Xi，j (t)-X i+1，j+1 (t)-L A >D safe t∈[t c ,T c ) (twenty two)

[0171] X i，j (t)-X i+1，j+1 (t)>D safe +L A t∈[t c ,T c ) (twenty three)

[0172] Dynamic and real-time information interaction in the vehicle-road cooperative environment provides accurate and real-time traffic information for vehicles. If current conditions permit, C i，j Active acceleration can be used to avoid rear-end collision and improve lane-changing efficiency; the vehicle behind the target lane actively decelerates to widen the gap. Therefore, the speeds of the lane-changing vehicle and the vehicle behind the target lane at time t are

[0173]

[0174] Among them, V and a are the speed and acceleration of the vehicle respectively; τ is the integral variable.

[0175] Then C i，j With C i+1，j+1 D without any collision B The minimum value of (t) is the minimum safe distance D that needs to be maintained between the lane-changing vehicle and the vehicle behind in the target lane. b for

[0176]

[0177] Among them, d min The safe distance that the rear vehicle needs to maintain from the front vehicle after emergency braking, usually 3 to 7 meters; a m Indicates the maximum deceleration of the following vehicle (m / s 2 ), generally 6~8m / s2; t d is the reaction delay of the intelligent connected car, which is generally 0.2 to 0.3 seconds. This formula is used to constrain the minimum safe distance between the rear vehicle and the front vehicle during cooperative lane change.

[0178] It can be seen from the above formula that the minimum safe distance between the lane-changing vehicle and the vehicle behind the target lane is mainly related to the speed and acceleration of the vehicle behind the target lane.

[0179] In summary, the present invention proposes a vehicle-road cooperative lane changing decision and control method based on potential conflict analysis. This method takes into account the arrival randomness and spatiotemporal evolution characteristics of lane-changing vehicles, analyzes the potential conflicts between vehicles and surrounding vehicles at different lane-changing stages, and constructs sub-models such as lane-changing decision triggering, lane-changing impact range calculation, lane-changing conflict vehicle screening, and lane-changing priority calculation. The safety distance between the lane-changing vehicle and the vehicle behind in the target lane is corrected based on the minimum safe distance model. By dynamically adjusting the lane-changing order of vehicles, lane-changing conflicts can be reduced to a large extent, adapting to different lane-changing scenarios and lane-changing requirements, and reducing the impact of lane-changing on the stability of intersection traffic, thereby improving overall traffic efficiency.

[0180] The above are preferred embodiments of the present invention. Any changes made according to the technical solution of the present invention, as long as the resulting functions do not exceed the scope of the technical solution of the present invention, belong to the protection scope of the present invention.

Claims

1. A vehicle-road cooperative lane-changing decision and control method based on potential conflict analysis, characterized in that: The specific steps include: S1. Collect and process vehicle and surrounding environment information, including vehicle position, vehicle spacing, lane change requirements, intersection signal light status, distance between the vehicle and the intersection stop line, and upstream lane change area; and determine whether the vehicle is in the lane change area. If the vehicle has entered the solid line segment of the lane divider, lane change is prohibited. S2. Start lane-changing decision calculation; obtain the lane-changing demand of the vehicle in real time through detection and calculation of the vehicle-road cooperative system, and record it in the dynamic lane-changing array; Design a lane-changing decision trigger model to determine when to start and stop the lane-changing calculation, and calculate the feasibility of lane-changing vehicles in a dynamic array based on spatiotemporal evolution; S3, calculating the lane change impact range: when the vehicle has a lane change demand, the lane change impact range is calculated based on the received driving speed, vehicle spacing and safety distance of surrounding vehicles; S4, screening of vehicles with lane-changing conflicts; Analyze the relationship between vehicles within the lane-changing range of each lane-changing vehicle to determine whether there is a potential conflict between the lane-changing vehicles; S5. Calculate lane-changing priority; quantify various influencing factors in the lane-changing process of vehicles, so that vehicles with less disturbance to the entrance lane have higher priority, and determine the final lane-changing array and lane-changing order; S6, lane change execution: make the best lane change decision based on the comprehensive calculation results, and execute the lane change according to the decision results while ensuring safety; if the distance between the lane-changing vehicle and the target lane vehicle meets the minimum safety distance required for lane change, directly execute the lane change; Otherwise, it is necessary to plan the coordination mode of surrounding vehicles based on the surrounding real-time environmental information and vehicle motion status to complete the lane changing task in cooperation with the lane changing vehicle; When all vehicles in the final lane-changing array have completed their lane-changing, a new round of lane-changing decision calculation is restarted.

2. The vehicle-road cooperative lane-changing decision and control method based on potential conflict analysis according to claim 1, characterized in that: The lane change decision triggering model performs the following steps: (1) When a vehicle is detected to have a lane change requirement, its vehicle number, position, speed, distance from the stop line, and target lane are stored in a dynamic lane change array; (2) When the lane change array is not zero, the lane change decision calculation switch is started; Sort the lane-changing vehicles in descending order of distance from the stop line, calculate the lane-changing impact range of each lane-changing vehicle, determine the interaction relationship between vehicles, and determine the final lane-changing array for subsequent lane-changing planning: ① If there are no other lane-changing vehicles within the affected range, the lane change is performed after the safety conditions are met, and the lane-changing vehicle number and total number N1 in this case are recorded; ② If there are other lane-changing vehicles within the impact range but the lane-changing trajectories between the vehicles do not conflict, each vehicle will change lanes after the safety conditions are met, and the lane-changing vehicle numbers and total number N2 in this case will be recorded; ③ When there are conflicting vehicles with intersecting trajectories within the lane change influence range, the number of lane change conflicts between them needs to be counted: when the number of conflicts is greater than or equal to 2, the lane change conflict is large and the vehicles will temporarily keep following; when the number of conflicts between vehicles is equal to 1, the lane change priority needs to be calculated, and the vehicle with the highest priority is selected to perform the lane change, and the lane change vehicle number and total number N3 in this case are recorded; Then the total number of lane-changing vehicles in the final lane-changing array is N=N1+N2+N3; the final non-lane-changing vehicles include: vehicles with a conflict number greater than or equal to 2 and vehicles with a conflict number of 1 but a smaller priority; Finally, the non-lane-changing vehicle will rejoin the lane-changing array after the current lane-changing round is completed; (3) Declare a temporary variable cl_flag to control the start and end of the lane change calculation; set the initial value of cl_flag to 0. When the final lane change array is not zero, set the value of cl_flag to N. When cl_flag is not 0, execute the vehicle lane change operation; when each vehicle completes the lane change, reduce the value of cl_flag by 1. This process will continue to loop until cl_flag is equal to 0 again, that is, all vehicles in the final lane change array have completed the lane change, and then a new round of lane change calculation can be started.

3. The vehicle-road cooperative lane changing decision and control method based on potential conflict analysis according to claim 1, characterized in that: The lane change impact range is a rectangular area with the lane change impact distance as the length and the total lane width as the width. The lane change impact distance is calculated as follows: The expected distance between the vehicle and the target vehicle in front during the lane change process is obtained by the following formula: The distance affected by vehicle lane change is: Where: T is the safety time interval; V i，j is the speed of the jth vehicle in the i-th lane; D safe is the safety space distance that needs to be met; ΔV (i+1,j)-(i,j) is the speed difference between the lane-changing vehicle and the vehicle in front of the target lane; a max is the maximum acceleration of the lane-changing vehicle; b is the comfortable deceleration of the lane-changing vehicle; W is the required horizontal safety distance; When the speed of the vehicle in front of the target lane is greater than the expected lane-changing speed, the calculation of the vehicle lane-changing impact distance is changed from equation (2) to equation (3): Where: L exp ΔV is the lane-changing impact distance of the vehicle in front of the target lane at the expected speed; exp-(i,j) is the difference between the lane-changing vehicle speed and the expected speed of the preceding vehicle in the target lane.

4. The vehicle-road cooperative lane-changing decision and control method based on potential conflict analysis according to claim 3 is characterized in that: The required lateral safety distance is the lane width.

5. The vehicle-road cooperative lane changing decision and control method based on potential conflict analysis according to claim 1, characterized in that: The lane-changing conflict vehicle screening is specifically as follows: according to the vehicle lane-changing impact range, determine whether there are conflicting cross-lane-changing vehicles; if there are no other lane-changing vehicles within the impact range, determine that they meet the safety conditions and directly execute the lane change; if there are other lane-changing vehicles but there is no conflict between the vehicles, determine that they meet the safety conditions and each executes the lane change; if there are lane-changing conflict vehicles with intersecting trajectories, it is necessary to count their respective lane-changing conflict numbers: ① When the conflict numbers are all greater than or equal to 2, the lane change has a greater interference on the entrance lane and the vehicle does not execute the lane change temporarily; ② When the conflict numbers between vehicles are all equal to 1, it is necessary to calculate the lane-changing priority, and the lane change is executed with a larger priority.

6. The vehicle-road cooperative lane-changing decision and control method based on potential conflict analysis according to claim 1, characterized in that: The lane changing priority comprehensively considers the influence of factors such as the number of lanes that a vehicle needs to cross when changing lanes, the traffic flow density, and the distance from the stop line on the lane changing priority.

7. The vehicle-road cooperative lane-changing decision and control method based on potential conflict analysis according to claim 6 is characterized in that: The lane change priority is calculated as follows: Where P r is the lane-changing priority; λ, μ are weight coefficients; P ln is the influencing factor of the number of lanes; P kl is the influencing factor of traffic flow density and distance to the stop line; In the formula, Δn is the number of lanes that the vehicle needs to cross; n is the total number of lanes where the vehicle is currently located; In the formula, K, K j is the traffic density and blocking density; α is the influence coefficient of traffic density; β is the influence coefficient of the distance to the stop line; γ is the process parameter; L is the length of the entrance lane; ΔL is the distance between the lane-changing vehicle and the stop line.

8. The vehicle-road cooperative lane-changing decision and control method based on potential conflict analysis according to claim 7 is characterized in that: The traffic flow density influence coefficient α is set to 3.597, the stop line distance influence coefficient β is set to 0.104, and the process parameter γ is set to 1.

9. The vehicle-road cooperative lane-changing decision and control method based on potential conflict analysis according to claim 1, characterized in that: If the distance between the lane-changing vehicle and the target lane vehicle meets the minimum safety distance required for lane changing, the lane changing is performed directly; otherwise, it is necessary to plan the coordination mode of surrounding vehicles based on the surrounding real-time environmental information and vehicle motion status to cooperate with the lane-changing vehicle to complete the lane changing task; Specifically: The distance between the lane-changing vehicle and the vehicle in front of it must always satisfy Δd (i，j)-(i+1，j+1) ≥D safe ; The lane-changing vehicle and the vehicle in front of the target lane need to maintain a minimum diagonal safety distance D f : Set the time when the distance between the vehicle behind the target lane and the lane-changing vehicle exceeds D tg When the lane-changing vehicle is 0, there is no need to implement a cooperative strategy. The lane-changing vehicle changes lanes directly without considering the impact of the lane-changing vehicle on the following vehicle. D tg =V max T c +D safe (17) Where: D tg No lane change cooperation interval is required; V max is the road speed limit; T c The time for vehicles to change lanes; If the distance between the current position of the lane-changing vehicle and the vehicle behind the target lane does not meet the lane-changing requirements, the vehicle behind the target lane will actively respond to the lane-changing request by slowing down and widening the gap after receiving the lane-changing signal. When the lane-changing vehicle determines that the gap can meet the requirements for safe lane changing, it starts to execute the lane-changing operation.

Images