Automatic driving vehicle lane changing game method based on hypergraph information interoperation

Abstract

The invention provides an automatic driving vehicle lane changing game method based on hypergraph information interoperation. The method comprises the steps that an automatic driving vehicle selects available gaps for vehicle lane changing, and the gap priority sequence is determined through gap benefit evaluation; according to the associated vehicle driving state, the lane changing vehicle driving intention and the priority of available gaps, establishing a lane changing vehicle driving intention safety boundary, and setting interoperation safety boundary constraints of vehicle lane changing; constructing an automatic driving vehicle driving interoperation hyperedge and an associated vehicle derivative hyperedge according to interoperation safety boundary constraints of vehicle lane changing; according to the interoperation hyperedge and the associated vehicle derivative hyperedge, the autonomous vehicle games with all other vehicles, and a lane changing scheme which can be executed by the autonomous vehicle is obtained. According to the automatic driving vehicle lane changing game method based on hypergraph information interoperation, the interaction relation of the automatic driving vehicle in the lane changing process can be represented, and therefore a safe, efficient and accurate lane changing decision can be achieved.

Description

Technical Field

[0001] The present invention relates to the technical fields of autonomous driving interoperability, graph theory, and game theory, and in particular to a lane changing game method for an autonomous driving vehicle based on hypergraph information interoperability. Background Art

[0002] Autonomous driving is considered an effective solution to alleviate traffic congestion and improve traffic safety. Autonomous driving will become the main mode of travel in the future, and there are significant differences in driving behavior between autonomous driving and manually driven vehicles. Traditional lane-changing models are designed based on driver characteristics, while autonomous vehicles with V2V (Virtual to Virtual) communication functions have interoperable behaviors and produce more complex interactive effects on surrounding vehicles. Therefore, studying the lane-changing model of autonomous driving based on human-like behavior is one of the key technologies that need to be solved for its large-scale application, and has become a current research hotspot at home and abroad.

[0003] In previous studies, traditional lane-changing models only considered the one-way impact of surrounding vehicles on lane-changing vehicles, such as rule-based lane-changing models and lane-changing models based on utility theory. These models take lane-changing vehicles as the main body and mainly consider the impact of the traffic environment on the vehicle's lane-changing decision. Connected self-driving cars are equipped with advanced on-board sensors, controllers, actuators and other equipment that can perceive environmental information and make intelligent driving decisions. In a connected driving environment, not only can the dynamic state of the vehicle be exchanged and shared, but also the vehicle's driving intentions, behaviors and surrounding environment information. This makes multi-vehicle interoperability in complex environments possible.

[0004] Therefore, in recent years, with the development of computing resources and communication capabilities, the mainstream method has become distributed decision-making and centralized decision-making and other interoperable methods. Among them, distributed decision-making, such as the game lane-changing method, treats each vehicle as an independent decision-making unit, has relatively small requirements for computing resources, can adapt to dynamically changing traffic environments, and has shown certain potential.

[0005] At present, the disadvantages of the existing game lane changing methods include: it is difficult to express the complex interactive relationships in actual autonomous driving vehicles, for example, when a vehicle completes an action, it may conflict with multiple vehicles. The existing game lane changing methods cannot accurately and comprehensively model the complex relationships in the vehicle lane changing process, which greatly limits the vehicle's lane changing efficiency and makes it difficult to well control the cascading effect caused by lane changing. Summary of the invention

[0006] The present invention provides an autonomous driving vehicle lane changing game method based on hypergraph information interoperation to achieve effective vehicle lane changing decision-making.

[0007] In order to achieve the above object, the present invention adopts the following technical scheme.

[0008] A lane-changing game method for an autonomous driving vehicle based on hypergraph information interoperation, comprising:

[0009] The autonomous driving vehicle selects available gaps for vehicles to change lanes and determines the gap priority through gap benefit evaluation;

[0010] According to the driving status of the associated vehicles, the driving intention of the lane-changing vehicle and the priority of the available gap, the driving intention safety boundary of the lane-changing vehicle is established, and the interoperability safety boundary constraints of the vehicle lane-changing are set;

[0011] Constructing the autonomous driving vehicle driving interoperability hyperedge and the associated vehicle derived hyperedge according to the interoperability safety boundary constraint of the vehicle lane change;

[0012] According to the interoperable hyperedge and the associated vehicle derived hyperedge, the autonomous driving vehicle respectively negotiates with all other vehicles to obtain a lane changing plan that can be executed by the autonomous driving vehicle.

[0013] Preferably, the autonomous driving vehicle selects available gaps for lane change and determines the gap priority order through gap benefit evaluation, including:

[0014] Select g gaps closest to the lane-changing vehicle in the target lane as the target gaps for lane changing, where the gap is the distance between two adjacent vehicles traveling in the same lane.

[0015] The calculation formulas for the distance between the lane-changing vehicle and the gap are shown in equations (1)-(3).

[0016] D i =min(|x j -x m |,|(x j+1 -l)-x m |) (1)

[0017]

[0018] Among them, j+1 and j are the preceding and following vehicles of vehicle i, D i Indicates lane-changing vehicle j 0 The distance between the jth gap and the lane-changing vehicle is the minimum distance between the gap and the center of the lane-changing vehicle. m Indicates lane-changing vehicle j 0 The center position of the lane-changing vehicle, l represents the length of the lane-changing vehicle, and x j+1 represents the front position of the vehicle in the jth gap, x j represents the front position of the vehicle behind the jth gap, Indicates lane-changing vehicle j 0The front position of the vehicle;

[0019] The evaluation functions for sorting gaps are shown in equations (4)-(6).

[0020]

[0021] Among them, E i represents the benefit of the i-th gap. The greater the benefit, the more inclined the lane-changing vehicle is to change lanes to the gap. i represents the length of the i-th gap, EV i It represents the distance between the gap and the lane-changing vehicle. The larger the gap, the farther the gap is from the vehicle and the lower the benefit of the gap. The smaller the gap, the closer the gap is to the lane-changing vehicle and the greater the benefit of the gap. 0 Indicates the obstacle location, ED i It indicates the urgency of lane changing, which is the inverse of the product of the distance between the rear vehicle in the gap and the obstacle and the distance between the lane changing vehicle and the obstacle. α and β are parameters.

[0022] Preferably, the above-mentioned establishing the lane-changing vehicle driving intention safety boundary and setting the interoperability safety boundary constraint of the vehicle lane-changing according to the driving state of the associated vehicles, the lane-changing vehicle driving intention and the priority of the available gap includes: assuming that the speed and acceleration of the leading vehicle and the following vehicle are both v j and a j , the vehicle redundant reaction time is τ, as shown in formula (7).

[0023] S j =x j+1 -x j =v j ×τ+d j (7)

[0024] Among them, x j+1 and x j They represent the positions of the front and rear vehicles at a certain moment, respectively, j represents the speed of the two vehicles, τ represents the vehicle redundant reaction time d j Indicates the minimum headway distance of the vehicle;

[0025] When the vehicle is following the vehicle, the linear expression of the headway and speed time is S n =τ×V j +d, where d represents the minimum headway between the two vehicles after emergency braking, which is a constant value, and the slope τ is the redundant reaction time of the variable autonomous driving vehicle, v j Indicates the speed of the following vehicle;

[0026] With the minimum headway d as the height and the redundant reaction time τ as the base, a rectangle is constructed and slid over the trajectory of the rear vehicle to obtain the minimum critical trajectory of the front vehicle to the rear vehicle. This trajectory represents the critical trajectory of the front vehicle closest to the rear vehicle when the space-time trajectory of the rear vehicle is fixed.

[0027] The longitudinal spatiotemporal safety boundary of the following model is extended to a two-dimensional plane. The calculation formulas of the lateral safety boundary SC and the longitudinal safety boundary SF at the time of vehicle lane change are shown in equations (8)-(9) respectively:

[0028] SC=vcτ+dc (8)

[0029] SF=vτ+d (9)

[0030] Among them, v is the longitudinal velocity, vc is the lateral velocity, τ is the reaction time, the minimum longitudinal head distance is d, and the minimum lateral vehicle spacing is dc;

[0031] The lane-changing trajectory of the vehicle is constructed, and the relationship between the lateral distance, longitudinal speed and position of the vehicle when changing lanes is derived, as shown in formula (10):

[0032]

[0033] Among them, yen d Indicates the lateral distance from the start to the end of the lane change. d Indicates lane width, x end Indicates the longitudinal distance from the start to the end of the lane change. Given x end and end The relationship between the vehicle's lateral position y and longitudinal position x can be obtained, and the relationship between the lateral velocity and time during lane change vc is shown in formula (11):

[0034]

[0035] Among them, v x The relationship between the longitudinal velocity of the vehicle and the displacement x is expressed as follows. The relationship between the lateral velocity of the vehicle and the longitudinal velocity when changing lanes is derived as shown in equation (12):

[0036]

[0037] The maximum lateral acceleration ac in the above process is shown in equation (13), where v end Indicates the longitudinal speed at the end of the lane change:

[0038]

[0039] Assume that the lane-changing vehicle prepares for lane-changing with uniformly variable speed before changing lanes. The lane-changing process is divided into two stages: preparation stage and lane-changing stage. The background vehicle does not have any lane-changing behavior, and its lateral velocity is always 0. Therefore, the lateral safety margin of the background vehicle is a constant value, that is, its shortest lateral spacing. The lateral and longitudinal safety margins of the background vehicle are shown in equations (26)-(27):

[0040] SC j，t =dc (26)

[0041] SF j，t =v j，t τ+d (27)

[0042] Among them, SC j，t represents the lateral safety margin of vehicle j at time t, SF j，t represents the longitudinal safety boundary of vehicle j at time t;

[0043] Based on the right front of the vehicle, and Respectively represent the longitudinal and lateral positions of the lane-changing vehicle, x j，t and j，t Respectively represent the longitudinal and lateral positions of the vehicle giving way. and They represent the lateral and longitudinal safety margins of the lane-changing vehicle respectively. When the relationship (28) exists, vehicle j 0 The safety boundary is not disturbed by vehicle j, otherwise, the safety boundary of the vehicle is disturbed and the two vehicles interact;

[0044]

[0045] When the longitudinal safety boundary and the lateral safety boundary of any one of the two vehicles are occupied by the other, there is interaction between the two vehicles, as shown in formula (29):

[0046]

[0047] in, and They represent the longitudinal and lateral positions of the lane-changing vehicle at time t, respectively, and x j，t and j，t They represent the longitudinal and lateral positions of the vehicle giving way at time t, and They represent the lateral and longitudinal safety margins of the lane-changing vehicle, lc and l represent the vehicle width and length, respectively;

[0048] The lateral and longitudinal safety boundaries of lane-changing vehicles are shown in equations (24)-(25):

[0049]

[0050] in, Represents vehicle j 0 The horizontal safety margin at time t, Represents vehicle j 0 The longitudinal safety boundary at time t, the lateral and longitudinal safety boundaries of the vehicle together constitute the safety boundary of the vehicle. Preferably, the said establishing the lane-changing vehicle driving intention safety boundary and setting the interoperability safety boundary constraint of the vehicle lane changing according to the priority of the associated vehicle driving state, the lane-changing vehicle driving intention and the available gap, further includes:

[0051] In the preparation stage of the vehicle changing lanes, let the longitudinal speed before changing lanes be v 0 , the speed in the second stage is v′, and the speed after the lane change is completed is v 2 , the vehicle longitudinal velocity is shown in formula (14), where a 1 、a 2 is the longitudinal acceleration in the first and third stages;

[0052]

[0053] t∈(0, t 1 ) when the vehicle moves from v 1 Perform uniformly accelerated motion to v′, and the speed reaches v′; t∈(t 1 , t 2 ) when the vehicle moves at a constant speed v′; t∈(t 2 , t 3 ) when the vehicle speed changes from v′ to v 2 , during the lane-changing phase, i.e., t∈(t 3 , t 4 ) when the vehicle maintains a uniform longitudinal speed and merges into the target gap;

[0054] When the vehicle is preparing to change lanes, the lateral velocity vc is 0. The lateral velocity in the lane changing stage is shown in formula (12). The lateral velocity of the vehicle is obtained by sorting it out as shown in formula (15):

[0055]

[0056] Distance travelled during the lane-changing vehicle preparation phase x str The constraints are shown in equations (16), (17) and (18):

[0057]

[0058] x str ≤(x 2 -x 0 )+v 2 ·t 3 +SF j0，t3 -lj+1 (18)

[0059] Among them, t 3 represents the time when the vehicle completes the lane change preparation phase, x str It indicates that the lane-changing vehicle is in the lane-changing preparation stage t∈(0, t 3 ) The longitudinal distance travelled, x 1 , x 2 , x 0 They represent the longitudinal positions of the vehicle behind the gap, the vehicle in front, and the lane-changing vehicle at time t=0, respectively. 1 and v 2 Respectively represent the longitudinal speed of the vehicle behind and in front of the gap, and SF j0，t3 They represent the gap behind the vehicle and the lane-changing vehicle at t 3 The vertical safety margin at the moment, l j0 and l j+1 They represent the lengths of the lane-changing vehicle and the vehicle in front of the gap respectively;

[0060] The lane-changing vehicle should complete the lane change before reaching the obstacle, satisfying constraints (19)-(20), where x 0 The distance between the vehicle and the obstacle before changing lanes;

[0061] x end =v 2 (t 4 -t 3 ) (19)

[0062] x str +x end <x 0 (20)

[0063] The maximum velocity and maximum acceleration constraints are shown in equations (21)-(22).

[0064]

[0065] v′∈(v min , v max ) (twenty two)

[0066] The maximum lateral acceleration is ac max , the lane-changing process must satisfy equation (23):

[0067] ac≤ac max (twenty three).

[0068] Preferably, the constructing of the autonomous driving vehicle driving interoperability hyperedge and the associated vehicle derived hyperedge according to the interoperability safety boundary constraint of the vehicle lane change further includes:

[0069] The hyperedge in hypergraph theory is used to represent the non-paired relationship when the autonomous driving vehicle changes lanes. The vehicles are taken as hypergraph nodes. The lane-changing vehicles and the vehicles whose lane-changing intentions conflict with their own safety boundaries constitute interoperable hyperedges. The vehicle interoperable hypergraph is constructed. Each vehicle has an interoperable hyperedge during driving. The interoperable hyperedges include lane-changing interoperable hyperedges, following interoperable hyperedges, and free-driving hyperedges.

[0070] When vehicles j and j+2 follow each other, the safety boundary and the front vehicle interact to form a convoy, then e 1 ={C j , C j+1}、e 2 ={C j+2 , C j+3} is the following interaction superedge, when vehicle j 0 The lane change intention safety boundary and vehicle j 0 When +1 and j+2 interact, a hyperedge e is constructed 1 is the lane-changing interoperability hyperedge, vehicle and vehicles For free driving, build a free driving interoperable hyperedge and when After giving way to the vehicle changing lanes, the safety margin will be The two vehicles have an interoperable relationship and build a following interoperable hyperedge This type of hyperedge is generated by the lane-changing intention and is a derived hyperedge of the lane-changing intention. Each interoperable hyperedge constitutes a game environment, and there is a sequence in the construction of hyperedges.

[0071] Preferably, the autonomous driving vehicle playing games with all other vehicles respectively according to the interoperable hyperedge and the associated vehicle derived hyperedge to obtain a lane change plan executable by the autonomous driving vehicle includes:

[0072] The multi-party game model is divided into multiple two-party games, namely, the two-party games between the target vehicle and each interacting vehicle. A multi-party game model is established for the vehicles in the interoperable hyperedge. The vehicle benefits are shown in formula (30), including individual benefits, group benefits and safety benefits, where λ E , G is the parameter;

[0073] I j =λ E IE j +λ G IG j +IS j (30)

[0074] In the formula, IE is individual benefit, as shown in formula (31).

[0075]

[0076] Where j = j 0 represents the lane-changing vehicle, whose individual benefits are divided into efficiency benefits and utility benefits. represents the average speed of the target lane, v′ represents the average speed of the lane where the lane-changing vehicle is located, and the speed gain is used. To express the efficiency benefit of vehicle lane change; x 0 Indicates the location of the obstacle. Indicates the longitudinal position of the lane-changing vehicle at the time of decision making, using the distance between the lane-changing vehicle and the obstacle is used to represent the utility benefit of a vehicle changing lanes, γ is a known parameter, j = j′ represents the vehicle that gives way, and its individual benefit is expressed as the speed difference before and after the lane change. represents the speed of the vehicle giving way at the end time, v j，0 Indicates the speed of the vehicle giving way at the start time;

[0077] IG is the group benefit. The greater the cascade effect caused by lane changing, the lower the group benefit of the subject vehicle, as shown in formula (32).

[0078]

[0079] Where j = j 0 represents a lane-changing vehicle, whose group benefit is the negative value of the number of derived hyperedges caused by the lane-changing -|e|. When the number of derived hyperedges generated by the lane-changing intention of the lane-changing vehicle is too large, it means that the cascade effect of the intention on the traffic flow is large, and the negative group benefit caused is high. j=j′ represents a yielding vehicle. When yielding is a passive decision, the group benefit is 0.

[0080] IS is the safety benefit, as shown in formula (33):

[0081]

[0082] in, represents the relationship between vehicle j′ and vehicle j under this strategy 0 A 0-1 variable indicating whether the safety margins of the two vehicles affect each other. When the two vehicles affect each other, Otherwise, it is 0. If the safety boundaries of the two vehicles affect each other, there will be a safety hazard. Let M be a very large number, so that the safety benefits of both parties are extremely low.

[0083] The strategy set of lane-changing vehicles is S 1 ={p 1 , p 2}, where p 1 represents the lane changing strategy, p 2represents the no lane change strategy, and the strategy set of other vehicles is S 2 = {q 1 ,q 2}, where q 1 represents the yield strategy, q 2 represents the no-yield strategy, and the benefits obtained by the two parties in the game when adopting different strategies are I, Indicates that the lane-changing vehicle adopts strategy p 1 The yielding vehicle adopts strategy q 1 The benefit obtained by the lane-changing vehicle when ;

[0084] The payoff matrix of the lane-changing game is shown in Table 2:

[0085] Table 2 Profit Matrix

[0086]

[0087] If there is a strategy (p n ,q n ) makes equation (34) true, then the strategy (p n ,q n ) is the Nash equilibrium point of the game;

[0088]

[0089] The point with the highest total benefit of both parties is selected as the final equilibrium solution. 1 ,q 1 ) is the final equilibrium solution, the lane-changing vehicle game is successful, and the lane-changing vehicle will play games with all other vehicles respectively, and all the game results (p 1 ,q 1 ) are all final equilibrium solutions. The vehicle is able to change lanes and output a lane-changing plan that can be executed by the autonomous driving vehicle. The lane-changing plan includes the decision of whether to change lanes and the trajectories of the lane-changing vehicle and surrounding vehicles.

[0090] It can be seen from the technical solutions provided by the above-mentioned embodiments of the present invention that the lane changing game method of autonomous driving vehicles based on hypergraph information interoperation of the present invention can characterize the interactive relationship of autonomous driving vehicles during the lane changing process and reduce the cascading effect caused by vehicle lane changing, thereby achieving safe, efficient and accurate lane changing decisions.

[0091] Additional aspects and advantages of the present invention will be given in part in the following description, which will become obvious from the following description, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0093] Figure 1 A schematic diagram of an implementation principle of a lane-changing game method for an autonomous driving vehicle based on hypergraph information interoperation provided in an embodiment of the present invention;

[0094] Figure 2 A schematic diagram of the relationship between a gap and a lane-changing vehicle provided in an embodiment of the present invention;

[0095] Figure 3 A three-dimensional schematic diagram of emergency braking provided by an embodiment of the present invention;

[0096] Figure 4 A schematic diagram of a safety boundary provided by an embodiment of the present invention;

[0097] Figure 5 A schematic diagram of a space-time trajectory provided by an embodiment of the present invention;

[0098] Figure 6 A schematic diagram of a lane changing trajectory provided by an embodiment of the present invention

[0099] Figure 7 Schematic diagram of four speed changes provided by an embodiment of the present invention

[0100] Figure 8 A schematic diagram of a vehicle lane-changing relationship provided in an embodiment of the present invention

[0101] Fig. 9 A schematic diagram of the relationship between a horizontal safety boundary and a vertical safety boundary provided in an embodiment of the present invention

[0102] Fig.10 Schematic diagram of four situations of lane-changing vehicles and vehicles giving way provided by an embodiment of the present invention

[0103] Fig.11 A schematic diagram of different interoperable hyperedges of a vehicle of the present invention provided by an embodiment of the present invention;

[0104] Fig.12 In an embodiment of the present invention, when a vehicle changes lanes, the vehicles that are directly affected by the yielding strategy are in four directions that do not interfere with each other. DETAILED DESCRIPTION

[0105] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0106] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or coupling. The term "and / or" used herein includes any unit and all combinations of one or more associated listed items.

[0107] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0108] To facilitate understanding of the embodiments of the present invention, several specific embodiments will be further explained below with reference to the accompanying drawings, and each embodiment does not constitute a limitation on the embodiments of the present invention.

[0109] Using interoperable hyperedges to model and characterize complex interactions between vehicles has obvious advantages and is in line with future development trends. According to the available gaps in the target lane, a benefit evaluation function of the available gaps in the target lane is designed to implement the lane change priority sorting of the target lane gaps. Next, a safety boundary is established based on the vehicle intentions. The intentional interacting vehicles construct interoperable hyperedges, and finally a game is played. The implementation principle of the lane change game method for autonomous driving vehicles based on hypergraph information interoperability provided in an embodiment of the present invention is as follows: Figure 1 As shown, the processing steps include the following:

[0110] Step S1: Select an available gap for the vehicle to change lanes.

[0111] The data used in the examples of the present invention are described in Table 1.

[0112] Table 1 Data description

[0113]

[0114] Step 1.1: Select available gap;

[0115] The g gaps closest to the lane-changing vehicle in the target lane are selected as the target gaps for lane changing. The above gaps are the distances between two adjacent vehicles traveling in the same lane.

[0116] Among them, the distance relationship between the lane-changing vehicle and the gap is as follows: Figure 2 As shown, the calculation formulas are shown in formulas (1)-(3).

[0117] D i =min(|x j -x m |,|(x j+1 -l)-x m |) (1)

[0118]

[0119] j+1j is the front and rear vehicle of vehicle i, l is the length of the lane-changing vehicle, D i Indicates lane-changing vehicle j 0 The distance between the jth gap is the minimum distance between the center of the lane-changing vehicle before and after the gap. m Indicates lane-changing vehicle j 0 The central location Indicates lane-changing vehicle j 0 The length of the vehicle, x j+1 represents the front position of the vehicle in the jth gap, x j represents the front position of the vehicle behind the jth gap, Indicates lane-changing vehicle j 0 The front position of the vehicle.

[0120] Step 1.2: Sort by available gaps;

[0121] By considering factors such as the gap size, the longitudinal distance between the gap and the lane-changing vehicle, and the distance between the gap and the obstacle in front, the evaluation function of the lane-changing gap sorting model is designed as shown in Equations (4)-(6).

[0122]

[0123] Among them, E i represents the benefit of the i-th gap. The greater the benefit, the more inclined the lane-changing vehicle is to change lanes to the gap. i represents the length of the i-th gap, Figure 2A schematic diagram of the relationship between a gap and the position of a lane-changing vehicle provided in an embodiment of the present invention is shown in FIG. Figure 2 As shown in the figure, the greater the gap length, the greater the benefit, and the shorter the gap length, the smaller the benefit. i Indicates the distance between the gap and the lane-changing vehicle. The larger the gap, the farther the gap is from the vehicle and the lower the benefit of the gap. The smaller the gap, the closer the gap is to the lane-changing vehicle and the greater the benefit of the gap. 0 Indicates the location of obstacles, such as the no lane change zone before the intersection, ED i Indicates the urgency of lane change, which is the inverse of the product of the distance between the rear vehicle and the obstacle and the distance between the lane change vehicle and the obstacle. The closer the distance, the greater the inverse, the more difficult the gap is to use, and the lower the benefit. α and β are parameters.

[0124] Step S2: Delineate the interoperability safety boundary for vehicle lane changing.

[0125] Step 2.1: Principles of safety boundary demarcation

[0126] Taking the following car as an example, the first-order expression of speed and reaction time is derived to represent the longitudinal safety boundary of the CAV (Connected autonomous vehicle) convoy. Assuming that the leading car and the following car are following each other, since the following cars are close together, the speed and acceleration of the two cars are the same, both v n and a n , when the front vehicle brakes urgently due to the situation, the rear vehicle follows the emergency braking. The present invention considers that the reaction time of the vehicle due to technical limitations is almost 0. However, in order to reduce the potential safety hazard caused by communication delay, the present invention adds a redundant reaction time τ, as shown in formula (7).

[0127] S j =x j+1 -x j = x′ j+1 -x′ j +v j ×τ=V j ×τ+d j (7)

[0128] Therefore, we can get the linear expression S of the headway and speed time when the vehicle is following the vehicle: n =τ×V j +d, where d represents the minimum headway between the two vehicles after emergency braking, which is a constant value, and the slope τ is the redundant reaction time of the variable autonomous driving vehicle, v j Indicates the speed of the following vehicle. Figure 3 A three-dimensional schematic diagram of emergency braking provided by an embodiment of the present invention, such as Figure 3As shown, by approximating Line2 and Line4 to Line1, Line3, the linear relationship of the safety margin can be obtained.

[0129] With the minimum headway d as the height and the redundant reaction time τ as the base, a rectangle is constructed and slid over the trajectory of the rear vehicle to obtain the minimum critical trajectory of the front vehicle to the rear vehicle. This trajectory represents the critical trajectory of the front vehicle closest to the rear vehicle when the space-time trajectory of the rear vehicle is fixed. Figure 4 A schematic diagram of a security boundary provided by an embodiment of the present invention, such as Figure 4 As shown in the figure, since the vehicle can transmit the space-time trajectory, it is different from human drivers. Instead of only judging the current safety situation, it can judge whether an interaction occurs based on the space-time trajectory diagram, thereby identifying whether the vehicle strategy is safe within the entire time window. Figure 5 A schematic diagram of a space-time trajectory provided by an embodiment of the present invention, Figure 5 In, C j represents the intermediate vehicle trajectory, C j+1 and C j-1 Denote the trajectories of the front and rear vehicles respectively. According to the τ×d block, the safety boundary SF of the middle vehicle and the rear vehicle is obtained. j and SF j-1 Two trajectories. Figure 5 The safety margins of the three vehicles in (a) are not affected, and Figure 5 (b) Car C in the middle and rear j-1 The safety margin is C j The vehicle intrudes, so there is a safety hazard, which is regarded as an interactive behavior between the two vehicles in the longitudinal dimension.

[0130] Based on this, the present invention extends the longitudinal space-time safety boundary of the vehicle following model to a two-dimensional plane. The lateral and longitudinal safety boundaries (SC, SF) of the vehicle at the time of lane change are shown in equations (8)-(9) respectively.

[0131] SC=vcτ+dc (8)

[0132] SF=vτ+d (9)

[0133] Among them, v is the longitudinal velocity, vc is the lateral velocity, τ is the reaction time, the minimum longitudinal head distance is d, and the minimum lateral vehicle spacing is dc.

[0134] Step 2.2: Construct the lane-changing trajectory of the vehicle.

[0135] According to the trajectory of the lane-changing process, the relationship between the lateral distance, longitudinal speed and position of the vehicle when changing lanes is derived.

[0136] The lane changing trajectory is shown in formula (10).

[0137]

[0138] Among them, yen d Indicates the lateral distance from the start to the end of the lane change. d Indicates lane width, x end Indicates the longitudinal distance from the start to the end of the lane change. Given x end and end The relationship between the vehicle's lateral position y and longitudinal position x can be obtained, and the relationship between the lateral velocity and time during lane change vc can be obtained, as shown in formula (11).

[0139]

[0140] Among them, v x It represents the relationship between the longitudinal velocity of the vehicle and the displacement x. Therefore, the relationship between the lateral velocity and the longitudinal velocity of the vehicle when changing lanes is derived as shown in equation (12).

[0141]

[0142] The maximum lateral acceleration ac in the above process is shown in equation (13), where v end Indicates the longitudinal speed at the end of the lane change.

[0143]

[0144] The present invention assumes that the lane-changing vehicle prepares for the lane-changing by uniformly changing speed before the lane-changing. Figure 6 A lane changing trajectory schematic diagram provided by an embodiment of the present invention, such as Figure 6 The lane changing process is divided into two stages: preparation stage and lane changing stage.

[0145] Figure 7 Schematic diagram of four speed changes provided by an embodiment of the present invention. The preparation stage consists of three parts, using a "trapezoidal" speed curve as shown in FIG. Figure 7 As shown, there are four types of speed changes. Let the longitudinal speed before lane change be v 0 , the speed in the second stage is v′, and the speed after the lane change is completed is v 2 The vehicle longitudinal velocity is shown in equation (14), where a 1 、a 2 is the longitudinal acceleration of the first and third stages.

[0146]

[0147] t∈(0, t 1 ) when the vehicle moves from v 1 Perform uniformly accelerated motion to v′, and the speed reaches v′; t∈(t 1 , t 2) when the vehicle moves at a constant speed v′; t∈(t 2 , t 3 ) when the vehicle speed changes from v′ to v 2 . During the lane-changing phase, t∈(t 3 , t 4 ) when the vehicle maintains a uniform longitudinal speed and merges into the target gap.

[0148] When the vehicle is preparing to change lanes, the lateral velocity vc is 0. The lateral velocity during the lane change phase is shown in equation (12). Therefore, the lateral velocity of the vehicle is shown in equation (15).

[0149]

[0150] Distance travelled by the vehicle in the lane-changing preparation phase x str The constraints are shown in formula (16).

[0151]

[0152] To ensure that the longitudinal position of the lane-changing vehicle is between the target gaps after the preparation phase is completed (at time t3), and the safety margin is not affected, Figure 8 A schematic diagram of a vehicle lane-changing relationship provided by an embodiment of the present invention, such as Figure 8 As shown, therefore x str The constraints (17)-(18) should be satisfied.

[0153]

[0154] x str ≤(x 2 -x 0 )+v 2 ·t 3 +SF j0，t3 -l j+1 (18)

[0155] Among them, t 3 represents the time when the vehicle completes the lane change preparation phase, x str It indicates that the lane-changing vehicle is in the lane-changing preparation stage t∈(0, t 3 ) The longitudinal distance travelled, x 1 , x 2 , x 0 They represent the longitudinal positions of the vehicle behind the gap, the vehicle in front, and the lane-changing vehicle at time t=0, respectively. 1 and v 2 Respectively represent the longitudinal speed of the vehicle behind and in front of the gap, and SF j0，t3 They represent the gap behind the vehicle and the lane-changing vehicle at t 3 The vertical safety margin at the moment, lj0 and l j+1 They respectively represent the length of the lane-changing vehicle and the vehicle in front of the gap.

[0156] The lane-changing vehicle should complete the lane change before reaching the obstacle, so the constraints (19)-(20) are satisfied, where x 0 It is the distance between the vehicle and the obstacle before changing lanes.

[0157] x end =v 2 (t 4 -t 3 ) (19)

[0158] x str +x end <x 0 (20)

[0159] The maximum velocity and maximum acceleration constraints are shown in equations (21)-(22).

[0160]

[0161] v′∈(v min , v max ) (twenty two)

[0162] To ensure driving safety and passenger comfort, the maximum lateral acceleration is ac max , so the lane-changing process needs to satisfy equation (23):

[0163] ac≤ac max (twenty three)

[0164] Therefore, the lateral and longitudinal safety boundaries of lane-changing vehicles are shown in equations (24)-(25):

[0165]

[0166] in, Represents vehicle j 0 The horizontal safety margin at time t, Represents vehicle j 0 The longitudinal safety margin at time t.

[0167] The background vehicle has no lane-changing behavior, and its lateral speed is always 0. Therefore, the lateral safety margin of the background vehicle is a constant, that is, its shortest lateral distance. The lateral and longitudinal safety margins of the background vehicle are shown in equations (26)-(27):

[0168] SC j，t =dc (26)

[0169] SF j，t =v j,tτ+d (27)

[0170] Among them, SC j，t represents the lateral safety margin of vehicle j at time t, SF j，t represents the longitudinal safety boundary of vehicle j at time t, Fig. 9 A schematic diagram of the relationship between a horizontal security boundary and a vertical security boundary provided by an embodiment of the present invention is shown in FIG. Fig. 9 As shown, based on the right front of the vehicle, and Respectively represent the longitudinal and lateral positions of the lane-changing vehicle, x j，t and j，t Respectively represent the longitudinal and lateral positions of the vehicle giving way. and They represent the lateral and longitudinal safety margins of the lane-changing vehicle respectively. When the relationship in equation (28) exists, Fig. 9 The vehicle shown in 0 The safety boundary is not disturbed by vehicle j, otherwise, the safety boundary of vehicle j is disturbed and the two vehicles interact.

[0171]

[0172] There are four situations in which two vehicles need to change lanes or give way. Fig.10 A schematic diagram of four situations of lane-changing vehicles and vehicles giving way provided by an embodiment of the present invention. Fig.10 In the four cases, the safety boundaries interact with each other. In Figures (a) and (b), the horizontal and vertical safety boundaries of the red car are occupied by the white car. In Figures (c) and (d), the horizontal boundary of the red car and the vertical boundary of the white car are occupied by each other. In the above four cases, there is an interaction between the two cars, and it is necessary to build an interoperable hyperedge to play the game. Therefore, the interaction relationship between the two cars is summarized. When the longitudinal safety boundary of either car and the horizontal safety boundary of either car are occupied by the other car, there is an interaction between the two cars, as shown in formula (29):

[0173]

[0174] in, and They represent the longitudinal and lateral positions of the lane-changing vehicle at time t, respectively, and x j，t and j，t They represent the longitudinal and lateral positions of the vehicle giving way at time t, and They represent the lateral and longitudinal safety boundaries of the lane-changing vehicle respectively, lc and l represent the vehicle width and length respectively.

[0175] Step S3: Construct the interoperability hyperedge of vehicle lane change.

[0176] During the lane-changing process, there is a possibility that the vehicle's own safety boundary is invaded or invades the safety boundary of other vehicles, thus affecting the safety of the vehicle. To ensure driving safety, one of the vehicles that affect each other must compromise and give way to the other. Similarly, when the vehicle's lane-changing intention conflicts with the safety boundaries of multiple surrounding vehicles, the lane-changing vehicle needs to negotiate with multiple vehicles. Only when all the negotiation results of the lane-changing vehicle are feasible can the lane-changing intention be realized. Otherwise, the vehicle will give up the intention and look for a gap in the lane-changing interval to negotiate again, or give up the lane-changing.

[0177] Since it is difficult to represent non-paired relationships in ordinary graphs, and the lane-changing process often involves multiple vehicles, the present invention uses hyperedges in hypergraph theory to represent non-paired relationships when autonomous driving vehicles change lanes. Taking vehicles as hypergraph nodes, vehicles that change lanes and vehicles that conflict with their own lane-changing intention safety boundaries constitute interoperable hyperedges to construct a vehicle interoperable hypergraph. Each vehicle has an interoperable hyperedge during driving, including lane-changing interoperable hyperedges, following interoperable hyperedges (convoys), free driving hyperedges, etc. Fig.11 A schematic diagram of different interoperable hyperedges of a vehicle of the present invention provided by an embodiment of the present invention is as follows: Fig.11 When vehicles j and j+2 follow each other, the safety boundary and the front vehicle interact to form a convoy, then e 1 ={C j , C j+1}、e 2 ={C j+2 , C j+3} is the following interaction superedge, when vehicle j 0 The lane change intention safety boundary and vehicle j 0 When +1 and j+2 interact, a hyperedge e is constructed 1 is the lane-changing interoperability hyperedge, vehicle and vehicles For free travel, we build a free travel interoperability hyperedge and when After giving way to the vehicle changing lanes, the safety margin will be Therefore, the two vehicles have an interoperable relationship and build a following interoperable hyperedge This type of hyperedge is generated by the lane-changing intention and is a derived hyperedge of the lane-changing intention. Each interoperable hyperedge constitutes a game environment, and the construction of hyperedges has a sequence. The game strategy of vehicles in the interoperable hyperedge may lead to the generation of new hyperedges.

[0178] Step S4: Interoperable hyperedge agent game.

[0179] Step 4.1: Calculate the game payoff matrix;

[0180] The driving intentions of vehicles within the interoperable hyperedge conflict and the safety of vehicles is threatened, so one party must give in. Based on this, this section establishes a multi-party game model for vehicles within the interoperable hyperedge.

[0181] Since the decision-making process of vehicles giving way does not interfere with each other, Fig.12 In an embodiment of the present invention, when a vehicle changes lanes, the vehicle yielding strategy directly affected is four directions that do not interfere with each other, such as Fig.12 As shown, this method divides the multi-party game model into multiple two-party games, namely, two-party games between the target vehicle and each interacting vehicle.

[0182] The benefits of most lane-changing vehicles are generally divided into individual efficiency benefits and safety benefits. The hyperedge model described in the present invention enables vehicles to simultaneously take into account the impact of lane-changing behavior on the regional traffic flow system environment. Therefore, the present invention adds group benefits to represent the impact of lane-changing behavior on the traffic flow state. The vehicle benefits are shown in formula (30), including individual benefits, group benefits, and safety benefits, where λ E , G as a parameter.

[0183] I j =λ E IE j +λ G IG j +IS j (30)

[0184] In the formula, IE is individual benefit, as shown in formula (31).

[0185]

[0186] Where j = j 0 Represents the lane-changing vehicle, whose individual benefits are divided into efficiency benefits and utility benefits. represents the average speed of the target lane, v′ represents the average speed of the lane where the lane-changing vehicle is located, and the present invention adopts the speed gain To express the efficiency benefit of vehicle lane change; x 0 Indicates the location of the obstacle. Indicates the longitudinal position of the lane-changing vehicle at the decision moment. The present invention uses the distance between the lane-changing vehicle and the obstacle is used to represent the utility benefit of a vehicle changing lanes, and γ is a known parameter. j=j′ represents the vehicle that gives way, and its individual benefit is expressed as the speed difference before and after the lane change. represents the speed of the vehicle giving way at the end time, v j，0 Indicates the speed of the vehicle giving way at the start time.

[0187] IG is the group benefit. The greater the cascade effect caused by lane changing, the lower the group benefit of the subject vehicle, as shown in formula (32).

[0188]

[0189] Where j = j 0 represents a lane-changing vehicle, whose group benefit is the negative value of the number of derived hyperedges caused by the lane change -|e|. When the yielding vehicle makes a decision, the interoperable safety boundary generated by the decision intention may affect other vehicles, thereby deriving new interoperable hyperedges and interoperating vehicle intentions. Therefore, when the number of derived hyperedges generated by the lane-changing intention of the lane-changing vehicle is too large, it means that the intention has a large cascading effect on the traffic flow, resulting in a high negative group benefit. j=j′ represents the yielding vehicle. Since it is a passive decision when yielding, the group benefit is 0.

[0190] IS is the safety benefit, as shown in formula (33).

[0191]

[0192] in, represents the relationship between vehicle j′ and vehicle j under this strategy 0 A 0-1 variable indicating whether the safety margins of the two vehicles affect each other. When the two vehicles affect each other, Otherwise, it is 0. If the safety margins of the two vehicles affect each other, there will be potential safety hazards. To prevent this dangerous phenomenon, let M be a very large number so that the safety benefits of both parties are extremely low.

[0193] In the lane-changing game, the lane-changing vehicle has two basic driving strategies: {do not change lanes, change lanes}, and other vehicles have two basic driving strategies: {give way, do not give way}. Therefore, the strategy set of the lane-changing vehicle is S 1 ={p 1 , p 2}, where p 1 represents the lane changing strategy, p 2 represents the no lane change strategy. The strategy set of other vehicles is S 2 = {q 1 ,q 2}, where q 1 represents the yield strategy, q 2 represents the no-yield strategy, and the benefits obtained by the two parties in the game when adopting different strategies are I, for example Indicates that the lane-changing vehicle adopts strategy p 1 The yielding vehicle adopts strategy q 1 The benefit obtained by the lane-changing vehicle when .

[0194] The payoff matrix of the lane-changing game is shown in Table 2.

[0195] Table 2 Profit Matrix

[0196]

[0197] Step 4.2: Solve the Nash equilibrium strategy;

[0198] Nash equilibrium is the main solution strategy in the game. In this strategy, both parties cannot unilaterally change their strategies to increase their own benefits. The strategy of each player is based on their common strategic judgment to achieve the optimal solution of the game. The result is to maximize the interests of both parties in the game.

[0199] If there is a strategy (p n ,q n ) makes equation (34) true, then the strategy (p n ,q n ) is the Nash equilibrium point of the game.

[0200]

[0201] The main difference between self-driving vehicles and manually driven vehicles is that there is no randomness in the decision-making of self-driving vehicles, that is, there is no situation where the choice violates the optimal strategy. Therefore, this method selects the point with the highest total benefit of both parties as the final equilibrium solution. Therefore, when (p 1 ,q 1 ) is the final equilibrium solution, the lane-changing vehicle game is successful.

[0202] For the lane-changing vehicle, since there may be multiple (>2) superpoints in the lane-changing interoperability hyperedge, the lane-changing vehicle will play games with all other vehicles separately, and all the game results (p 1 ,q 1 ) are all final equilibrium solutions, and the car can change lanes. Finally, a lane-changing plan that can be executed by the autonomous driving vehicle is output, which includes the decision of whether to change lanes and the trajectories of the lane-changing vehicle and surrounding vehicles.

[0203] In summary, the embodiments of the present invention can effectively characterize the interactive relationships generated during the lane changing process of an autonomous driving vehicle. Due to the defects of the traditional game lane changing model itself, it is difficult to characterize the complex interactions of autonomous driving vehicles. The current lane changing decision method for autonomous driving vehicles based on traditional game theory is greatly constrained. Compared with the traditional game theory method, the interoperable hyperedge game method used in the present invention can not only characterize the direct impact between the lane changing vehicle and the directly related vehicle, but also characterize the cascading impact of the lane changing process on the adjacent vehicles of the directly related vehicle, ultimately achieving a comprehensive improvement in the lane changing considerations.

[0204] The present invention can effectively overcome the defects of traditional lane-changing methods, significantly improve the interactive safety of autonomous vehicles, and effectively reduce the cascading effect in complex lane-changing processes. The method can be applied to lane-changing decisions of autonomous vehicles in real road networks, and will obtain a safe, accurate, and efficient lane-changing decision solution.

[0205] The present invention can effectively reduce the cascading effect produced during the lane changing process of an autonomous driving vehicle. The traditional game lane changing method does not take into account the shock effect of the lane changing process on the traffic flow, and cannot characterize the speed volatility of the traffic flow after the lane changing. Based on the game theory, the present invention adopts the hypergraph theory, the safety boundary model, the interoperable hyperedge method, and the derived hyperedge method to achieve the characterization and characterization of the cascading effect during the lane changing process of the autonomous driving vehicle, and finally safely and effectively decides the position and target gap of the autonomous driving vehicle for lane changing, reducing the cascading effect produced by the lane changing process on the traffic flow.

[0206] Those skilled in the art can understand that the accompanying drawings are only schematic diagrams of an embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0207] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention or certain parts of the embodiments.

[0208] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0209] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A lane-changing game method for autonomous driving vehicles based on hypergraph information interoperation, characterized in that: include: The autonomous driving vehicle selects available gaps for vehicles to change lanes and determines the gap priority through gap benefit evaluation; According to the driving status of the associated vehicles, the driving intention of the lane-changing vehicle and the priority of the available gap, the driving intention safety boundary of the lane-changing vehicle is established, and the interoperability safety boundary constraints of the vehicle lane-changing are set; Constructing the autonomous driving vehicle driving interoperability hyperedge and the associated vehicle derived hyperedge according to the interoperability safety boundary constraint of the vehicle lane change; According to the interoperable hyperedge and the associated vehicle derived hyperedge, the autonomous driving vehicle respectively plays games with all other vehicles to obtain a lane changing plan that can be executed by the autonomous driving vehicle.

2. The method according to claim 1, characterized in that The autonomous driving vehicle selects available gaps for vehicle lane change and determines the gap priority order through gap benefit evaluation, including: Select g gaps closest to the lane-changing vehicle in the target lane as the target gaps for lane changing, where the gap is the distance between two adjacent vehicles traveling in the same lane. The calculation formulas for the distance between the lane-changing vehicle and the gap are shown in equations (1)-(3). D i =min(|x j -x m |,|(x j+1 -l)-x m |) (1) Among them, j+1 and j are the preceding and following vehicles of vehicle i, D i represents the distance between the lane-changing vehicle j0 and the jth gap, which is the minimum distance between the center of the lane-changing vehicle before and after the gap, x m represents the center position of the lane-changing vehicle j0, l represents the length of the lane-changing vehicle, x j+1 represents the front position of the vehicle in the jth gap, x j represents the front position of the vehicle behind the jth gap, represents the front position of the lane-changing vehicle j0; The evaluation functions for sorting gaps are shown in equations (4)-(6). Among them, E i represents the benefit of the i-th gap. The greater the benefit, the more inclined the lane-changing vehicle is to change lanes to the gap. i represents the length of the i-th gap, EV i Indicates the distance between the gap and the lane-changing vehicle. The larger the gap, the farther the gap is from the vehicle and the lower the benefit of the gap. The smaller the gap, the closer the gap is to the lane-changing vehicle and the greater the benefit of the gap. x0 indicates the obstacle position, and ED i It indicates the urgency of lane changing, which is the inverse of the product of the distance between the rear vehicle in the gap and the obstacle and the distance between the lane changing vehicle and the obstacle. α and β are parameters.

3. The method according to claim 2, characterized in that The above-mentioned establishment of a lane-changing vehicle driving intention safety boundary and setting of the interoperability safety boundary constraints for lane changing of vehicles based on the driving status of the associated vehicles, the driving intention of the lane-changing vehicle and the priority of the available gaps includes: assuming that the speed and acceleration of the leading vehicle and the following vehicle are both v j and a j , the vehicle redundant reaction time is τ, as shown in formula (7). S j =x j+1 -x j =v j ×τ+d j (7) Among them, x j+1 and x j They represent the positions of the front and rear vehicles at a certain moment, respectively, j represents the speed of the two vehicles, τ represents the vehicle redundant reaction time d j Indicates the minimum headway distance of the vehicle; When the vehicle is following the vehicle, the linear expression of the headway and speed time is S n =τ×v j +d, where d represents the minimum headway between the two vehicles after emergency braking, which is a constant value, and the slope τ is the redundant reaction time of the variable autonomous driving vehicle, v j Indicates the speed of the following vehicle; With the minimum headway d as the height and the redundant reaction time τ as the base, a rectangle is constructed and slid over the trajectory of the rear vehicle to obtain the minimum critical trajectory of the front vehicle to the rear vehicle. This trajectory represents the critical trajectory of the front vehicle closest to the rear vehicle when the space-time trajectory of the rear vehicle is fixed. The longitudinal spatiotemporal safety boundary of the following model is extended to a two-dimensional plane. The calculation formulas of the lateral safety boundary SC and the longitudinal safety boundary SF at the time of vehicle lane change are shown in equations (8)-(9) respectively: SC=vcτ+dc (8) SF=vτ+d (9) Among them, v is the longitudinal velocity, vc is the lateral velocity, τ is the reaction time, the minimum longitudinal head distance is d, and the minimum lateral vehicle spacing is dc; The lane-changing trajectory of the vehicle is constructed, and the relationship between the lateral distance, longitudinal speed and position of the vehicle when changing lanes is derived, as shown in formula (10): Among them, y end Indicates the lateral distance from the start to the end of the lane change. end Indicates lane width, x end Indicates the longitudinal distance from the start to the end of the lane change. Given x end and end The relationship between the vehicle's lateral position y and longitudinal position x can be obtained, and the relationship between the lateral velocity and time during lane change vc is shown in formula (11): Among them, v x The relationship between the longitudinal velocity of the vehicle and the displacement x is expressed as follows. The relationship between the lateral velocity of the vehicle and the longitudinal velocity when changing lanes is derived as shown in equation (12): The maximum lateral acceleration ac in the above process is shown in equation (13), where v end Indicates the longitudinal speed at the end of the lane change: Assume that the lane-changing vehicle prepares for lane-changing with uniformly variable speed before changing lanes. The lane-changing process is divided into two stages: preparation stage and lane-changing stage. The background vehicle does not have any lane-changing behavior, and its lateral velocity is always 0. Therefore, the lateral safety margin of the background vehicle is a constant value, that is, its shortest lateral spacing. The lateral and longitudinal safety margins of the background vehicle are shown in equations (26)-(27): SC j,t =dc (26) SF j,t =v j,t τ+d (27) Among them, SC j,t represents the lateral safety margin of vehicle j at time t, SF j,t represents the longitudinal safety boundary of vehicle j at time t; Based on the right front of the vehicle, and Respectively represent the longitudinal and lateral positions of the lane-changing vehicle, x j,t and j,t Respectively represent the longitudinal and lateral positions of the vehicle giving way. and They represent the lateral and longitudinal safety boundaries of the lane-changing vehicle, respectively. When the relationship in formula (28) exists, the safety boundary of vehicle j0 is not disturbed by vehicle j. Otherwise, the safety boundary of the vehicle is disturbed and the two vehicles interact. When the longitudinal safety boundary and the lateral safety boundary of any one of the two vehicles are occupied by the other, there is interaction between the two vehicles, as shown in formula (29): in, and They represent the longitudinal and lateral positions of the lane-changing vehicle at time t, respectively, and x j,t and j,t They represent the longitudinal and lateral positions of the vehicle giving way at time t, and They represent the lateral and longitudinal safety margins of the lane-changing vehicle, lc and l represent the vehicle width and length, respectively; The lateral and longitudinal safety boundaries of lane-changing vehicles are shown in equations (24)-(25): in, represents the lateral safety boundary of vehicle j0 at time t, It represents the longitudinal safety boundary of vehicle j0 at time t. The lateral and longitudinal safety boundaries of the vehicle together constitute the safety boundary of the vehicle.

4. The method according to claim 3, characterized in that The method of establishing a lane-changing vehicle driving intention safety boundary and setting an interoperability safety boundary constraint for vehicle lane changing according to the driving state of the associated vehicle, the lane-changing vehicle driving intention and the priority of the available gap, further includes: In the preparation stage of the vehicle lane change, let the longitudinal speed before lane change be v0, the speed in the second stage be v′, and the speed after lane change be v2. The longitudinal speed of the vehicle is shown in formula (14), where a1 and a2 are the longitudinal accelerations in the first and third stages; At t∈(0,t1), the vehicle moves from v1 to v′ with a uniform speed, and the speed reaches v′; at t∈(t1,t2), the vehicle moves at a uniform speed of v′; at t∈(t2,t3), the vehicle speed moves from v′ to v2 with a uniform speed. During the lane-changing phase, i.e., at t∈(t3,t4), the vehicle maintains a uniform longitudinal speed and merges into the target gap; When the vehicle is preparing to change lanes, the lateral velocity vc is 0. The lateral velocity in the lane changing stage is shown in formula (12). The lateral velocity of the vehicle is obtained by sorting it out as shown in formula (15): Distance travelled by the vehicle in the lane-changing preparation phase x str The constraints are shown in equations (16), (17) and (18): <h2 style=";text-align:left;direction:ltr">x<h2 style=";text-align:left;direction:ltr"> str <h2 style=";text-align:left;direction:ltr"> ≤(x2-x0)+v2·t3+SF<h2 style=";text-align:left;direction:ltr"> j0,t3 <h2 style=";text-align:left;direction:ltr"> -l<h2 style=";text-align:left;direction:ltr"> j+1 <h2 style=";text-align:left;direction:ltr"> (18) Among them, t3 represents the time when the vehicle completes the lane change preparation phase, x str represents the longitudinal distance traveled by the lane-changing vehicle in the lane-changing preparation phase t∈(0,t3), x1, x2, x0 represent the longitudinal positions of the vehicle behind the gap, the vehicle in front of the gap, and the lane-changing vehicle at time t=0, v1 and v2 represent the longitudinal speeds of the vehicle behind the gap and the vehicle in front of the gap, respectively. and SF j0,t3 They represent the longitudinal safety margins of the vehicle behind the gap and the lane-changing vehicle at time t3, l j0 and l j+1 They represent the lengths of the lane-changing vehicle and the vehicle in front of the gap respectively; The lane-changing vehicle should complete the lane change before reaching the obstacle, satisfying constraints (19)-(20), where x0 is the distance between the vehicle and the obstacle before the lane change; x end =v2(t4-t3) (19) x str +x end <x0 (20) The maximum velocity and maximum acceleration constraints are shown in equations (21)-(22). v′∈(v min ,v max ) (22) The maximum lateral acceleration is ac max , the lane-changing process must satisfy equation (23): ac≤ac max (23)。 5. The method according to claim 4, characterized in that The constructing of the autonomous driving vehicle driving interoperability hyperedge and the associated vehicle derived hyperedge according to the interoperability safety boundary constraint of the vehicle lane change further includes: The hyperedge in hypergraph theory is used to represent the non-paired relationship when the autonomous driving vehicle changes lanes. The vehicles are taken as hypergraph nodes. The lane-changing vehicles and the vehicles whose lane-changing intentions conflict with their own safety boundaries constitute interoperable hyperedges. The vehicle interoperable hypergraph is constructed. Each vehicle has an interoperable hyperedge during driving. The interoperable hyperedges include lane-changing interoperable hyperedges, following interoperable hyperedges, and free-driving hyperedges. When vehicles j and j+2 follow each other, the safety boundary and the front vehicle interact to form a convoy, then e1 = {C j ,C j+1 }、e2={C j+2 ,C j+3 } is a following interoperable hyperedge. When the lane-changing intention safety boundary of vehicle j0 interacts with vehicles j0+1 and j+2, the hyperedge e1 is constructed as a lane-changing interoperable hyperedge. vehicle and vehicles For free driving, build a free driving interoperable hyperedge and when After giving way to the vehicle changing lanes, the safety margin will be affected by the vehicle in front. The two vehicles have an interoperable relationship and build a following interoperable hyperedge This type of hyperedge is generated by the lane-changing intention and is a derived hyperedge of the lane-changing intention. Each interoperable hyperedge constitutes a game environment, and there is a sequence in the construction of hyperedges.

6. The method according to claim 5, characterized in that The autonomous driving vehicle playing games with all other vehicles respectively according to the interoperable hyperedge and the associated vehicle derived hyperedge to obtain a lane change plan executable by the autonomous driving vehicle includes: The multi-party game model is divided into multiple two-party games, namely, the two-party games between the target vehicle and each interacting vehicle. A multi-party game model is established for the vehicles in the interoperable hyperedge. The vehicle benefits are shown in formula (30), including individual benefits, group benefits and safety benefits, where λ E , G is the parameter; I j =λ E ·IE j +λ G ·IG j +IS j (30) In the formula, IE is individual benefit, as shown in formula (31). In the formula, j=j0 represents the lane-changing vehicle, and its individual benefits are divided into efficiency benefits and utility benefits. represents the average speed of the target lane, v′ represents the average speed of the lane where the lane-changing vehicle is located, and the speed gain is used. to represent the efficiency of vehicle lane change; x0 represents the obstacle position, Indicates the longitudinal position of the lane-changing vehicle at the time of decision making, using the distance between the lane-changing vehicle and the obstacle is used to represent the utility benefit of a vehicle changing lanes, γ is a known parameter, j = j′ represents the vehicle that gives way, and its individual benefit is expressed as the speed difference before and after the lane change. represents the speed of the vehicle giving way at the end time, v j,0 Indicates the speed of the vehicle giving way at the start time; IG is the group benefit. The greater the cascade effect caused by lane changing, the lower the group benefit of the subject vehicle, as shown in formula (32). Among them, j=j0 represents the lane-changing vehicle, and its group benefit is the negative value of the number of derived hyperedges caused by the lane change -|e|. When the number of derived hyperedges generated by the lane-changing intention of the lane-changing vehicle is too large, it means that the cascade effect of the intention on the traffic flow is large, and the negative group benefit caused is high. j=j′ represents the yielding vehicle. When yielding is a passive decision, the group benefit is 0. IS is the safety benefit, as shown in formula (33): in, A 0-1 variable indicating whether the safety margins of vehicles j′ and j0 affect each other under this strategy. When the two vehicles affect each other, Otherwise, it is 0. If the safety boundaries of the two vehicles affect each other, there will be a safety hazard. Let M be a very large number, so that the safety benefits of both parties are extremely low. The strategy set of lane-changing vehicles is S1 = {p1, p2}, where p1 represents the lane-changing strategy and p2 represents the non-lane-changing strategy. The strategy set of other vehicles is S2 = {q1, q2}, where q1 represents the yielding strategy and q2 represents the non-yielding strategy. Let the benefits obtained by the two parties in the game when adopting different strategies be I. It represents the benefit obtained by the lane-changing vehicle when the lane-changing vehicle adopts strategy p1 and the yielding vehicle adopts strategy q1; The payoff matrix of the lane-changing game is shown in Table 2: Table 2 Profit Matrix If there is a strategy (p n ,q n ) makes equation (34) true, then the strategy (p n ,q n ) is the Nash equilibrium point of the game; The point with the highest total benefit for both parties is selected as the final equilibrium solution. When (p1, q1) is the final equilibrium solution, the lane-changing vehicle successfully plays the game. The lane-changing vehicle will play games with all other vehicles respectively, and (p1, q1) in all game results is the final equilibrium solution. The vehicle can change lanes and output a lane-changing plan that can be executed by the autonomous driving vehicle. The lane-changing plan includes the decision on whether to change lanes and the trajectories of the lane-changing vehicle and surrounding vehicles.

Images