A Lane-Changing Trajectory Planning Method Based on Cyber-Physical Iterative Game

Abstract

The present invention discloses a lane - changing trajectory planning method based on cyber - physical iterative game, including: S1: Divide the longitudinal distance between the current lane and the target lane into N horizontal lines; S2: Define the revenue functions of the lane - changing vehicle and the competing vehicle; S3: Collect the driving state information of the vehicle itself and surrounding vehicles and road environment information, solve the evolutionary stable strategy to obtain the optimal lane - changing strategy. If it is to continue lane - changing, perform S4; if it is to maintain the lane, perform S5; otherwise, perform S6; S4: Generate a lane - changing trajectory according to the lane - changing strategy planned in S3; S5: Generate a lane - maintaining trajectory according to the lane - maintaining strategy planned in S3; S6: Generate a vehicle operation trajectory that moves to the lateral position at the previous moment according to the fallback strategy planned in S3; S7: Repeat the process of S3 - S6 until the lane - changing is successful. The present invention considers the traffic conditions at each moment, completes the optimal trajectory at each moment in a progressive form, and finally realizes the lane - changing trajectory planning for autonomous vehicles.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle intelligent driving or assisted driving, and particularly relates to a lane-changing trajectory planning method based on cyber-physical iterative game. Background Art

[0002] With the increasing number of automobiles year by year, traffic accidents and traffic congestion problems have become increasingly serious. Autonomous vehicles integrating multiple technologies such as communication, sensing, and computer have played a huge role in improving traffic safety, enhancing energy efficiency, and alleviating traffic congestion, and have attracted much attention from the academic and industrial circles. Active lane-changing decision-making is a basic task of autonomous driving. The safety and effectiveness of lane-changing decisions affect the safety and traffic efficiency of vehicles. Effective lane-changing decisions can improve the traffic efficiency of vehicles to a certain extent and reduce the occurrence of traffic accidents.

[0003] By consulting relevant patents and papers, it is found that most of the existing lane-changing trajectory planning is a static trajectory planning model, that is, it is assumed that the speeds of surrounding vehicles do not change during the entire lane-changing process, and the vehicle itself does not change according to the real-time changes of surrounding vehicles, which is inconsistent with the traffic characteristics of the real world. Patent CN113276848B discloses a method for generating multiple lane-changing trajectories based on the collected roadside information and the vehicle state information of the vehicle itself, and selecting an optimal trajectory therefrom; Patent Application CN114852105A discloses a method for generating a lane-changing trajectory by collecting vehicle information for game-playing and taking the lane-changing efficiency and fuel economy as the goals. However, neither of the two patents considers the dynamic changes of other vehicles, and the actual application effect is poor. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a lane-changing trajectory planning method based on cyber-physical iterative game.

[0005] The purpose of the present invention is achieved by the following technical solutions:

[0006] A lane-changing trajectory planning method based on cyber-physical iterative game, comprising the following steps:

[0007] S1: Divide the lateral distance between the current lane and the target lane into N horizontal lines along the vehicle driving direction, and store the ordinate y of each horizontal line i in an array L as the nodes for the vehicle to adjust the lane-changing trajectory, where the current lane refers to the lane where the lane-changing vehicle is currently located. Taking the current position of the lane-changing vehicle as the origin, the driving direction as the X-axis, and establishing a Cartesian coordinate system in the direction perpendicular to the X-axis from the current lane to the target lane, and setting N horizontal lines at equal intervals along the Y-axis direction. The ordinate y of each horizontal line i , where i represents the serial number of each horizontal line;

[0008] S2: Define the revenue functions of the lane-changing vehicle and the competing vehicles, where the competing vehicles refer to the vehicles in the target lane that may compete for position with the lane-changing vehicle.

[0009] S3: Collect the driving state information and road environment information of the lane-changing vehicle and the competing vehicles, and solve for the evolutionary stable strategy. If the optimal strategy obtained is to continue lane-changing, then proceed to S4; if it is to maintain the lane, then proceed to S5; if it is to reverse, then proceed to S6.

[0010] S4: Generate the optimal cubic polynomial lane-changing trajectory according to the lane-changing strategy planned in S3.

[0011] S5: Generate the lane-keeping trajectory according to the lane-keeping strategy planned in S3.

[0012] S6: Generate the vehicle driving trajectory to the lateral position at the previous moment according to the reverse strategy planned in S3.

[0013] S7: Repeat the process of S3 - S6 until the lane change is successful.

[0014] Furthermore, the specific content of S2 is as follows:

[0015] S21: Define the lane-changing vehicle as vehicle No. 1, the vehicle behind in the target lane as vehicle No. 2, also known as the competing vehicle, the vehicle in front in the target lane as vehicle No. 3, and the vehicle in front of the lane-changing vehicle as vehicle No. 4.

[0016] S22: Calculate the revenue of vehicle No. 1, and its revenue function is:

[0017]

[0018] where β is the weight coefficient, representing the aggressiveness of the driver, and here it is taken as 0.4, U safety1 is the safety revenue of vehicle No. 1, and U velocity is the speed revenue of vehicle No. 1;

[0019]

[0020]

[0021]

[0022] where d is the lateral distance between vehicle No. 1 and vehicle No. 2, l is the vehicle width, generally taken as 1.6 meters, a represents the safe lateral distance between the two vehicles, generally taken as 1 meter, u is the safety factor, T headway is the headway of vehicle No. 1 relative to vehicle No. 2, T1 represents the headway between vehicle No. 1 and vehicle No. 4, T2 represents the headway between vehicle No. 2 and vehicle No. 3, and T e1 represents the desired headway of vehicle No. 1, which can be obtained from T e1= min(3, T1), where T1 represents the headway between Vehicle 1 and Vehicle 4, v3 is the speed of Vehicle 3, and v4 is the speed of Vehicle 4;

[0023] S23: Calculate the revenue of Vehicle 2, and its revenue function is:

[0024]

[0025] where, U safety2 is the safety revenue of Vehicle 2, and U space is the spatial revenue of Vehicle 2;

[0026]

[0027]

[0028]

[0029] where, T e2 is the expected headway of Vehicle 2.

[0030] Further, the specific steps of S3 are as follows:

[0031] S31: Determine the game revenue matrix of Vehicle 1 and Vehicle 2 in each game strategy according to the revenue functions determined by S22 and S23 where, q = {1, 2} respectively represent Vehicle 1 and Vehicle 2, x = {1, 2}, y = {1, 2, 3}, n1 represents Vehicle 2 accelerating, n2 represents Vehicle 2 decelerating, m1 represents Vehicle 1 changing lanes, m2 represents Vehicle 1 driving along the current horizontal line y i traveling, m3 represents Vehicle 1 retreating to the longitudinal position at the previous moment, (n x , m y ) represents the game strategy; When Vehicle 1 and Vehicle 2 conduct a lane-changing game, six game results will be obtained, which are: Vehicle 1 changes lanes and Vehicle 2 in the lane decelerates; Vehicle 1 changes lanes and Vehicle 2 in the lane accelerates; Vehicle 1 keeps its lane and Vehicle 2 in the lane accelerates; Vehicle 1 keeps its lane and Vehicle 2 in the lane decelerates; Vehicle 1 retreats and Vehicle 2 accelerates; Vehicle 1 retreats and Vehicle 2 decelerates;

[0032] S32: Use evolutionary game to solve the revenue matrix and find the evolutionary stable strategy of the matrix. The replication dynamic equation is as follows:

[0033]

[0034] In the formula, s j represents the strategy set of Vehicle j, and z j represents the proportion of the population choosing the s j strategy at time t, e(s j) Represents the expected payoff of individual choice s j The expected payoff, and e(s) represents the average expected payoff of all strategy sets s of the individual;

[0035] S33: Add the minimum distance between vehicles to the constraint conditions, and set the minimum acceptable time headway between two vehicles to 1 s. Then we have:

[0036]

[0037] Among them, T safety Is the safe time headway. If the safety payoff does not satisfy U safety1 ≥T safety , Execute the lane keeping operation.

[0038] Furthermore, the specific content of S4 is as follows:

[0039] S41: Fit the lane change trajectory through a cubic polynomial as follows:

[0040]

[0041] Among them, x(t) and y(t) are the longitudinal position and lateral position of the vehicle at time t, and a0, a1, a2, a3, b0, b1, b2, b3 are the coefficients of the curve polynomial;

[0042] S42: At each time step, update the coordinate system with the current vehicle position as the origin, and assume that the speed v is constant. Then we have:

[0043] x(0) = 0, x(t total ) = X, x'(0) = vcosθ i , x'(t total ) = v

[0044] y(0) = 0, y(t total ) = Y, y'(0) = vsinθ i , y'(t total ) = 0

[0045] Finally, we have:

[0046]

[0047] Among them, t toatal Is the total lane change time of the vehicle, θ i Is the vehicle heading angle at the current moment, X is the longitudinal displacement of the vehicle lane change trajectory, and Y is the lateral displacement of the vehicle lane change trajectory;

[0048] S43: Optimal path planning needs to consider both the comfort and efficiency of lane change. Then define the lane change cost function J as follows:

[0049] J = βatotal +(1 - β)t total

[0050] Where a total is the tangential acceleration at the end of lane change, at which time the lane - changing trajectory has the maximum curvature, t total is the time taken to complete the lane change, β is the aggressiveness of the driver, taking 0.4 here, then a total represents the driver's comfort, and t total represents the efficiency of lane change;

[0051]

[0052] Where J * is the minimum value of the cost function, t * total is the time required for the optimal lane - changing trajectory, y”(t total ) represents taking t as t after taking the second - order derivative of y(t), total , y'(t total ) represents taking t as t after taking the first - order derivative of y(t), total , x”(t total ) represents taking t as t after taking the second - order derivative of x(t), total , x'(t total ) represents taking t as t after taking the first - order derivative of x(t), total , so the longitudinal displacement X of lane change is:

[0053] X = f(v, t total , θ i , y total ),

[0054] where f(·) represents the solution process of the longitudinal displacement X when the lane - changing cost function takes the minimum value, and y total represents y(t total ).

[0055] Furthermore, the S5 includes instructing the lane - changing vehicle to continue driving at the current lateral position.

[0056] Furthermore, the S6 further includes: updating the Y value to the lateral displacement required to complete the vehicle lane change when the No. 1 vehicle is at the lateral position at the previous moment.

[0057] The beneficial effects of the present invention are as follows: The lane-changing process is discretized. The information of surrounding vehicles is collected at each time node, and based on the collected information, the optimal lane-changing decision is found through the form of vehicle game. If the lane-changing condition is met, the optimal lane-changing trajectory is planned by a cubic polynomial, otherwise, the vehicle maintains its lane at the current position or returns to the position at the previous moment. The vehicle can react in real time according to the dynamic changes of other vehicles during the lane-changing process, ensuring the safety and efficiency of the lane-changing process.

[0058] Other advantages, objectives, and features of the present invention will be described to some extent in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the following specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings, where:

[0060] Figure 1 is the flowchart of the present invention;

[0061] Figure 2 is the lane-changing trajectory diagram of the lane-changing trajectory planning method based on cyber-physical iterative game;

[0062] Figure 3 is the optimal trajectory of each plan of the lane-changing trajectory planning method based on cyber-physical iterative game. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0063] The following will refer to the accompanying drawings to describe the preferred embodiments of the present invention in detail. It should be understood that the preferred embodiments are only for illustrating the present invention, rather than for limiting the protection scope of the present invention.

[0064] This embodiment proposes a lane-changing trajectory planning method based on cyber-physical iterative game. The perception of environmental information is realized through sensors such as on-vehicle cameras, lidars, and millimeter-wave radars carried by autonomous vehicles. Considering the position and speed information of the vehicle in front in the current lane, the vehicle in front and the vehicle behind in the target lane, the optimal strategy is given through the dynamic game between vehicles, and the optimal trajectory of the vehicle is planned based on different strategies considering the comfort of the driver and the lane-changing efficiency.

[0065] As Figure 1 shown, this embodiment may include the following steps:

[0066] S1: Collect the lateral position of the current vehicle of the host vehicle from the center line of the target lane through the sensors of the host vehicle. Taking the current vehicle position as the origin, the driving direction as the X-axis, establish a Cartesian coordinate system with the direction perpendicular to the X-axis from the current lane to the target lane, and select N horizontal lines (such as Figure 2 shown by the dashed line) at equal intervals along the Y-axis direction as the vehicle lane-changing nodes S i , and save the ordinates of the N horizontal lines in the array L = {S0, S1,..., S N-1}. Among them, the host vehicle refers to the vehicle to perform lane change, that is, the lane-changing vehicle. The current lane refers to the lane where the host vehicle is currently located at the current time, and the target lane refers to the lane that the host vehicle wants to switch to from the current lane.

[0067] S2: For the purpose of improving the lane-changing efficiency and ensuring the lane-changing safety, establish a benefit function reflecting vehicle safety and driving efficiency. The benefit functions of the lane-changing vehicle and the competing vehicle are defined as follows. The following takes the Figure 2 state shown as an example for illustration. In Figure 2 , Car 1 represents the lane-changing vehicle, also known as Vehicle No. 1 or the host vehicle; Car 2 represents the vehicle behind in the target lane, also known as Vehicle No. 2 or the competing vehicle; Car 3 represents the vehicle in front in the target lane, also known as Vehicle No. 3. When Vehicle No. 1 changes lanes from the current lane, the target position is between Vehicle No. 2 and Vehicle No. 3; Car4 represents the vehicle in front of the lane-changing vehicle.

[0068] For the lane-changing vehicle, its benefit function can be expressed as:

[0069]

[0070] Among them, β is the weight coefficient, representing the aggressiveness of the driver, U safety1 is the longitudinal safety benefit of the lane-changing vehicle, U velocity is the speed benefit of the lane-changing vehicle; among them,

[0071]

[0072]

[0073]

[0074] Among them, d is the lateral distance between Vehicle No. 1 and Vehicle No. 2, l is the vehicle width, generally taken as 1.6 meters, a represents the safe lateral distance between the two vehicles, generally taken as 1 meter, u is the safety factor, T headway is the headway of Vehicle No. 1 relative to Vehicle No. 2, T1 represents the headway between Vehicle No. 1 and Vehicle 4, T2 represents the headway between Vehicle No. 2 and Vehicle 3, T e1 represents the desired headway of Vehicle No. 1, which can be obtained from T e1It is obtained by = min(3, T1), where T1 represents the headway between Vehicle 1 and Vehicle 4, v3 is the speed of Vehicle 3, and v4 is the speed of Vehicle 4.

[0075] For the competing vehicle (i.e., Figure 2 Vehicle 2 in ), its revenue function

[0076]

[0077] can be expressed as: safety2 where β is the weight coefficient representing the aggressiveness of the driver, U space is the safety revenue of the competing vehicle, and U

[0078]

[0079]

[0080]

[0081] where T e2 is the expected headway of Vehicle 2.

[0082] S3: Obtain the game revenue of each lane-changing strategy according to the game revenue calculation formula in S2, list the game revenue matrix, and the strategy with the maximum revenue in the game revenue matrix is the current game lane-changing decision.

[0083] The game revenue matrix of Vehicle 1 and Vehicle 2 in each game strategy is determined by the revenue functions determined in S22 and S23 where q = {1, 2} represents Vehicle 1 and Vehicle 2 respectively, x = {1, 2}, y = {1, 2, 3}, n1 represents Vehicle 2 accelerating, n2 represents Vehicle 2 decelerating, m1 represents Vehicle 1 changing lanes, m2 represents Vehicle 1 driving along the current horizontal line y i traveling, m3 represents Vehicle 1 retreating to the longitudinal position at the previous moment, and (n x , m y ) represents the game strategy; when Vehicle 1 and Vehicle 2 conduct a lane-changing game, six game results will be obtained, namely: Vehicle 1 changes lanes and Vehicle 2 in the adjacent lane decelerates; Vehicle 1 changes lanes and Vehicle 2 in the adjacent lane accelerates; Vehicle 1 keeps its lane and Vehicle 2 in the adjacent lane accelerates; Vehicle 1 keeps its lane and Vehicle 2 in the adjacent lane decelerates; Vehicle 1 retreats and Vehicle 2 accelerates; Vehicle 1 retreats and Vehicle 2 decelerates.

[0084] The game revenue matrix can be shown in Table 1 below.

[0085] Table 1 Game Revenue Matrix

[0086]

[0087] Calculate the revenue values of lane-changing vehicles and competing vehicles considering safety and timeliness under each strategy to obtain the game revenue matrix. Use evolutionary game to solve the revenue matrix and find the evolutionary stable strategy of the matrix. The replicator dynamics equation is as follows:

[0088]

[0089] In the formula, s j represents the strategy set of vehicle No. j, and z j represents the proportion of the group choosing strategy s j at time t. e(s j ) represents the expected revenue of an individual choosing s j , and e(s) represents the average expected revenue of all strategy sets s of an individual.

[0090] Add the minimum distance between vehicles to the constraint conditions, and set the minimum acceptable headway between two vehicles to 1 s. Then we have:

[0091]

[0092] U safety1 ≥T safety

[0093] where T safety is the safe headway. If the safety revenue does not satisfy U safety1 ≥T safety , perform the lane-keeping operation.

[0094] S4: When the vehicle strategy is to change lanes, the lane-changing trajectory can be fitted by a cubic polynomial. The cubic polynomial is as follows:

[0095]

[0096] where x(t) and y(t) are the longitudinal position and lateral position of the vehicle at time t, and a0, a1, a2, a3, b0, b1, b2, b3 are the coefficients of the curve polynomial. At each time step, update the coordinate system with the current vehicle position as the origin, and assume that the speed v is constant. Then we have:

[0097] x(0) = 0, x(t total ) = X, x'(0) = vcosθ i , x'(t total ) = v

[0098] y(0) = 0, y(t total ) = Y, y'(0) = vsinθ i , y'(t total ) = 0

[0099] Finally, there is:

[0100]

[0101] wherein, t toatal is the total vehicle lane - changing time, θ i is the vehicle heading angle at the current moment (i.e., when the lateral position of the vehicle is at y i ), X is the lateral displacement of the vehicle lane - changing trajectory, and Y is the longitudinal displacement of the vehicle lane - changing trajectory; considering that θ i , v, and Y are all known quantities, so X and t total determine the vehicle's driving trajectory. Figure 3 represents the driving trajectories determined at different time points. It should be noted that Figure 3 this is only exemplary and does not limit the present invention.

[0102] Optimal path planning needs to consider both the comfort and efficiency of lane - changing. The lane - changing cost function can be defined as follows:

[0103] J = βa total +(1 - β)t total

[0104] wherein, a total is the tangential acceleration at the end of lane - changing, at which time the lane - changing trajectory has the maximum curvature, t total is the time taken to complete lane - changing, β is the aggressiveness of the driver, taking 0.4 here, then a total can represent the driver's comfort, and t total can represent the efficiency of lane - changing;

[0105]

[0106] wherein, J * is the minimum value of the cost function, is the time required for the optimal lane - changing trajectory, y”(t total ) represents taking t as t total after taking the second - order derivative of y(t), y'(t total ) represents taking t as t total after taking the second - order derivative of y(t), x”(t total ) represents taking t as t total after taking the second - order derivative of x(t), x'(t total ) represents taking t as t total after taking the first - order derivative of x(t). Thus, the longitudinal distance X for lane - changing is:

[0107] X = f(v, t total , θ i , y total ),

[0108] f(·) represents the solution process of the longitudinal displacement X when the lane-changing cost function takes the minimum value, and y total represents y(t total ).

[0109] The lane-changing vehicle continues to travel at the current lateral position.

[0110] Further, step S6 further includes: updating the Y value to the remaining lateral displacement of vehicle No. 1 during lane-changing at the previous moment.

[0111] S5: When it is currently determined that the optimal strategy is lane keeping, generate a lane-keeping trajectory and make the vehicle (i.e., vehicle No. 1) travel straight at the current lateral position.

[0112] S6: When it is determined that the optimal strategy is to retreat, generate an instruction to make the vehicle (i.e., vehicle No. 1) retreat to the lateral position at the previous moment, and generate the following cubic polynomial lane-changing trajectory:

[0113]

[0114] where Y1 represents the lateral displacement required for the vehicle to change lanes when vehicle No. 1 is at the lateral position at the previous moment. According to the cost function defined above, the optimal longitudinal distance X1 in the case of retreat can be obtained as:

[0115] X1 = f(v, t total , θ i , y total ).

[0116] S7: Repeat the above steps S3 - S6 until the lane change is successful.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the purpose and scope of the present technical solution, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A lane-changing trajectory planning method based on cyber-physical iterative game, characterized in that Including: S1: Divide the lateral distance between the current lane and the target lane into N horizontal lines along the vehicle's driving direction, and store the ordinate y of each horizontal line i in the array L as the nodes for the vehicle to adjust the lane-changing trajectory. Here, the current lane refers to the lane where the lane-changing vehicle is currently located. Taking the current position of the lane-changing vehicle as the origin, the driving direction as the X-axis, a Cartesian coordinate system is established in the direction perpendicular to the X-axis from the current lane to the target lane, and N horizontal lines are equally spaced along the Y-axis direction. The ordinate y of each horizontal line i , where i represents the serial number of each horizontal line; S2: Define the revenue functions of the lane-changing vehicle and the competing vehicles, where the competing vehicles refer to the vehicles in the target lane that may compete for positions with the lane-changing vehicle; S3: Collect the driving state information and road environment information of the lane-changing vehicle and the competing vehicles, and solve the evolutionary stable strategy. If the obtained optimal strategy is to continue lane-changing, then proceed to S4; if it is to maintain the lane, then proceed to S5; if it is to retreat, then proceed to S6; S4: Generate the optimal cubic polynomial lane-changing trajectory according to the lane-changing strategy planned in S3; S5: Generate the lane-keeping trajectory according to the lane-keeping strategy planned in S3; S6: Generate the vehicle driving trajectory to the lateral position at the previous moment according to the retreat strategy planned in S3; S7: Repeat the process of S3 - S6 until the lane change is successful.

2. The lane-changing trajectory planning method based on cyber-physical iterative game according to claim 1, characterized in that: The specific content of S2 is as follows: S21: Define the lane-changing vehicle as vehicle No. 1, the vehicle behind in the target lane as vehicle No. 2, also known as the competing vehicle, the vehicle in front in the target lane as vehicle No. 3, and the vehicle in front of the lane-changing vehicle as vehicle No. 4; S22: Calculate the revenue of vehicle No. 1, and its revenue function is: Among them, β is the weight coefficient, representing the aggressiveness of the driver, which is taken as 0.4 here, and U safety1 is the safety benefit of vehicle No. 1, and U velocity is the speed benefit of vehicle No. 1; Among them, d is the lateral distance between vehicle No. 1 and vehicle No. 2, l is the vehicle width, generally taken as 1.6 meters, a represents the safe lateral distance between the two vehicles, generally taken as 1 meter, u is the safety factor, T headway is the headway of vehicle No. 1 relative to the front of vehicle No. 2, T1 represents the headway between vehicle No. 1 and vehicle No. 4, T2 represents the headway between vehicle No. 2 and vehicle No. 3, T e1 represents the desired headway of vehicle No. 1, which can be obtained from T e1 = min(3, T1), where T1 represents the headway between vehicle No. 1 and vehicle No. 4, v3 is the speed of vehicle No. 3, and v4 is the speed of vehicle No. 4; S23: Calculate the revenue of vehicle No. 2, and its revenue function is: Among them, U safety2 is the safety benefit of vehicle No. 2, and U space is the space benefit of vehicle No. 2; Among them, T e2 is the expected headway of vehicle No.

2.

3. The lane-changing trajectory planning method based on cyber-physical iterative game according to claim 2, characterized in that: The specific content of S3 is as follows: S31: Determine the game payoff matrix of Vehicle 1 and Vehicle 2 in each game strategy according to the payoff function determined by S22 and S23 where q = {1, 2} respectively represent Vehicle 1 and Vehicle 2, x = {1, 2}, y = {1, 2, 3}, n1 represents that Vehicle 2 accelerates, n2 represents that Vehicle 2 decelerates, m1 represents that Vehicle 1 changes lanes, m2 represents that Vehicle 1 travels along the current horizontal line y i and m3 represents that Vehicle 1 retreats to the longitudinal position at the previous moment. (n x , m y ) represents the game strategy. When Vehicle 1 and Vehicle 2 conduct a lane-changing game, six game results will be obtained, namely: Vehicle 1 changes lanes and Vehicle 2 in the adjacent lane decelerates; Vehicle 1 changes lanes and Vehicle 2 in the adjacent lane accelerates; Vehicle 1 keeps its lane and Vehicle 2 in the adjacent lane accelerates; Vehicle 1 keeps its lane and Vehicle 2 in the adjacent lane decelerates; Vehicle 1 retreats and Vehicle 2 accelerates; Vehicle 1 retreats and Vehicle 2 decelerates; S32: Use evolutionary game theory to solve the revenue matrix and find the evolutionary stable strategy of the matrix. The replication dynamic equation is as follows: where s j represents the set of strategies of the j-th vehicle, and z j represents the proportion of the group choosing strategy s j at time t, and e(s j ) represents the expected payoff of an individual choosing s j , and e(s) represents the average expected payoff of all strategy sets s of an individual; S33: Add the minimum distance between vehicles to the constraint conditions, and set the minimum acceptable headway between two vehicles to 1 s. Thus, we have: U safety1 ≥T safety Among them, T safety is the safe headway. If the safety benefit does not satisfy U safety1 ≥ T safety , perform the lane keeping operation.

4. The lane-changing trajectory planning method based on cyber-physical iterative game according to claim 3, characterized in that: The specific content of S4 is as follows: S41: Fit the lane-changing trajectory through a cubic polynomial as follows: where x(t) and y(t) are the longitudinal and lateral positions of the vehicle at time t, and a0, a1, a2, a3, b0, b1, b2, b3 are the coefficients of the curve polynomial; S42: At each time step, update the coordinate system with the current vehicle position as the origin, and assume a constant speed v. Thus, we have: x(0) = 0, x(t total ) = X, x'(0) = vcosθ i , x'(t total ) = v y(0) = 0, y(t total ) = Y, y'(0) = v sinθ i , y'(t total ) = 0 Finally, we have: where t total is the total vehicle lane - changing time, θ i is the vehicle heading angle at the current moment, X is the longitudinal displacement of the vehicle lane - changing trajectory, and Y is the lateral displacement of the vehicle lane - changing trajectory; S43: The optimal path planning needs to consider both the comfort and efficiency of lane-changing. Then define the lane-changing cost function J as follows: J = βa total +(1 - β)t total where a total is the tangential acceleration at the end of lane change, at which the lane change trajectory has the maximum curvature, and β is the aggressiveness of the driver, which is taken as 0.4 here; Among them, J * is the minimum value of the cost function, is the time required for the optimal lane-changing trajectory, y”(t total ) represents taking t total after taking the second derivative of y(t), y'(t total ) represents taking t total after taking the first derivative of y(t), x”(t total ) represents taking t total after taking the second derivative of x(t), x'(t total ) represents taking t total after taking the first derivative of x(t), and thus the longitudinal displacement X of lane changing is: X = f(v, t total , θ i , y total ), Among them, f(·) represents the solution process of the longitudinal displacement X when the lane-changing cost function takes the minimum value, and y total represents y(t total ).

5. The lane-changing trajectory planning method based on cyber-physical iterative game according to claim 4, characterized in that S5 includes instructing the lane-changing vehicle to continue driving at the current lateral position.

6. The lane-changing trajectory planning method based on the cyber-physical iterative game according to claim 4, characterized in that S6 also includes: Update the Y value to the lateral displacement required for lane-changing when the lateral position where vehicle No. 1 was located at the previous moment is reached.

Images