Lane locking assisted lane changing method based on combination of game lane changing and forced lane changing

Abstract

The present invention relates to a lane locking assisted lane changing method based on the combination of game lane changing and forced lane changing, and belongs to the field of traffic informatization. Aiming at the problem that the control strategy constructed based on game theory can complete the coordinated lane changing of vehicles, but the game benefits are difficult to support efficient lane changing when the traffic is deadlocked or stagnant, a method for combining game lane changing and forced lane changing to command efficient assisted lane changing in a cloud scenario is proposed. Prediction is performed according to the following steps: 1) traffic flow modeling; 2) description of the game lane changing process; 3) analysis of the game lane changing benefits; 4) division of game lane changing and forced lane changing; 5) construction of a combined lane changing model. Considering combining forced lane changing with game lane changing, when the distance between vehicles is too small, switching to a forced lane changing strategy can better improve the lane changing efficiency, thereby effectively reducing energy consumption, emissions and the effect of reducing the queue length upstream of the accident area on the highway.

Description

Technical Field

[0001] The present invention belongs to the field of traffic informatization and relates to a lane locking assisted lane changing method based on the combination of game lane changing and forced lane changing. Background Art

[0002] When a highway incident reduces lanes, traffic upstream of the fault zone becomes congested, and significant speed fluctuations further exacerbate the queue. To ensure safe and efficient highway operation, controlling traffic upstream of the fault zone and incorporating effective management strategies can significantly reduce the negative impact of bottlenecks on highway traffic, lower energy consumption and emissions, and support smart and green transportation initiatives.

[0003] Traffic accidents on highways are characterized by sporadic and sudden occurrences. Comparative experiments with rule-based and game-based lane-changing strategies have revealed their inability to cope with highly variable and uncertain traffic environments. Furthermore, unlike coordinated guidance at intersections, traffic on highways operates in a single direction and lacks signal timing. Controlling lane changes simultaneously reduces vehicle speeds. Therefore, this invention focuses on traffic flow conditions and control strategies upstream of locked lanes at short distances from highway exit ramps. It clarifies that forced lane changes can, under certain conditions, compel vehicles to change their original driving state, making them well-suited to be incorporated into other control strategies as an auxiliary measure. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a lane locking assisted lane changing method based on the combination of game lane changing and forced lane changing. This method conducts comparative experiments on target vehicle lane changing and merging into traffic in different lanes using rule-based and game theory-based control strategies. It is concluded that the control strategy constructed based on game theory can complete vehicle coordinated lane changing, but when the traffic is deadlocked or stagnant, the game benefits are difficult to support efficient lane changing. Game lane changing and forced lane changing are combined, and a forced lane changing model is constructed based on the need for ramp vehicles to leave the highway. The length of the accident impact area and the set of lane changing points are obtained, realizing the transition from game strategy to forced lane changing. Under the premise of ensuring no collision, safe and efficient lane changing of vehicles upstream of the accident area is achieved.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] The present invention specifically comprises the following steps:

[0007] S1: Traffic flow modeling: Build the scenario system and functional framework, analyze the characteristics of the exit ramp, and analyze the queue length of vehicles upstream of the accident area to build multiple mode control methods in the cloud platform.

[0008] S2: Game-based lane-changing process description: Describe the vehicle behavior based on the game theory, and derive the lane-changing process accordingly.

[0009] S3: Game lane-changing benefit analysis: Based on the lane-changing process of S2, the benefit moment and lane-changing benefit function are obtained.

[0010] S: Divide the game lane changing and forced lane changing stages: According to the exit stage division method mentioned in S1, the exit process of forced lane changing is divided into two stages.

[0011] S5: Construct a combined lane-changing model: Construct a game-forced combined lane-changing model for ramp vehicles and main road vehicles in two driving situations: with and without vehicles ahead, to complete lane-changing optimization.

[0012] S1 specifically involves traffic flow modeling. A highway multi-vehicle collaborative scenario consists of at least a roadside unit (RSU), an onboard unit (OVU), and cloud-based devices. In this scenario, the IoV cloud is mapped to the cyberspace in real time, leveraging both physical and informational components. It then analyzes and makes decisions about the scenario from a holistic perspective in the cyberspace, returning the resulting control commands to the physical space to control vehicles and other entities.

[0013] The multi-vehicle collaborative scenario function on highways consists of at least several parts, including vehicles entering the highway, queuing, leaving the highway, and accident handling.

[0014] Single-lane exit ramps are designed in a direct style, while dual-lane exit ramps are designed in a parallel style. When the traffic volume exiting the highway is large, in order to ensure smooth operation of vehicles, reasonable diversion and continuity of lane connections, an auxiliary lane will be added between the right lane and the exit ramp for vehicles to change lanes.

[0015] The traffic diversion process of the highway is shown in Table 1 and Figure 1 shown.

[0016] Analyze the changes in traffic flow after the accident:

[0017] Assume that the maximum traffic volume allowed on the highway section is Q, and the actual traffic volume on the left is βQ, where 0<β<1.

[0018] At a certain moment, an accident occurred on a section of the right side of the highway. The reference dividing surfaces were Section A and Section B, which resulted in some lanes being blocked and traffic unable to pass.

[0019] At this time, the allowed traffic volume of the downstream section of the accident becomes αQ, where 0<α<β<1.

[0020] Since αQ<βQ, the remaining traffic capacity downstream cannot meet the original traffic demand upstream of the traffic accident, so congestion and queuing will occur on this section of road.

[0021] Use queuing theory to analyze the queue length caused by traffic accidents and other reasons:

[0022] When a sudden accident such as lane locking occurs t time later, let q a is the original traffic demand, q b is the remaining traffic volume of the road section, q c is the road traffic volume after the accident disappears. a With q b The difference L(T) between them is the change of queue length over time. The specific calculation method is as follows:

[0023]

[0024] Among them, T x is the total time of the accident, which is the time from the lane blocking to the lane unblocking, t≤T x and t>T x They are the judgment signs before and after the accident is resolved. Taking the queue diffusion speed as the rating indicator, the average diffusion speed of the vehicle queue upstream of the accident area can be calculated as follows:

[0025]

[0026] S2: Specifically: Game description of lane-changing process, using complete information game to describe the lane-changing behavior between vehicles in the Internet of Vehicles environment, and this game is a complete information static non-cooperative game. The vehicle behavior description based on the game idea is as follows:

[0027] The target lane-changing vehicle and vehicles in other lanes correspond to participant elements. The target vehicle (whether to change lanes) and other vehicles (whether to change speed or lanes) correspond to strategy set elements. Position, speed, and acceleration correspond to information. A successful lane change corresponds to a payoff. Stable driving state corresponds to equilibrium.

[0028] Based on this, the lateral lane change process is implemented (such as Figure 2 ): When a vehicle requests a lane change, it calculates the benefits of changing lanes based on information collected from other vehicles and, based on the game results, determines whether it can successfully change lanes. If the lane change request is rejected or the benefits are lower than the benefits of not changing lanes, it adjusts its driving state, i.e., changes its speed, and then collects and compares the driving states of other vehicles multiple times until it can successfully change lanes.

[0029] S3 specifically involves analyzing the benefits of lane changes. On the road, vehicle safety is the primary consideration. Therefore, when vehicle M considers changing lanes, it must determine whether the distance between M and other vehicles meets the safety distance requirement. The decision to change lanes is then made based on the safety benefits. If the safety benefits and decision-making benefits are considered equivalent, the safety benefits are as follows:

[0030]

[0031] Among them, S(t) is the distance between the vehicle and the surrounding vehicles in the current state, S min (t) is the minimum safe driving distance.

[0032] Taking a one-way three-lane straight road as the background, common lane-changing situations can be divided into the following three types:

[0033] Driving situation 1: There is no obvious interaction between vehicles: Vehicle M changes lanes to an empty lane (no competition), and there is no need to write a payoff matrix.

[0034] Driving scenario 2: There is a potential conflict between vehicles: Vehicle M changes lanes to a lane without vehicles (there is a competitive relationship), forming a payoff matrix as shown in Table 1.

[0035] Table 1 Profit Matrix 1

[0036]

[0037] Driving situation 3: The vehicle merges into the traffic flow of the other lane: When the speed of vehicle A behind the right is less than the speed of vehicle M, the profit matrix 2 is obtained as shown in Table 2.

[0038] Table 2 Profit Matrix II

[0039]

[0040] When the speed of vehicle A is greater than the speed of vehicle M, the profit matrix 3 is obtained as shown in Table 3.

[0041] Table 3 Profit Matrix III

[0042]

[0043] According to the current information of each vehicle, the comparison is made in the rules. If the equation (4) is satisfied, it is determined that the participant has the intention to change lanes:

[0044]

[0045] Where Δh i is the distance of the target vehicle to the detour point ahead, is the expected speed of the target vehicle, is the current speed of the target vehicle, v baris the moving speed of the point to be avoided, v i-1 is the current speed of the preceding vehicle, T safe is the safety time interval, T min is the minimum reaction time. Since the fault area is a fixed location, v bar =0.

[0046] Based on the three lane-changing scenarios provided above, it is found that vehicles are very likely to collide obliquely during lane changes due to speed and distance differences. To avoid collisions caused by insufficient distance, the following inequality is constructed:

[0047]

[0048] Where x is the horizontal coordinate difference, ω is a sensitive parameter, and θ is the angle between the vehicle's target turning direction and the horizontal lane. Let the vehicle spacing h satisfy equations (6) and (7):

[0049]

[0050]

[0051] Where a is the acceleration of the vehicle along the road. To ensure that no collision occurs, equation (8) must be satisfied:

[0052]

[0053] The minimum safe distance and Calculate according to formula (9) and (10):

[0054]

[0055]

[0056] The time benefit is formula (11):

[0057]

[0058] Among them, t is the time required to reach the target point after changing the strategy, and t0 is the time required to reach the target point while maintaining the original state.

[0059] According to the minimum safety distance S between vehicles min The benefit function W is designed based on the safety benefit (Equation (3)) and the time benefit (Equation (11)). Different weights are assigned to each of them to ensure that the driver, after weighing safety and timeliness, chooses a reasonable lane-changing behavior and reaches the destination as quickly as possible. Therefore, the benefit function shown in Equation (12) is obtained:

[0060]

[0061] where w i is the weight value, representing the driver's driving style, and the value is non-negative. The driver's personality will affect his decision-making, and thus affect the driving results. Therefore, we discard the two extreme personality cases and set the weight to a moderate value. Figure 3 The idea of ​​constructing the game profit algorithm based on the above formula is shown in Table 4.

[0062] Table 4 Game payoff algorithm

[0063]

[0064] S4 is specifically divided into the game lane changing and forced lane changing stages, Figure 4 and Figure 5 These are two typical forced lane change scenarios. Based on the exit phase division method mentioned in S1, the forced lane change process is divided into two phases. As shown in Table 5, when vehicle M is located on section E1, the surrounding environment is favorable, and there is ample space for lane changes. Therefore, a lane change strategy based on game theory is implemented. As driving time and location continue to advance, vehicle M approaches the accident blockade area, so this section is classified as E2, the forced lane change phase.

[0065] Table 5 Exit process of forced lane change

[0066]

[0067] S5 specifically involves building a combined lane-changing model. Based on the two stages divided in S4, a game-forced combined lane-changing model is built for ramp vehicles and main road vehicles under two driving conditions: with and without vehicles ahead.

[0068] In the Internet of Vehicles environment, the characteristic of the forced lane changing process of multi-vehicle cooperation is that the vehicle performing the lane changing behavior and the surrounding vehicles maintain continuous communication.

[0069] After the accident occurs, based on the information transmission of the Internet of Vehicles system, the cloud can obtain the accident time t0 in time, and after Δt, the blocked area is released and the location of vehicle i [x i (t0),y i (t0)], speed v i (t0), the lane change optimization is completed with the goal of minimizing the sum of lane change distances. Figure 5 When shown, the length of the upstream impact area of ​​the blockade area satisfies the following formula:

[0070] S=0.5*W(t f -t0) (13)

[0071] Where W is the velocity of the shock wave diffusion, which is calculated as shown in formula (2).

[0072] The set of lane change points when there is no following is as follows:

[0073] a. The set of lane-changing points for ramp vehicles. Figure 5 The forced lane change situation shown in the figure is that the vehicle exiting the highway needs to change lanes to the right lane. If there is no vehicle in front at the current moment, then vehicle i chooses the lane change point G to change lanes, and at T i The lane change is completed at time t, and after the lane change is completed, the front and rear vehicles of vehicle i in the target lane are P g and P g+1 The feasible lane-changing points of vehicle i belong to the lane-changing point set, denoted as r iG ∈R iG , as follows:

[0074] r iG =x i (t0)+v i (t0)(T i0 -t0) (14)

[0075] The following formula must be met:

[0076]

[0077] Where r is the horizontal coordinate of the lane-changing point G of vehicle i, t di is the time required for vehicle i to complete the entire lane-changing process, Δx is the horizontal distance change during the lane-changing process, and S min is the minimum safe distance between two workshops. di and Δx i Satisfy equations (16) and (17) respectively:

[0078]

[0079]

[0080] After vehicle i is affected upstream of the blockade area, it satisfies the following equation:

[0081]

[0082] b. Lane-changing point set of main road vehicles. Let the main road vehicle be represented as vehicle j and the lane-changing point be represented as H. Similarly, the lane-changing point set r can be obtained jH ∈R jH When there are lane-changing points in both the left and right lanes on the main road, vehicles on the main road should give priority to turning into the left lane.

[0083] The set of lane change points when there is a following situation is as follows:

[0084] a. Lane change point set for ramp vehicles. There is vehicle i-1 in front of vehicle i, that is, vehicle i follows vehicle i-1. There are multiple lane change points for the target vehicle to choose from, and there is a lane change point r' iG ∈R' iG , lane change point r (i-1)U ∈R (i-1)U . It is necessary to satisfy that after changing lanes, vehicle i and vehicle i-1 are at the same P g and P g+1 . And the distance between lane-changing points should be greater than S min . The following formula can be obtained:

[0085]

[0086] Where r' is the horizontal coordinate of the lane-changing point G of vehicle i.

[0087] b. Lane-changing point set of main road vehicles. Denoting the main road vehicle as vehicle j, we can get the lane-changing point set r' jH ∈R' jH When there are lane-changing points in both the left and right lanes for vehicles on the main road, it is preferred to turn to the left lane.

[0088] Select multiple vehicles that perform forced lane changes and combine their lane change points. That is, use the lane change point as the decision variable for this lane change decision. Combine the lane change points of multiple vehicles to determine the optimal lane change point combination at the current moment, which is the following formula:

[0089]

[0090] The beneficial effect of this invention is that it addresses the problem in highway emergency situations where driving behavior based on the game payoff matrix fails to meet actual needs, potentially leading to new rounds of congestion and queues, and the entire traffic flow once again falling into a deadlock. Consider combining forced lane changes with game lane changes, switching to a forced lane change strategy when the distance between vehicles becomes too close.

[0091] Other advantages, objects, and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art upon examination of the following description or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings, in which:

[0093] Figure 1 Schematic diagram of the diversion area;

[0094] Figure 2 Implementation process diagram for lane-changing behavior;

[0095] Figure 3 This is a diagram of the lane-changing process based on game theory;

[0096] Figure 4 For forced lane change scenarios (left lane closed);

[0097] Figure 5 For forced lane change scenarios (middle lane closed);

[0098] Figure 6 Implement algorithm flow chart for forced lane change;

[0099] Figure 7 (a) is the total fuel consumption diagram of each method when the traffic volume is 3000 vehicles / h; (b) is the total fuel consumption diagram of each method when the traffic volume is 4000 vehicles / h; (c) is the total fuel consumption diagram of each method when the traffic volume is 5000 vehicles / h; (d) is the total fuel consumption diagram of each method when the traffic volume is 6000 vehicles / h; (e) is the total fuel consumption diagram of each method when the traffic volume is 7000 vehicles / h; (f) is the total fuel consumption diagram of each method when the traffic volume is 8000 vehicles / h; (g) is the total fuel consumption diagram of each method when the traffic volume is 3000 vehicles / h (h) is the total carbon dioxide emissions of each method when the traffic volume is 4000 vehicles / h; (i) is the total carbon dioxide emissions of each method when the traffic volume is 5000 vehicles / h; (j) is the total carbon dioxide emissions of each method when the traffic volume is 6000 vehicles / h; (k) is the total carbon dioxide emissions of each method when the traffic volume is 7000 vehicles / h; (l) is the total carbon dioxide emissions of each method when the traffic volume is 8000 vehicles / h;

[0100] Figure 8(a) is the queue length diagram of each method in the left lane when the traffic volume is 3000 vehicles / h; (b) is the queue length diagram of each method in the middle lane when the traffic volume is 3000 vehicles / h; (c) is the queue length diagram of each method in the right lane when the traffic volume is 3000 vehicles / h; (d) is the queue length diagram of each method in the left lane when the traffic volume is 4000 vehicles / h; (e) is the queue length diagram of each method in the middle lane when the traffic volume is 4000 vehicles / h; (f) is the queue length diagram of each method in the right lane when the traffic volume is 4000 vehicles / h; (g) is the queue length diagram of each method in the left lane when the traffic volume is 5000 vehicles / h; (h) is the queue length diagram of each method in the middle lane when the traffic volume is 5000 vehicles / h; (i) is the queue length diagram of each method in the right lane when the traffic volume is 5000 vehicles / h ; (j) is the queue length diagram of each method in the left lane when the traffic volume is 6000 vehicles / h; (k) is the queue length diagram of each method in the middle lane when the traffic volume is 6000 vehicles / h; (l) is the queue length diagram of each method in the right lane when the traffic volume is 6000 vehicles / h; (m) is the queue length diagram of each method in the left lane when the traffic volume is 7000 vehicles / h; (n) is the queue length diagram of each method in the middle lane when the traffic volume is 7000 vehicles / h; (o) is the queue length diagram of each method in the right lane when the traffic volume is 7000 vehicles / h; (p) is the queue length diagram of each method in the left lane when the traffic volume is 8000 vehicles / h; (q) is the queue length diagram of each method in the middle lane when the traffic volume is 8000 vehicles / h; (r) is the queue length diagram of each method in the right lane when the traffic volume is 8000 vehicles / h;

[0101] Figure 9 (a) is the total fuel consumption of each method when the traffic volume is 3000 vehicles / h (the middle lane and the left lane are closed); (b) is the total carbon dioxide emission of each method when the traffic volume is 3000 vehicles / h (the middle lane and the left lane are closed); (c) is the box plot of carbon dioxide emission of each method when the traffic volume is 3000 vehicles / h (the middle lane and the left lane are closed);

[0102] Figure 10 (a) shows the queue lengths of various methods in the left lane when the traffic volume is 3000 vehicles / h (the middle lane and the left lane are closed); (b) shows the queue lengths of various methods in the middle lane when the traffic volume is 3000 vehicles / h (the middle lane and the left lane are closed); (c) shows the queue lengths of various methods in the right lane when the traffic volume is 3000 vehicles / h (the middle lane and the left lane are closed). DETAILED DESCRIPTION

[0103] The following describes the embodiments of the present invention by means of specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0104] Among them, the accompanying drawings are only for illustrative purposes and represent only schematic diagrams rather than actual pictures, and should not be understood as limiting the present invention. In order to better illustrate the embodiments of the present invention, some parts of the accompanying drawings may be omitted, enlarged or reduced, and do not represent the dimensions of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions may be omitted in the accompanying drawings.

[0105] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", "front", "back", etc. indicating directions or positional relationships, they are based on the directions or positional relationships shown in the drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operate in a specific direction. Therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0106] The implementation process of the present invention includes:

[0107] 1) Traffic Flow Modeling: Build a scenario system and functional framework, analyze exit ramp characteristics, and analyze the length of vehicle queues upstream of the accident area to build multiple mode control methods on the cloud platform.

[0108] 2) Game-based lane-changing process description: Describe the vehicle behavior based on game theory and derive the lane-changing process accordingly.

[0109] 3) Game lane-changing benefit analysis: Based on the lane-changing process in 2), the benefit moment and lane-changing benefit function are derived.

[0110] 4) Dividing the game lane changing and forced lane changing stages: According to the exit stage division method mentioned in 1), the exit process of forced lane changing is divided into two stages.

[0111] 5) Construct a combined lane-changing model: A game-forced combined lane-changing model is constructed for ramp vehicles and main road vehicles under two driving conditions: one with no vehicle ahead and the other with a vehicle ahead, to achieve lane-changing optimization.

[0112] experiment

[0113] This example uses SUMO and Python co-simulation to combine forced lane changes with game-based lane changes. When vehicles are too close together to change through game-based lane changes, forced lane changes are used to control vehicles upstream of the blocked area. The road node coordinates are set with (0,0) as the origin. The length of the main road E0 is 3000 meters, the length of the merging area road E1 is 300 meters, the length of the main road E2 is 300 meters, and the length of the exit ramp E4 is 225 meters. E0 and E2 are three-lane one-way roads, E1 is a three-lane road with an auxiliary lane, and E3 is a single-lane exit ramp. The speed limit on the main road is 100 km / h (approximately 27.78 m / s), the speed limit on the ramp is fixed at 40 km / h (approximately 11.11 m / s), and the road priority is set to the default value of -1. The initial traffic volume is set to 3000 vehicles per hour, with a vehicle allocation ratio of 7:3 between the main road traffic (main flow) and the exit ramp traffic (offramp). The traffic flow values ​​were set between 3000 and 8000, with a delta of 1000. The following model type for all vehicles was set to CACC, with a random appearance. The main road traffic route was E0-E1-E2, colored green; the exit ramp traffic route was E0-E1-E3, colored red. The detector type was set to e1, located 2 meters from the entrance of each lane, and the detection frequency was set to 900. The simulation started at 0, with vehicles free to travel on the road for the first 300 seconds. After 300 seconds, the lane on road E1 was closed, and after 2400 seconds, the fault zone was unblocked. The traffic density was set between 3000 and 8000 vehicles per hour. Lane closures varied from only the middle lane to both the middle and upper lanes. Experiments were conducted using an uncontrolled approach, a game-based lane-changing strategy, and a game-based forced lane-changing strategy.

[0114] 1) Traffic flow modeling: The structure of highway multi-vehicle cooperative scene includes at least roadside unit, vehicle unit and cloud equipment. In this scene, relying on physical structure and information structure,

[0115] The traffic diversion process of the expressway is shown in Table 6 and Figure 1 shown.

[0116] Table 6 Traffic diversion process of highway

[0117]

[0118] Analyze the changes in traffic flow after the accident:

[0119] Suppose the maximum traffic volume allowed on the highway section is Q, and the actual traffic volume on the left is βQ, where 0<β<1.

[0120] At a certain moment, an accident occurred on a section of the right side of the highway. The reference dividing surfaces were Section A and Section B, which resulted in some lanes being blocked and traffic unable to pass.

[0121] At this time, the allowed traffic volume of the downstream section of the accident becomes αQ, where 0<α<β<1.

[0122] Since αQ<βQ, the remaining traffic capacity downstream cannot meet the original traffic demand upstream of the traffic accident, so congestion and queuing will occur on this section of road.

[0123] Use queuing theory to analyze the queue length caused by traffic accidents and other reasons:

[0124] When a sudden accident such as lane locking occurs t time later, let q a is the original traffic demand, q b is the remaining traffic volume of the road section, q c is the road traffic volume after the accident disappears. a With q b The difference L(T) between them is the change of queue length over time. The specific calculation method is as follows:

[0125]

[0126] Among them, T x is the total time of the accident, which is the time from the lane blocking to the lane unblocking, t≤T x and t>T x They are the judgment signs before and after the accident is resolved. Taking the queue diffusion speed as the rating indicator, the average diffusion speed of the vehicle queue upstream of the accident area can be calculated as follows:

[0127]

[0128] 2) Game-based lane-changing process description: A complete information game is used to describe the lane-changing behavior between vehicles in the connected vehicle environment. This game is a complete information static non-cooperative game. The vehicle behavior description based on the game idea is as follows:

[0129] The target lane-changing vehicle and vehicles in other lanes correspond to participant elements. The target vehicle (whether to change lanes) and other vehicles (whether to change speed or lanes) correspond to strategy set elements. Position, speed, and acceleration correspond to information. A successful lane change corresponds to a payoff. Stable driving state corresponds to equilibrium.

[0130] Based on this, the lateral lane change process is implemented (such as Figure 2 ): When a vehicle requests a lane change, it calculates the benefits of changing lanes based on information collected from other vehicles and, based on the game results, determines whether it can successfully change lanes. If the lane change request is rejected or the benefits are lower than the benefits of not changing lanes, it adjusts its driving state, i.e., changes its speed, and then collects and compares the driving states of other vehicles multiple times until it can successfully change lanes.

[0131] 3) Lane-Changing Benefit Analysis: On the road, vehicle safety is the primary consideration. Therefore, when vehicle M considers changing lanes, it must determine whether the distance between M and other vehicles meets the safety distance requirement. The decision to change lanes is then made based on the safety benefit. If the safety benefit and decision-making benefit are considered equivalent, the safety benefit is as follows:

[0132]

[0133] Among them, S(t) is the distance between the vehicle and the surrounding vehicles in the current state, S min (t) is the minimum safe driving distance.

[0134] Taking a one-way three-lane straight road as the background, common lane-changing situations can be divided into the following three types:

[0135] Driving situation 1: There is no obvious interaction between vehicles: Vehicle M changes lanes to an empty lane (no competition), and there is no need to write a payoff matrix.

[0136] Driving situation 2: There is a potential conflict between vehicles: Vehicle M changes lanes to a lane without vehicles (there is a competitive relationship), and the payoff matrix is ​​shown in Table 2.

[0137] Driving situation 3: The vehicle merges into the traffic flow of the different lane: When the speed of vehicle A behind the right is less than that of vehicle M, the profit matrix is ​​as shown in Table 3; when the speed of vehicle A is greater than that of vehicle M, the profit matrix is ​​as shown in Table 4.

[0138] According to the current information of each vehicle, the comparison is made in the rules. If the equation (4) is satisfied, it is determined that the participant has the intention to change lanes:

[0139]

[0140] Where Δh i is the distance of the target vehicle to the detour point ahead, is the expected speed of the target vehicle, is the current speed of the target vehicle, v bar is the moving speed of the point to be avoided, v i-1 is the current speed of the preceding vehicle, T safe is the safety time interval, T minis the minimum reaction time. Since the fault area is a fixed location, v bar =0.

[0141] Based on the three lane-changing scenarios provided above, it is found that vehicles are very likely to collide obliquely during lane changes due to speed and distance differences. To avoid collisions caused by insufficient distance, the following inequality is constructed:

[0142]

[0143] Where x is the horizontal coordinate difference, ω is a sensitive parameter, and θ is the angle between the vehicle's target turning direction and the horizontal lane. Let the vehicle spacing h satisfy equations (6) and (7):

[0144]

[0145]

[0146] Where a is the acceleration of the vehicle along the road. To ensure that no collision occurs, equation (8) must be satisfied:

[0147]

[0148] The minimum safe distance and Calculate according to formula (9) and (10):

[0149]

[0150]

[0151] The time benefit is formula (11):

[0152]

[0153] Among them, t is the time required to reach the target point after changing the strategy, and t0 is the time required to reach the target point while maintaining the original state.

[0154] According to the minimum safety distance S between vehicles min The benefit function W is designed based on the safety benefit (Equation (3)) and the time benefit (Equation (11)). Different weights are assigned to each of them to ensure that the driver, after weighing safety and timeliness, chooses a reasonable lane-changing behavior and reaches the destination as quickly as possible. Therefore, the benefit function shown in Equation (12) is obtained:

[0155]

[0156] where w iis the weight value, representing the driver's driving style, and the value is non-negative. The driver's personality will affect his decision-making, and thus affect the driving results. Therefore, we discard the two extreme personality cases and set the weight to a moderate value. Figure 3 The above formula is used to construct the game profit algorithm.

[0157] 4) Divide the game lane change and forced lane change stages: Figure 4 and Figure 5 These are two typical forced lane change scenarios. Based on the exit phase classification method mentioned in 1), the forced lane change process is divided into two phases. As shown in Table 5, when vehicle M is located on section E1, the surrounding environment is favorable, and there is ample space for lane changes. Therefore, a game-based lane change strategy is implemented. As driving time and location progress, vehicle M approaches the accident blockade, so this section is classified as E2, the forced lane change phase.

[0158] 5) Constructing a combined lane-changing model: Based on the two phases described in 4), a game-forced combined lane-changing model is constructed for ramp vehicles and main road vehicles under two driving conditions: one with no vehicle ahead and the other with a vehicle ahead.

[0159] In the Internet of Vehicles environment, the characteristic of the forced lane changing process of multi-vehicle cooperation is that the vehicle performing the lane changing behavior and the surrounding vehicles maintain continuous communication.

[0160] After the accident occurs, based on the information transmission of the Internet of Vehicles system, the cloud can obtain the accident time t0 in time, and after Δt, the blocked area is released and the location of vehicle i [x i (t0),y i (t0)], speed v i (t0), with the goal of minimizing the sum of lane change distances, the lane change optimization is completed. Figure 5 When shown, the length of the upstream impact area of ​​the blockade area satisfies the following formula:

[0161] S=0.5*W(t f -t0) (13)

[0162] Where W is the velocity of the shock wave diffusion, which is calculated as shown in formula (2).

[0163] The set of lane change points when there is no following is as follows:

[0164] c. The set of lane-changing points for ramp vehicles. Figure 5 The forced lane change situation shown in the figure is that the vehicle exiting the highway needs to change lanes to the right lane. If there is no vehicle in front at the current moment, then vehicle i chooses the lane change point G to change lanes, and at T iThe lane change is completed at time t, and after the lane change is completed, the front and rear vehicles of vehicle i in the target lane are P g and P g+1 The feasible lane-changing points of vehicle i belong to the lane-changing point set, denoted as r iG ∈R iG , as follows:

[0165] r iG =x i (t0)+v i (t0)(T i0 -t0) (14)

[0166] The following formula must be met:

[0167]

[0168] Where r is the horizontal coordinate of the lane-changing point G of vehicle i, t di is the time required for vehicle i to complete the entire lane-changing process, Δx is the horizontal distance change during the lane-changing process, and S min is the minimum safe distance between two workshops. di and Δx i Satisfy equations (16) and (17) respectively:

[0169]

[0170]

[0171] After vehicle i is affected upstream of the blockade area, it satisfies the following equation:

[0172]

[0173] d. The lane-changing point set of main road vehicles. Let the main road vehicle be represented as vehicle j and the lane-changing point be represented as H. Similarly, the lane-changing point set r can be obtained. jH ∈R jH When there are lane-changing points in both the left and right lanes on the main road, vehicles on the main road should give priority to turning into the left lane.

[0174] The set of lane change points when there is a following situation is as follows:

[0175] c. Lane change point set for ramp vehicles. There is vehicle i-1 in front of vehicle i, that is, vehicle i follows vehicle i-1. There are multiple lane change points for the target vehicle to choose from, and there is a lane change point r' iG ∈R' iG , lane change point r (i-1)U ∈R (i-1)U . It is necessary to satisfy that after changing lanes, vehicle i and vehicle i-1 are at the same P g and P g+1. And the distance between lane-changing points should be greater than S min . The following formula can be obtained:

[0176]

[0177] Where r' is the horizontal coordinate of the lane-changing point G of vehicle i.

[0178] d. Lane-changing point set of main road vehicles. Denoting the main road vehicle as vehicle j, we can get the lane-changing point set r' jH ∈R' jH When there are lane-changing points in both the left and right lanes for vehicles on the main road, it is preferred to turn to the left lane.

[0179] Select multiple vehicles that perform forced lane changes and combine their lane change points. That is, use the lane change point as the decision variable for this lane change decision. Combine the lane change points of multiple vehicles to determine the optimal lane change point combination at the current moment, which is the following formula:

[0180]

[0181] The corresponding algorithm flow chart is as follows Figure 6 shown.

[0182] Close the middle lane E1_2 and compare the experimental results. Figure 7 and Figure 8 shown. Figure 7 (a)~ Figure 7 (l) is the total fuel consumption diagram of each method under different traffic flow. Figure 8 (a)~ Figure 8 (r) is the queue length diagram of each method in the left and right lanes when the traffic volume is different.

[0183] Close the middle lane E1_2 and the left lane E1_3, and compare the experimental results. Figure 9 (a)~ Figure 9 (c) and Figure 10 (a)~ Figure 10 (c) shown.

[0184] Experiments have shown that both the game-based lane-changing strategy and the game-based-forced combined lane-changing strategy have good control effects when closing the middle single lane and the left and middle lanes at the same time. The combined lane-changing strategy prevents and alleviates traffic stagnation and locking, which cannot be solved by using the game-based lane-changing strategy alone. It effectively reduces energy consumption, emissions, and queue lengths upstream of the highway accident area.

[0185] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions, which should all be included in the scope of the claims of the present invention.

Claims

1. A lane-locking assisted lane-changing method based on a combination of game-based lane-changing and forced lane-changing, characterized by: The method comprises the following steps: S1: Traffic flow modeling: Build the scenario system and functional framework, analyze the characteristics of the exit ramp, analyze the length of the vehicle queue upstream of the accident area, and build the mode control method of the cloud platform; S2: Game-based lane-changing process description: Describe the vehicle behavior based on the game theory and derive the lane-changing process; S3: Game lane-changing benefit analysis: Based on the lane-changing process in S2, the benefit matrix and lane-changing benefit function are obtained; S4: Divide the lane-changing process into two phases: The forced lane-changing process is divided into two phases. Specifically, when vehicle M is located on section E1, the surrounding environment is good and there is ample space for lane changes, so a lane-changing strategy based on game theory is implemented. As driving time and location continue to advance, vehicle M gets closer to the accident blockade area, and the section is divided into E2, which is the forced lane-changing phase. S5: Constructing a combined lane-changing model: Constructing a game-forced combined lane-changing model for ramp vehicles and main road vehicles under two driving conditions: no vehicle ahead and vehicle ahead, to complete lane-changing optimization. Specifically, according to the two stages divided in S4, constructing a game-forced combined lane-changing model for ramp vehicles and main road vehicles under two driving conditions: no vehicle ahead and vehicle ahead: After the accident occurs, based on the information transmission of the Internet of Vehicles system, the cloud can obtain the accident time t0 in time, and after Δt, the blocked area is released and the location of vehicle i [x i (t0),y i (t0)], speed v i (t0), with the goal of minimizing the sum of lane-changing distances, the lane-changing optimization is completed; the length of the impact area of ​​the blockade on the upstream satisfies the following formula: S=0.5*W(t f -t0) (13) Where W is the velocity of the shock wave diffusion; The set of lane change points when there is no following is as follows: a. Lane change point set for ramp vehicles; a vehicle exiting the highway on the exit ramp needs to change lanes to the right lane; if there is no vehicle ahead at the current moment, then vehicle i selects lane change point G to change lanes, and at T i The lane change is completed at time t, and after the lane change is completed, the front and rear vehicles of vehicle i in the target lane are P g and P g+1 ; The feasible lane-changing points of vehicle i belong to the lane-changing point set, denoted as r iG ∈R iG , as follows: r iG =x i (t0)+v i (t0)(T i0 -t0) (14) Satisfy the following formula: Among them, r iG is the horizontal coordinate of the lane-changing point G of vehicle i, t di is the time required for vehicle i to complete the entire lane changing process, Δx i is the horizontal distance change during lane changing, S min is the minimum safe distance between two workshops; t di and Δx i Satisfy equations (16) and (17) respectively: After vehicle i is affected upstream of the blockade area, it satisfies the following equation: b. Lane-changing point set of main road vehicles; denote the main road vehicle as vehicle j and the lane-changing point as H, and obtain the lane-changing point set r. jH ∈R jH When there are lane-changing points in both the left and right lanes on the main road, vehicles on the main road will prioritize turning to the left lane. The set of lane change points when there is a following situation is as follows: c. Lane-changing point set for ramp vehicles; there is vehicle i-1 in front of vehicle i, i.e., vehicle i follows vehicle i-1. There are multiple lane-changing points for the target vehicle to choose from, and there is a lane-changing point r′ iG ∈R′ iG , lane change point r (i-1)U ∈R (i-1)U ; It is necessary to satisfy that after changing lanes, vehicle i and vehicle i-1 are at the same P g and P g+1 ; and the distance between lane-changing points should be greater than S min , we get the following formula: Among them, r′ iG is the horizontal coordinate of the lane-changing point G of vehicle i; d. Lane-changing point set of main road vehicles; denote the main road vehicle as vehicle j, and obtain the lane-changing point set r′ jH ∈R′ jH ,When the main road vehicles have lane change points in both the left and right lanes, they give priority to turning to the left lane; Select multiple vehicles that perform forced lane changes and combine their lane change points. That is, use the lane change point as the decision variable for this lane change decision. Combine the lane change points of multiple vehicles to determine the optimal lane change point combination at the current moment, which is the following formula: Lane changing optimization is completed based on the constructed combined lane changing model.

2. The lane locking assisted lane changing method based on a combination of game lane changing and forced lane changing according to claim 1, characterized in that: S1 specifically states that the highway multi-vehicle collaboration scenario consists of a roadside unit, an onboard unit, and cloud devices. In this scenario, the functions of the highway multi-vehicle collaboration scenario include vehicle entry, platoon cruising, exiting the highway, and accident handling. Analyze the changes in traffic flow after the accident: Assume that the maximum traffic volume allowed on the highway section is The actual traffic flow on the left is Where 0<β<1; at a certain moment, an accident occurs on a section of the right side of the highway, with the reference dividing planes being sections A and B, resulting in partial lane closure and traffic being unable to pass; The allowed traffic volume of the downstream section of the accident becomes Where 0<α<β<1; The remaining capacity of the downstream section cannot meet the original traffic demand of the upstream section of the traffic accident, and congestion and queues will occur on the downstream section. The queue length caused by the traffic accident is analyzed using queuing theory: When the lane locking accident occurs t time later, let q a is the original traffic demand, q b is the remaining traffic volume of the road section, q c is the road traffic volume after the accident disappears; q a With q b The difference L(T) between them is the change of queue length over time. The specific calculation method is as follows: Among them, T x is the total time of the accident, which is the time from the beginning of lane blocking to the end of lane blocking, t≤T x and t>T x are the judgment signs before and after the accident is resolved; taking the queue diffusion speed as the rating indicator, the calculation method for the average diffusion speed of the vehicle queue upstream of the accident area is as follows:

3. The lane locking assisted lane changing method based on a combination of game lane changing and forced lane changing according to claim 1, characterized in that: Specifically, S3 is as follows: On the road, vehicle safety is the primary consideration. After vehicle M has the idea of ​​changing lanes, it is necessary to determine whether the distance between vehicle M and other vehicles meets the safety distance requirement. The decision on whether to change lanes is then made based on the safety benefit. If the safety benefit and decision benefit are considered equivalent, the safety benefit is as follows: Among them, S(t) is the distance between the vehicle and the surrounding vehicles in the current state, S min (t) is the minimum safe driving distance. Taking a one-way three-lane straight road as the background, common lane-changing situations are divided into the following three types: Driving situation 1: No obvious interaction between vehicles: Vehicle M changes lanes to a lane without vehicles, and the payoff matrix is ​​not written; Driving situation 2: There is a potential conflict between vehicles: vehicle M changes lanes to the empty lane, forming payoff matrix 1; Driving situation 3: The vehicle merges into the traffic flow of the other lane: When the speed of vehicle A behind the right is less than the speed of vehicle M, the profit matrix 2 is obtained; When the speed of vehicle A is greater than the speed of vehicle M, we get the payoff matrix three; According to the current information of each vehicle, the comparison is made in the rules. If the equation (4) is satisfied, it is determined that the participant has the intention to change lanes: Where Δh i is the distance of the target vehicle to the detour point ahead, is the expected speed of the target vehicle, is the current speed of the target vehicle, v bar is the moving speed of the point to be avoided, v i-1 is the current speed of the preceding vehicle, T safe is the safety time interval, T min is the minimum reaction time. Since the fault area is a fixed location, v bar =0; Based on the three driving scenarios provided above, it is found that vehicles are very likely to have oblique collisions due to speed differences and distance differences during lane changes. To avoid collisions due to insufficient distance, the following inequality is constructed: Where x is the horizontal coordinate difference, ω is a sensitive parameter, and θ is the angle between the vehicle in the target turning direction and the horizontal lane. Let the vehicle spacing h satisfy equations (6) and (7): Where a is the acceleration of the vehicle along the road; to ensure that no collision occurs, it is necessary to satisfy formula (8): The minimum safe distance and Calculate according to formula (9) and (10): The time benefit is formula (11): Among them, t is the time required to reach the target point after changing the strategy, and t'0 is the time required to reach the target point while maintaining the original state. min The profit function W is designed based on the safety benefit proposed in formula (3) and the time benefit proposed in formula (11). Different weights are assigned to the two respectively to ensure that the driver weighs safety and timeliness and chooses a reasonable lane-changing behavior to reach the target point as soon as possible. The profit function shown in formula (12) is obtained:

Images