A humanized lane change decision method

Abstract

The present application belongs to the field of automobile technology, and specifically relates to a kind of anthropomorphic lane change decision method. It includes: 1. Selecting the optimal lane; 2. If lane change intention occurs, selecting one of the three behaviors of waiting to change lanes, overtaking and changing lanes, and generating the path and speed of the host vehicle according to the state of the vehicle; 3. Evaluating the benefit index, which includes speed benefit and safety benefit. The present application divides the lane change process into lane change execution stages 1 and 2 and lane change preparation stage, and considers the overtaking and waiting to change lanes driving scheme into the lane change decision system, providing different driving schemes for automatic lane change. The determination process of the lane change timing is determined through event triggering, and finally a dynamic game model is established between the host vehicle and the following vehicle through the determination of the potential conflict area, ensuring the safe and efficient interaction between the two vehicles.

Description

Technical Field

[0001] This invention belongs to the field of automotive technology, specifically a human-like lane-changing decision-making method. Background Technology

[0002] With the rapid development of intelligent vehicles, lane changing is an essential driving behavior in intelligent vehicle decision-making systems. Human-like lane-changing decision-making is crucial for intelligent vehicles to drive efficiently in urban environments. Furthermore, the decision-making model needs to ensure the interaction between vehicles in complex traffic conditions.

[0003] Existing lane-changing decision-making methods are insufficient in determining the timing of returning to the original lane when changing lanes. Furthermore, they do not consider whether overtaking is possible when changing lanes, and rarely consider the behavior of slowing down to change lanes when there is insufficient space ahead. Summary of the Invention

[0004] This invention provides a human-like lane-changing decision-making method. The method divides the lane-changing process into three stages: lane-changing execution stage 1, lane-changing execution stage 2, and lane-changing preparation stage. It incorporates overtaking and waiting-to-change driving options into the lane-changing decision-making system and proposes conditions for selecting overtaking or waiting-to-change based on the current traffic scenario. This provides different driving options for autonomous driving lane-changing. The timing of lane-changing is determined through event triggering. Finally, by identifying potential conflict areas, a dynamic game model is established with the following vehicle to ensure safe and efficient interaction between the two vehicles.

[0005] The technical solution of this invention is described below in conjunction with the accompanying drawings:

[0006] A human-like lane-changing decision-making method includes the following steps:

[0007] Step 1: Select the optimal lane;

[0008] Step 2: If a lane-changing intention is generated, select one of the three behaviors: waiting to change lanes, overtaking, or changing lanes, and generate the path and speed of the main vehicle based on the vehicle's current state.

[0009] Step 3: Evaluate the benefit indicators, which include speed benefits and security benefits.

[0010] Furthermore, the specific method for step one is as follows:

[0011] 11) Select lanes based on the distance between the main vehicle and other road users in each lane ahead, and the speed of the vehicles ahead;

[0012] The lane assessment is obtained by the following formula:

[0013]

[0014]

[0015]

[0016]

[0017] In the formula, The speed evaluation function for candidate lanes; V is the speed of the nearest road user in the waiting lane in front of the main vehicle. traffic The speed limit is set by traffic regulations; Candidate lane ID is the candidate lane ID. For the spatial evaluation function of candidate lanes; D represents the distance between the nearest traffic participant in the candidate lane, located in front of the driver, and the driver vehicle. sensor The distance within the sensing range; i is the driving load function for the candidate lanes; lane The candidate lane number is used for evaluation; The lane number where the main vehicle is located; This is the sum of the cost functions of the candidate lanes;

[0018] 12) The candidate lane with the largest sum of substitution functions is the target lane Opt ID, and it is used as input to the decision of driving behavior.

[0019] Furthermore, the specific method for step two is as follows:

[0020] When an intention to change lanes is made, if there is insufficient space to change lanes, the vehicle should wait to change lanes; if there is sufficient space to change lanes, the vehicle should change lanes; if the lead vehicle is behind other interacting traffic participants, the vehicle should overtake.

[0021] Furthermore, when choosing to wait for a lane change, the main vehicle's path is along the centerline of the main lane, and the main vehicle's speed is calculated using the following formula:

[0022]

[0023] V wait =max[V wait V traffic ]

[0024] In the formula, D F1 The distance from the main vehicle to the vehicle in front; V wait The speed planned for the waiting lane-changing phase; t Delay The driver's braking reaction time; a pdThe acceleration of the main vehicle during braking; V F1 The speed of the vehicle being overtaken; a pdmax D represents the maximum braking acceleration of the vehicle being overtaken. stop V is the distance between the two vehicles after braking ends. traffic Speeds regulated by law.

[0025] Furthermore, when the lane-changing action is selected, lane-changing preparation and lane-changing execution actions are performed;

[0026] During the lane-changing preparation phase, the main vehicle's path is along the centerline of the main lane; the space for preparing to change lanes is D. pre =V ego t pre In the formula, V ego The speed of the main vehicle; t pre Preparation time for changing lanes;

[0027] When executing the lane change action, the required distance for lane change is first calculated. Once the lane change distance of the main vehicle meets the required distance, the lane change action is executed.

[0028] The distance required for a lane change is calculated using the following formula;

[0029]

[0030] In the formula, D LC V represents the space required for lane changing. ego The speed of the main vehicle; T LC The lane-changing time for the main vehicle is set to 4 seconds; D LC,limit The minimum space required for lane changing;

[0031] The global coordinate system for the start and end of a lane change is obtained using the following formula:

[0032]

[0033]

[0034] In the formula, This is the starting position for lane changing; S ego The projection of the main vehicle's position onto the road coordinate system; This is the end position of the lane change;

[0035] Determine the starting position of the lane change. and the end position Then, the lane-changing path is generated using a cubic polynomial, obtained through the following formula:

[0036] y = a3x 3 +a2x 2 +a1x+a0

[0037]

[0038]

[0039]

[0040] In the formula, x0 is the x-coordinate of the starting position of the lane change; y0 is the y-coordinate of the starting position of the lane change. x is the heading angle at the starting position of the lane change; d y is the x-coordinate of the lane change endpoint; d The ordinate of the lane change endpoint; The heading angle at the end of the lane change;

[0041] The lane change execution phase includes lane change execution phase 1 and lane change execution phase 2;

[0042] During lane-changing execution phase 1, before the primary vehicle enters the desired lane 1, it engages in a game of strategy with the following vehicle B1 in lane 1 to determine whether to contest or yield the right-of-way. If the primary vehicle is in lane-changing execution phase 1 and the outcome of the game is yielding, it will return to its original lane, thus avoiding a collision. The primary vehicle's entry into the adjacent lane requires D... A >W lane / 2, where D A The distance between the main vehicle's corner point A and the lane center reference line; W lane The width of the lane currently occupied by the main vehicle;

[0043] The distance required to return to the original lane is D. MB =V ego ·T MB In the formula, S ego The position of the main vehicle projected onto the road coordinate system; T MB Time to return to the original lane;

[0044] Return to the starting position of the original lane and the end position of returning to the original lane as follows:

[0045]

[0046]

[0047] In the formula, x ego The x-coordinate of the current position of the main vehicle; y ego The vertical coordinate of the current position of the main vehicle; The heading angle of the main vehicle's current position;

[0048] When returning to the original lane, the speed is determined based on the position and movement of traffic participants in the original lane; specifically as follows:

[0049]

[0050] V RB =max[V RB V traffic ]

[0051] In the formula, D F1 Distance to the vehicle in front in the main lane; V RB Speed ​​planning for returning to the original lane;

[0052] In the second phase of lane change execution, the lead vehicle has entered the potential conflict zone and no longer engages in a game with the following vehicle.

[0053] The speed of the main vehicle is calculated using the following formula:

[0054]

[0055] V wait =max[V wait V traffic ]

[0056] In the formula, D F1 The distance from the main vehicle to the vehicle in front; V wait The speed planned for the waiting lane-changing phase; t Delay The driver's braking reaction time; a pd The acceleration of the main vehicle during braking; V F1 The speed of the vehicle being overtaken; a pdmax D represents the maximum braking acceleration of the vehicle being overtaken. stop V is the distance between the two vehicles after braking ends. traffic Speeds regulated by law.

[0057] Furthermore, when choosing to overtake, the center line of the driver's lane is planned as a reference path; a target position S is first set. target ; Reach the target location S target Then, the process shifts to lane change preparation and lane change execution.

[0058] The target location S target The following formula can be used to obtain:

[0059] D Bsafe =L car +D F1safe

[0060]

[0061] S target =S F1 +D Bsafe

[0062] In the formula, D Bsafe The safe distance between the main vehicle's current speed and the vehicle being overtaken; L car D is the vehicle profile length of the vehicle being overtaken; F1safe The safe distance between the main vehicle's current speed and the vehicle being overtaken; V Opt The speed chosen by the main vehicle for overtaking; S F1 The projection of the overtaken vehicle onto the road coordinate system;

[0063] Speed ​​planning during the overtaking phase is obtained using the following formula:

[0064]

[0065] V overtaking =max[V overtaking V traffic ]

[0066] In the formula, D F2 The distance between the main vehicle and the nearest vehicle in front in the main lane; V overtaking The speed planned for overtaking; V F2 The speed of the vehicle closest to the main vehicle in front of the main lane.

[0067] Furthermore, the method for determining the overtaking behavior is as follows:

[0068] If the lead vehicle intends to change lanes to the left and the traffic situation ahead allows for overtaking, it will attempt to overtake and will not choose to change lanes to follow behind vehicle F1. When the lead vehicle is behind an interacting traffic participant, it will attempt to overtake.

[0069] When there are vehicles F2 and F3 in the main lane and the overtaking lane, the speed of vehicles F2 and F3 and the space of the target position will limit the possibility of reaching the target overtaking position, so overtaking is not chosen.

[0070] Overtaking acceleration is constrained in [0, a] max Within the range of constraints, the acceleration within the constraint range is analyzed by n-level separation to determine whether there is a feasible solution within the range of overtaking acceleration. If there is, overtaking is selected; otherwise, overtaking is not selected.

[0071] The following formula can be used to calculate the time required for the main vehicle to reach the target position under different accelerations for overtaking;

[0072] V opt=V ego +a opt t

[0073] S target -S ego =(V ego -V F1 )t+(a opt -a F1 )t 2 / 2

[0074] In the formula, V opt The speed at which the main vehicle reaches the target location; V ego The speed of the main vehicle; a opt The overtaking acceleration selected by the main vehicle; t is the overtaking time; a F1 The longitudinal acceleration of the overtaken vehicle F1;

[0075] Due to the constraints imposed by vehicles F2 and F3 ahead, although the main vehicle can reach the target position and overtake the overtaken vehicle F1 in time t, it still needs to maintain a safe distance from the vehicles ahead. If the constraint imposed by the vehicles ahead is not met, it indicates that the selected acceleration a... opt It is not feasible; only when the feasible acceleration space is traversed and a feasible acceleration exists can the overtaking behavior be determined, and the feasibility of the overtaking behavior is judged by Algorithm 1.

[0076] The judgment method of Algorithm 1 is as follows:

[0077] 11) Calculate the target position S target ;

[0078] 12) If the projection S of the position of traffic vehicle F3 onto the road coordinate system F3 >S target Furthermore, the projection S of the position of vehicle F2 onto the road coordinate system F2 >S target Then proceed to step 13);

[0079] 13) By selecting the acceleration a for overtaking... i Calculate the choice a i Overtaking time t under overtaking acceleration i , where a i ∈[0,a max (i = 1, 2, ..., n)

[0080] 14) Calculate in, This represents the projection of the current vehicle's position onto the road coordinate system. t represents the longitudinal speed of the main vehicle at the current moment. i To choose a iOvertaking time under overtaking acceleration; This represents the longitudinal acceleration of the main vehicle at the current moment;

[0081] 15) Through get

[0082] 16) Through get in, For vehicle F2, the overtaking time t i Then, the projection of the location onto the road coordinate system; For overtaking time t i Then, the projection of the position of the main vehicle onto the road coordinate system;

[0083] 17) Through get in, For traffic vehicle F3 during overtaking time t i Then, the projection of the location onto the road coordinate system;

[0084] 18) Through get in, For overtaking time t i Then, the distance between vehicle F2 and the main vehicle;

[0085] 19) Through get in, For overtaking time t i Then, the distance between vehicle F3 and the main vehicle;

[0086] 110) If Then proceed to step 111);

[0087] 111) If Then proceed to step 112); otherwise proceed to step 113).

[0088] 112) Overtaking;

[0089] 113) Do not overtake.

[0090] Furthermore, the method for determining the lane-changing behavior is as follows:

[0091] When the target lane is not the main lane, a lane-changing intention is generated; however, the driving plan for lane changing is determined by Algorithm 2. First, it is determined whether the lane can be changed by overtaking; if it cannot be changed by overtaking, it is determined whether there is enough space for lane changing. If the space for lane changing can meet the lane-changing needs of the main vehicle, then the lane-changing intention is entered; if the space for lane changing cannot meet the needs, then the lane-changing intention is selected when there is enough space in the main vehicle lane. The lane-changing method is determined by the condition triggering method, which is also the determination of the lane-changing timing.

[0092] The judgment method of Algorithm 2 is as follows:

[0093] 21) Determine whether to overtake;

[0094] 22) Predict the vehicles ahead in the adjacent lane using the following method:

[0095] 23) If the result of step 21) is to choose to overtake, then overtake; otherwise, proceed to step 24).

[0096] 24) If the conditions are met If the current behavior is not specified, then proceed to the waiting lane-changing action; otherwise, proceed to step 25.

[0097] 25) Intending to change lanes.

[0098] Furthermore, during lane changing, the potential conflict zone is determined by the position of the lead vehicle when the lead vehicle interacts with the following vehicle; the speed planning of the lead vehicle during lane changing is determined by the traffic participant ahead of the target lane, i.e., traffic vehicle F1; the time it takes for the following vehicle B1 to reach the potential conflict zone is calculated using the following formula.

[0099] ΔT=(S ego -S B1 ) / (V B1 -V ego )

[0100] In the formula, ΔT is the time it takes for the following vehicle to reach the potential conflict zone; S ego The projection of the main vehicle's position onto the road coordinate system; S B1 V is the projection of the rear vehicle's position B1 onto the road coordinates; B1 V represents the longitudinal speed of the following vehicle B1; ego The longitudinal speed of the main vehicle;

[0101] When the speed of the following vehicle B1 is less than that of the main vehicle, no game occurs; when the time it takes for the following vehicle to reach the conflict area is greater than 3 seconds, no game also occurs; only when the speed of the following vehicle is greater than that of the main vehicle and the time it takes to reach the conflict area is less than 3 seconds, the main vehicle and the following vehicle will engage in a game, and the calculation process is shown in Algorithm 3.

[0102] The judgment method of Algorithm 3 is as follows:

[0103] 31) Calculate the time ΔT for vehicles behind the adjacent lane to reach the conflict zone;

[0104] 32) If the arrival time in the conflict zone meets the conditions ΔT>0&ΔT<3, then the two cars will play a game, and the result of the game will be output as the intention result; otherwise, proceed to step 33).

[0105] 33) The driver's intention was not to yield;

[0106] When two vehicles engage in a game, the master vehicle calculates the aggressiveness of each vehicle based on its motion state towards the conflict point; the master vehicle's set of actions is... The set of actions of the following vehicle is in, The action of yielding to the main vehicle; The action of not yielding to the main vehicle; The action of yielding to the vehicle behind; The action of the following vehicle not yielding;

[0107] Furthermore, the specific method for step three is as follows:

[0108] Two benefit metrics were evaluated, including speed benefit and safety benefit:

[0109] The speed gains of the main vehicle are as follows:

[0110] When the main vehicle does not change lanes:

[0111] When the main vehicle chooses not to change lanes, it will continue to travel in the original lane and enter the return to the original lane phase; the speed planning result of the return to the original lane phase is taken as the speed gain of choosing not to overtake, which is obtained by the following formula.

[0112]

[0113] In the formula, The gain at a given speed without overtaking; v RB v represents the speed gain of the main vehicle at time t if it transitions from the overtaking phase to the return-to-original-lane phase; traffic The legal speed limit in this traffic scenario;

[0114] When the main vehicle selects a lane-changing strategy:

[0115] When the main vehicle chooses the lane-changing strategy, the speed gain is obtained by the following formula:

[0116]

[0117] in, The gain at a given speed without overtaking; v LC The speed obtained; the legal speed v in this traffic scenario. traffic =30m / s;

[0118] The speed gains of the following vehicle are as follows:

[0119] When the following vehicle is swerving to avoid a collision:

[0120] After the main vehicle arrives at the conflict zone, the following vehicle also arrives at the conflict zone; therefore, it is necessary to calculate the speed gain based on the time it takes for the main vehicle to arrive at the conflict zone, which can be obtained by the following formula.

[0121]

[0122]

[0123] In the formula, D B1 The target lane is the closest traffic participant vehicle to the driver vehicle, located behind it. The speed of the vehicle following when yielding to another vehicle; t Delay This refers to the driver's braking reaction time; Let be the payoff function for the following vehicle when it tries to cut in front of another vehicle;

[0124] When the vehicle behind is cutting in:

[0125] When the car behind cuts in, the speed the car behind will gain is The following vehicle's speed is determined by the distance between itself and the vehicle ahead (B1) and the speeds of both vehicles. The following vehicle will gain a greater speed, therefore its speed gain is calculated using the following formula.

[0126]

[0127]

[0128] In the formula, D B1-F1 The shortest distance from the following vehicle to the next road user in the lane in front of it; Let be the payoff function for the following vehicle when it tries to cut in front of another vehicle;

[0129] The safety benefits of the main vehicle are as follows:

[0130] When the intention of the lead vehicle is to overtake, and the intention of the following vehicle is also to overtake, the benefit of safety is obtained by the following formula;

[0131]

[0132] In the formula, The payoff function of the primary vehicle when it is cutting in and the following vehicle is not cutting in; ΔT is the time it takes for the following vehicle to reach the conflict zone; TTC. Threshold The time threshold for the interaction between the following vehicle and the leading vehicle;

[0133] Furthermore, the risk and benefit of the vehicle need to be normalized; if the time it takes for the main vehicle to reach the conflict zone is longer than the time it takes for the following vehicle to reach the conflict zone, the safety benefit obtained will be lower.

[0134] When the driver's intention is to change lanes and the following vehicle's intention is to yield, the safety benefit is 1, which is the maximum value. When the driver's intention is to yield, there is no conflict with the following vehicle, so the safety benefit is also 1. The safety benefit is then derived using the following formula:

[0135]

[0136] In the formula, The benefit of the following vehicle yielding when the main vehicle does not yield; The benefits a vehicle receives when yielding to another vehicle;

[0137] The safety benefits for the following vehicle are as follows:

[0138] When the following vehicle intends not to yield, and the leading vehicle intends to overtake, the safety benefit is obtained by the following formula:

[0139]

[0140] In the formula, The payoff function for the following vehicle when the following vehicle does not yield and the leading vehicle does not yield;

[0141] When the following vehicle intends to overtake, and the lead vehicle intends to yield, the safety benefit is 1, which is the maximum value. When the following vehicle intends to yield, there is no conflict with the lead vehicle, so the safety benefit is also 1, as shown by the following formula:

[0142]

[0143] In the formula, The payoff function for the following vehicle when the following vehicle does not yield and the leading vehicle yields; The benefit function of the following vehicle when yielding to another vehicle;

[0144] The payoff matrix for the two vehicles is as follows:

[0145] When the leading vehicle yields and the following vehicle also yields, the benefit is When the main vehicle yields or does not yield, the benefit is When the main vehicle changes lanes and the following vehicle yields, the benefit is When the main vehicle changes lanes and does not yield, the benefit is

[0146] In strategy combination Below, among which, S 11 A combination of strategies when the main vehicle yields and the following vehicle yields; The action of yielding to the main vehicle; The action of yielding to the vehicle behind;

[0147] The total revenue of the main vehicle and the following vehicle is shown below;

[0148]

[0149]

[0150] In the formula, Let be the revenue function of the lead vehicle when both the lead vehicle and the following vehicle yield. Let α1 and β1 be the weighting parameters of the main vehicle between speed and safety factor when the main vehicle yields and the following vehicle yields; α2 and β2 are the weighting parameters of the main vehicle between speed and safety factor; the sum of the speed weight and the safety weight is 1. The speed gain of the primary vehicle when yielding to the primary vehicle; The safety benefits of the primary vehicle when yielding to the primary vehicle; The speed gain of the following vehicle when yielding to it; The safety benefits to the following vehicle when yielding to it;

[0151] In strategy combination Below, among which, S 12 A combination of strategies when the main vehicle yields but the following vehicle does not yield; The action of yielding to the main vehicle; The action of the following vehicle not yielding;

[0152] The total revenue of the main vehicle and the following vehicle is obtained by the following formula:

[0153]

[0154]

[0155] In the formula, Let be the revenue function of the lead vehicle when the lead vehicle yields and the following vehicle does not yield; Let be the payoff function for the following vehicle when the leading vehicle yields but the following vehicle does not yield; Speed ​​gain when yielding to the main vehicle; Safety benefits when yielding to the main vehicle; Let be the speed gain function when the following vehicle does not yield; The safety benefit function for the following vehicle when the following vehicle does not yield and the leading vehicle yields;

[0156] In strategy combination Below, among which, S 21 Strategy combinations when the lead vehicle does not yield and the following vehicle yields; The action of not yielding to the main vehicle; The action of yielding to the vehicle behind;

[0157] The total revenue of the main vehicle and the following vehicle is obtained by the following formula:

[0158]

[0159]

[0160] In the formula, The benefit function of the main vehicle when the main vehicle does not yield and the following vehicle yields; The benefit function of the following vehicle when the main vehicle does not yield and the following vehicle yields; The speed benefit function when the main vehicle does not yield; When the main vehicle does not yield, the following vehicle benefits from the safety of the main vehicle that yields. The speed gain function of the following vehicle when yielding to another vehicle; The safety benefit function for the following vehicle when yielding to it;

[0161] In strategy combination Below, among which, S 22 A combination of strategies when the lead vehicle does not yield and the following vehicle does not yield. The action of not yielding to the main vehicle; The action of the following vehicle not yielding;

[0162] The total revenue of the main vehicle and the following vehicle is obtained by the following formula:

[0163]

[0164]

[0165] In the formula, The benefit function of the lead vehicle when the lead vehicle does not yield and the following vehicle does not yield; The payoff function for the following vehicle when the main vehicle does not yield and the following vehicle does not yield; The speed gain function of the main vehicle when the main vehicle does not yield; The safety benefits of the lead vehicle when both the lead vehicle and the following vehicle fail to yield; Let be the speed gain function of the following vehicle when the following vehicle does not yield; The safety benefit function for the following vehicle when both the main vehicle and the following vehicle fail to yield.

[0166] The beneficial effects of this invention are as follows:

[0167] This invention divides the lane-changing process into lane-changing execution phase 1, lane-changing execution phase 2, and lane-changing preparation phase. It also incorporates overtaking and waiting lane-changing driving options into the lane-changing decision-making system, and proposes conditions for selecting overtaking and waiting lane-changing based on the current traffic scenario, providing different driving options for autonomous driving lane-changing. The timing of lane-changing is determined through event triggering. Finally, by identifying potential conflict areas, a dynamic game model is established with the following vehicle to ensure the safe and efficient interaction between the two vehicles. Attached Figure Description

[0168] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0169] Figure 1 This is a schematic diagram of the framework of the present invention;

[0170] Figure 2 A diagram illustrating lane selection decisions;

[0171] Figure 3 A diagram illustrating the preparation phase for a lane change;

[0172] Figure 4a This is a schematic diagram of the lane-changing execution phase;

[0173] Figure 4b This is a schematic diagram of the critical position in the first stage of the lane-changing execution.

[0174] Figure 5 This is a diagram illustrating the process of returning to the original lane.

[0175] Figure 6 This is a diagram illustrating the overtaking process.

[0176] Figure 7a A diagram illustrating the situation of traffic participants in the adjacent lanes when changing lanes.

[0177] Figure 7b This is a diagram illustrating the waiting process for lane changing.

[0178] Figure 8 This is a diagram illustrating the overtaking process.

[0179] Figure 9 This is a schematic diagram of the game process during the lane-changing phase. Detailed Implementation

[0180] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0181] See Figure 1 A human-like lane-changing decision-making method includes the following steps:

[0182] See Figure 2 Step 1: Select the optimal lane;

[0183] The specific method is as follows:

[0184] Lane selection is based on the distance between the main vehicle and other road users in each lane ahead, as well as the speed of the vehicles ahead.

[0185] The lane assessment is obtained by the following formula:

[0186]

[0187]

[0188]

[0189]

[0190] In the formula, The speed evaluation function for the candidate lanes; V is the speed of the nearest road user in the waiting lane in front of the main vehicle. traffic The speed limit is set by traffic regulations; Candidate lane ID is the candidate lane ID. For the spatial evaluation function of candidate lanes; D represents the distance between the nearest traffic participant in the candidate lane, located in front of the driver, and the driver vehicle. sensor The distance within the sensing range; i is the driving load function for the candidate lanes; lane The candidate lane number is used for evaluation; The lane number where the main vehicle is located; ω1, ω2, and ω3 are the sum of the cost functions of the candidate lanes; ω1, ω2, and ω3 are the weight coefficients of each cost function, respectively.

[0191] The candidate lane with the largest sum of substitution functions is the target lane Opt ID, and it is used as input to the decision of driving behavior.

[0192]

[0193] Step 2: If a lane-changing intention is generated, select one of the three behaviors: waiting to change lanes, overtaking, or changing lanes, and generate the path and speed of the main vehicle based on the vehicle's current state.

[0194] The specific method is as follows:

[0195] When the intention to change lanes is made, if there is insufficient space for the lane change... Among them, S ego D is the projection of the vehicle's position onto the road coordinate system. LC The amount of space required for lane changing. If the projection of traffic participants in the adjacent lane onto the road coordinate system within the predicted time is insufficient to change lanes, then the player chooses to wait for a lane change; if there is sufficient space to change lanes... If the vehicle is behind other interacting traffic participants, then choose to change lanes; if the vehicle is behind them, then choose to overtake.

[0196] When a lane-changing intention is generated, if there is insufficient space to change lanes, the vehicle will enter a waiting lane-changing phase, as shown in Figure 7(a). When the waiting lane-changing behavior is selected, the path of the main vehicle is the center line of the main lane, and the recommended speed is calculated by formula (10), as shown in Figure 7(b).

[0197]

[0198] V wait =max[V wait V traffic (10)

[0199] In the formula, D F1 The distance from the main vehicle to the vehicle in front; V wait The speed planned for the waiting lane-changing phase; t Delay The driver's braking reaction time; a pd The acceleration of the main vehicle during braking; V F1 The speed of the vehicle being overtaken; a pdmax D represents the maximum braking acceleration of the vehicle being overtaken. stop V is the distance between the two vehicles after braking ends. traffic Speeds regulated by law.

[0200] When the lane change action is selected, the lane change preparation and lane change execution actions are performed.

[0201] See Figure 3 During the lane-changing preparation phase, the main vehicle's path is along the centerline of the main lane; the space for preparing to change lanes is D. pre =V ego tpre In the formula, V ego The speed of the main vehicle; t pre The preparation time for lane changing is set at 3 seconds in this invention.

[0202] When executing the lane change action, the required distance for lane change is first calculated. Once the lane change distance of the main vehicle meets the required distance, the lane change action is executed.

[0203] The distance required for lane changing is calculated using formula (3);

[0204]

[0205] In the formula, D LC V represents the space required for lane changing. ego The speed of the main vehicle; T LC The lane-changing time for the main vehicle is set to 4 seconds; D LC,limit The minimum space required for lane changing;

[0206] The Frenet coordinate system is transformed to the Cartesian coordinate system using formula (4), and the global coordinate system at the start and end of the lane change is obtained.

[0207]

[0208]

[0209] In the formula, This is the starting position for lane changing; S ego The projection of the main vehicle's position onto the road coordinate system; This is the end position of the lane change;

[0210] Determine the starting position of the lane change. and the end position Then, the lane-changing path is generated using a cubic polynomial, and obtained through formula (5):

[0211] y = a3x 3 +a2x 2 +a1x+a0

[0212]

[0213]

[0214]

[0215] In the formula, x0 is the x-coordinate of the starting position of the lane change; y0 is the y-coordinate of the starting position of the lane change. x is the heading angle at the starting position of the lane change; dy is the x-coordinate of the lane change endpoint; d The ordinate of the lane change endpoint; The heading angle at the end of the lane change;

[0216] The lane change execution phase includes lane change execution phase 1 and lane change execution phase 2;

[0217] Referring to Figure 4(a), in lane-changing execution phase 1, before the lead vehicle enters the lane 1 it wants to switch to, the lead vehicle determines whether to compete for right-of-way or yield through a game with the following vehicle B1 in lane 1. When the lead vehicle is in lane-changing execution phase 1 and the game result is yielding, the lead vehicle will enter the phase of returning to its original lane, thus avoiding a collision between the two vehicles. Figure 5 As shown.

[0218] The main vehicle needs to enter the adjacent lane. A >W lane / 2, as shown in Figure 4(b), D A W is the distance between the main vehicle corner point A (corner point A is the front corner point of the vehicle outline; for left lane changes, it is the left front corner point, and for right lane changes, it is the right front corner point) and the lane center reference line; lane The width of the lane currently occupied by the main vehicle;

[0219] The distance required to return to the original lane is D. MB =V ego ·T MB In the formula, S ego The position of the main vehicle projected onto the road coordinate system; T MB Time to return to the original lane;

[0220] Generate a path from the vehicle's current position back to the original lane using a cubic polynomial, returning to the starting position of the original lane. and the end position of returning to the original lane We can obtain the following using formula (6):

[0221]

[0222]

[0223] In the formula, x ego The x-coordinate of the current position of the main vehicle; y ego The vertical coordinate of the current position of the main vehicle; The heading angle of the main vehicle's current position;

[0224] See Figure 5 When returning to the original lane, the speed is determined based on the position and movement state of traffic participants in the original lane; obtained through formula (7):

[0225]

[0226] V RB =max[V RB V traffic (7)

[0227] In the formula, D F1 Distance to the vehicle in front in the main lane; V RB Speed ​​planning for returning to the original lane;

[0228] In lane-changing execution phase 2, the lead vehicle has entered the potential conflict zone. In this phase, it no longer engages in game theory with the following vehicle, as shown in Figure 4(a). The lead vehicle's speed is calculated using the following formula:

[0229]

[0230] V wait =max[V wait V traffic (10)

[0231] In the formula, D F1 The distance from the main vehicle to the vehicle in front; V wait The speed planned for the waiting lane-changing phase; t Delay The driver's braking reaction time; a pd The acceleration of the main vehicle during braking; V F1 The speed of the vehicle being overtaken; a pdmax D represents the maximum braking acceleration of the vehicle being overtaken. stop V is the distance between the two vehicles after braking ends. traffic Speeds regulated by law.

[0232] Furthermore, when choosing to overtake, the center line of the driver's lane is planned as a reference path; a target position S is first set. target ,like Figure 6 As shown; Reaching target position S target Then, the process shifts to lane change preparation and lane change execution.

[0233] The target location S target We can obtain the following using formula (8):

[0234] D Bsafe =L car +D F1safe

[0235]

[0236] S target =S F1 +D Bsafe (8)

[0237] In the formula, D Bsafe The safe distance between the main vehicle's current speed and the vehicle being overtaken; L car D is the vehicle profile length of the vehicle being overtaken; F1safe The safe distance between the main vehicle's current speed and the vehicle being overtaken; V Opt The speed chosen by the main vehicle for overtaking; S F1 The projection of the overtaken vehicle onto the road coordinate system;

[0238] The speed planning during the overtaking phase is obtained using formula (9):

[0239]

[0240] V overtaking =max[V overtaking V traffic (9)

[0241] In the formula, D F2 The distance between the main vehicle and the nearest vehicle in front in the main lane; V overtaking The speed planned for overtaking; V F2 The speed of the vehicle closest to the main vehicle in front of the main lane.

[0242] Furthermore, the method for determining the overtaking behavior is as follows:

[0243] If the lead vehicle intends to change lanes to the left and the traffic situation ahead allows for overtaking, it will attempt to overtake and will not choose to change lanes to follow behind vehicle F1. When the lead vehicle is behind an interacting traffic participant, it will attempt to overtake.

[0244] When there are vehicles F2 and F3 in the main lane and the overtaking lane, the speed of vehicles F2 and F3 and the space of the target position will limit the possibility of reaching the target overtaking position. This can be determined by Algorithm 1, so overtaking is not selected.

[0245] Overtaking acceleration is constrained in [0, a] max Within the range, a max The acceleration limit of the main vehicle is determined; and the acceleration within the constraint range is discretized by n (n takes the value of 10) to solve whether there is a feasible solution within the range of overtaking acceleration. If there is, overtaking is selected; if not, overtaking is not selected.

[0246] Formula (12) can be used to calculate the time required for the main vehicle to reach the target position under different accelerations for overtaking.

[0247] V opt =V ego+a opt t

[0248] S target -S ego =(V ego -V F1 )t+(a opt -a F1 )t 2 / 2 (12)

[0249] In the formula, V opt The speed at which the main vehicle reaches the target location; V ego The speed of the main vehicle; a opt The overtaking acceleration selected by the main vehicle; t is the overtaking time; a F1 S represents the longitudinal acceleration of the overtaken vehicle F1; target The projection of the target position onto the road coordinate system can be calculated using formula (8);

[0250] Due to the constraints imposed by vehicles F2 and F3 ahead, although the main vehicle can reach the target position and overtake the overtaken vehicle F1 in time t, it still needs to maintain a safe distance from the vehicles ahead. If the constraint imposed by the vehicles ahead is not met, it indicates that the selected acceleration a... opt It is not feasible; only when the feasible acceleration space is traversed and a feasible acceleration exists can the overtaking behavior be determined, and the feasibility of the overtaking behavior is judged by Algorithm 1.

[0251] The judgment method of Algorithm 1 is as follows:

[0252] 11) Calculate the target position S target ;

[0253] 12) If the projection S of the position of traffic vehicle F3 onto the road coordinate system F3 >S target Furthermore, the projection S of the position of vehicle F2 onto the road coordinate system F2 >S target Then proceed to step 13);

[0254] 13) By selecting the acceleration a for overtaking... i Calculate the choice a i Overtaking time t under overtaking acceleration i , where a i ∈[0,a max (i = 1, 2, ..., n)

[0255] 14) Calculate in, This represents the projection of the current vehicle's position onto the road coordinate system. t represents the longitudinal speed of the main vehicle at the current moment.i To choose a i Overtaking time under overtaking acceleration; This represents the longitudinal acceleration of the main vehicle at the current moment;

[0256] 15) Through get

[0257] 16) Through get in, For vehicle F2, the overtaking time t i Then, the projection of the location onto the road coordinate system; For overtaking time t i Then, the projection of the position of the main vehicle onto the road coordinate system;

[0258] 17) Through get in, For traffic vehicle F3 during overtaking time t i Then, the projection of the location onto the road coordinate system;

[0259] 18) Through get in, For overtaking time t i Then, the distance between vehicle F2 and the main vehicle;

[0260] 19) Through get in, For overtaking time t i Then, the distance between vehicle F3 and the main vehicle;

[0261] 110) If Then proceed to step 111);

[0262] 111) If Then proceed to step 112); otherwise proceed to step 113).

[0263] 112) Overtaking;

[0264] 113) Do not overtake.

[0265] Furthermore, the method for determining the lane-changing behavior is as follows:

[0266] When the target lane is not the main lane, a lane-changing intention is generated; however, the driving strategy for lane changing is determined by Algorithm 2. First, it is determined whether overtaking is possible to change lanes. If overtaking is not possible, it is assessed whether there is sufficient space for lane changing. If the space is sufficient for the main vehicle to change lanes, the lane-changing intention is initiated; otherwise, the vehicle waits in the main lane until there is enough space. Then, choose the intention to change lanes; determining the lane-changing method through conditional triggering is also a way to determine the timing of the lane change.

[0267] The judgment method of Algorithm 2 is as follows:

[0268] 21) Determine whether to overtake;

[0269] 22) Predict the vehicles ahead in the adjacent lane using the following method:

[0270] 23) If the result of step 21) is to choose to overtake, then overtake; otherwise, proceed to step 24).

[0271] 24) If the conditions are met If the current behavior is not specified, then proceed to the waiting lane-changing action; otherwise, proceed to step 25.

[0272] 25) Intending to change lanes.

[0273] Furthermore, during lane changing, when the main vehicle and the following vehicle interact, the potential conflict zone is determined by the position of the main vehicle; the speed planning of the main vehicle during lane changing is determined by the traffic participant in front of the target lane, i.e., traffic vehicle F1, and is obtained by solving formula (10); the time for the following vehicle B1 to reach the potential conflict zone is determined by the following formula.

[0274] ΔT=(S ego -S B1 ) / (V B1 -V ego (13)

[0275] In the formula, ΔT is the time it takes for the following vehicle to reach the potential conflict zone; S ego The projection of the main vehicle's position onto the road coordinate system; S B1 V is the projection of the rear vehicle's position B1 onto the road coordinates; B1 V represents the longitudinal speed of the following vehicle B1; ego The longitudinal speed of the main vehicle;

[0276] When the speed of the following vehicle B1 is less than that of the main vehicle, no game occurs; when the time it takes for the following vehicle to reach the conflict area is greater than 3 seconds, no game also occurs; only when the speed of the following vehicle is greater than that of the main vehicle and the time it takes to reach the conflict area is less than 3 seconds, the main vehicle and the following vehicle will engage in a game, and the calculation process is shown in Algorithm 3.

[0277] The judgment method of Algorithm 3 is as follows:

[0278] 31) Calculate the time ΔT for vehicles behind the adjacent lane to reach the conflict zone;

[0279] 32) If the arrival time in the conflict zone meets the conditions ΔT>0&ΔT<3, then the two cars will play a game, and the result of the game will be output as the intention result; otherwise, proceed to step 33).

[0280] 33) The driver's intention was not to yield;

[0281] When two vehicles engage in a game, the master vehicle calculates the aggressiveness of each vehicle based on its motion state towards the conflict point; the master vehicle's set of actions is... The set of actions of the following vehicle is in, The action of yielding to the main vehicle; The action of not yielding to the main vehicle; The action of yielding to the vehicle behind; The action of the following vehicle not yielding;

[0282] Step 3: Evaluate the benefit indicators, which include speed benefits and security benefits.

[0283] Furthermore, the specific method for step three is as follows:

[0284] Two benefit metrics were evaluated, including speed benefit and safety benefit:

[0285] The speed gains of the main vehicle are as follows:

[0286] When the main vehicle does not change lanes:

[0287] When the main vehicle chooses not to change lanes, it will continue to drive in the original lane and will enter the return to the original lane stage; the speed planning result of the return to the original lane stage is used as the speed benefit of choosing not to overtake, which is obtained by formula (14).

[0288]

[0289] In the formula, The gain at a given speed without overtaking; v RBTo calculate the speed gain of the main vehicle at time t if it transitions from the overtaking phase to the return-to-original-lane phase, use formula (7); traffic The legal speed limit in this traffic scenario;

[0290] When the main vehicle selects a lane-changing strategy:

[0291] When the main vehicle chooses the lane-changing strategy, the speed gain is obtained through formula (15):

[0292]

[0293] in, The gain at a given speed without overtaking; v LC The speed obtained; the legal speed v in this traffic scenario. traffic =30m / s;

[0294] The speed gains of the following vehicle are as follows:

[0295] When the following vehicle is swerving to avoid a collision:

[0296] After the main vehicle arrives at the conflict zone, the following vehicle arrives at the conflict zone; therefore, it is necessary to calculate the speed gain based on the time it takes for the main vehicle to arrive at the conflict zone, which can be obtained through formula (16).

[0297]

[0298]

[0299] In the formula, D B1 The target lane is the closest traffic participant vehicle to the driver vehicle, located behind it. The speed of the vehicle following when yielding to another vehicle; t Delay This refers to the driver's braking reaction time; Let be the payoff function for the following vehicle when it tries to cut in front of another vehicle;

[0300] When the vehicle behind is cutting in:

[0301] When the car behind cuts in, the speed the car behind will gain is The following vehicle is determined by the distance between itself and the vehicle ahead (B1) and the speeds of both vehicles. The following vehicle will gain a greater speed, and therefore the speed gain of the following vehicle is calculated using formula (17).

[0302]

[0303]

[0304] In the formula, D B1-F1 The shortest distance from the following vehicle to the next road user in the lane in front of it; Let be the payoff function for the following vehicle when it tries to cut in front of another vehicle;

[0305] The safety benefits of the main vehicle are as follows:

[0306] When the intention of the main vehicle is to overtake, and the intention of the following vehicle is to overtake, the benefit of safety is obtained by formula (18);

[0307]

[0308] In the formula, The payoff function of the primary vehicle when it is cutting in and the following vehicle is not cutting in; ΔT is the time it takes for the following vehicle to reach the conflict zone; TTC. Threshold The time threshold for the interaction between the following vehicle and the leading vehicle;

[0309] Furthermore, the risk and benefit of the vehicle need to be normalized; if the time it takes for the main vehicle to reach the conflict zone is longer than the time it takes for the following vehicle to reach the conflict zone, the safety benefit obtained will be lower.

[0310] When the driver's intention is to change lanes and the following vehicle's intention is to yield, the safety benefit is 1, which is the maximum value. When the driver's intention is to yield, there is no conflict with the following vehicle, so the safety benefit is also 1. The safety benefit is obtained through formula (19):

[0311]

[0312] In the formula, The benefit of the following vehicle yielding when the main vehicle does not yield; The benefits the primary vehicle receives when yielding to the primary vehicle;

[0313] The safety benefits for the following vehicle are as follows:

[0314] When the following vehicle intends not to yield, and the lead vehicle intends to overtake, the safety benefit can be obtained using formula (20):

[0315]

[0316] In the formula, The payoff function for the following vehicle when the following vehicle does not yield and the leading vehicle does not yield;

[0317] When the following vehicle intends to overtake, and the main vehicle intends to yield, the safety benefit is 1, which is the maximum value; when the following vehicle intends to yield, there is no conflict with the main vehicle, so the safety benefit is also 1, which can be obtained through formula (21):

[0318]

[0319] In the formula, The payoff function for the following vehicle when the following vehicle does not yield and the leading vehicle yields; The benefit function of the following vehicle when yielding to another vehicle;

[0320] The payoff matrix for the two vehicles is as follows:

[0321]

[0322] In strategy combination Below, among which, S 11 A combination of strategies when the main vehicle yields and the following vehicle yields; The action of yielding to the main vehicle; The action of yielding to the vehicle behind;

[0323] The total revenue of the main vehicle and the following vehicle is obtained by formula (22);

[0324]

[0325]

[0326] In the formula, Let be the revenue function of the lead vehicle when both the lead vehicle and the following vehicle yield. Let α1 and β1 be the weighting parameters of the main vehicle between speed and safety factor when the main vehicle yields and the following vehicle yields; α2 and β2 are the weighting parameters of the following vehicle between speed and safety factor; the sum of the two weighting coefficients (weighting coefficients of safety and speed benefits) is 1. The speed gain of the primary vehicle when yielding to the primary vehicle; The safety benefits of the primary vehicle when yielding to the primary vehicle; The speed gain of the following vehicle when yielding to it; The safety benefits to the vehicle behind when yielding to it;

[0327] In strategy combination Below, among which, S 12 A combination of strategies when the main vehicle yields but the following vehicle does not yield; The action of yielding to the main vehicle; The action of the following vehicle not yielding;

[0328] The total revenue of the main vehicle and the following vehicle is obtained by formula (23):

[0329]

[0330]

[0331] In the formula, Let be the revenue function of the lead vehicle when the lead vehicle yields and the following vehicle does not yield; Let be the payoff function for the following vehicle when the leading vehicle yields but the following vehicle does not yield; Speed ​​gain when yielding to the main vehicle; Safety benefits when yielding to the main vehicle; Let be the speed gain function when the following vehicle does not yield; The safety benefit function for the following vehicle when the following vehicle does not yield and the leading vehicle yields;

[0332] In strategy combination Below, among which, S 21 Strategy combinations when the lead vehicle does not yield and the following vehicle yields; The action of not yielding to the main vehicle; The action of yielding to the vehicle behind;

[0333] The total revenue of the main vehicle and the following vehicle is obtained by formula (24).

[0334]

[0335]

[0336] In the formula, The benefit function of the main vehicle when the main vehicle does not yield and the following vehicle yields; The benefit function of the following vehicle when the main vehicle does not yield and the following vehicle yields; The speed benefit function when the main vehicle does not yield; When the main vehicle does not yield, the following vehicle benefits from the safety of the main vehicle that yields. The speed gain function of the following vehicle when yielding to another vehicle; The safety benefit function of the following vehicle when yielding to another vehicle;

[0337] In strategy combination Below, among which, S 22 A combination of strategies when the lead vehicle does not yield and the following vehicle does not yield. The action of not yielding to the main vehicle; The action of the following vehicle not yielding;

[0338] The total revenue of the main vehicle and the following vehicle is obtained by formula (25):

[0339]

[0340]

[0341] In the formula, The benefit function of the lead vehicle when the lead vehicle does not yield and the following vehicle does not yield; The payoff function of the following vehicle when the main vehicle does not yield and the following vehicle does not yield; The speed gain function of the main vehicle when the main vehicle does not yield; The safety benefits of the lead vehicle when both the lead vehicle and the following vehicle fail to yield; Let be the speed gain function of the following vehicle when the following vehicle does not yield; The safety benefit function for the following vehicle when both the main vehicle and the following vehicle fail to yield.

[0342] This invention models the interaction between two vehicles using a dynamic game theory approach, with the other vehicle acting as the leader and the current vehicle as the follower. After establishing the dynamic game, the game equilibrium is solved using formula (26).

[0343]

[0344]

[0345] The subscripts L and F represent leader and follower, respectively. L Strategies representing leaders, s F The strategy of representing followers. This represents the optimal strategy for followers. The optimal strategy representing the leader. S F The strategy space representing the followers. S L The strategic space representing the leader. P represents the optimal set of strategies for followers given the leader's strategy. F Let P be the cost function for the followers. L The cost function for the leader.

[0346] Because drivers interact with different types of road users during driving, it is necessary to calculate the driver's aggressiveness. Different types of drivers have different weighting coefficients for safety indicators. This invention estimates the driving style coefficient of the following vehicle based on the acceleration of the interacting vehicles and calculates it using formula (27). max This is the maximum longitudinal acceleration.

[0347] ρ=a B1 / a max (27)

[0348] Where ρ represents the aggressiveness of the following vehicle, and a B1 The magnitude of the following vehicle's acceleration is ρ; the degree of aggression ρ will adjust the weights β1 and β2 of the safety factor. β′1 is the weight coefficient of the main vehicle after considering the degree of aggression. β′2 is the weight coefficient of the following vehicle after considering the degree of aggression, calculated by formula (28). ω1 is the magnitude of the influence of the degree of aggression on the weight coefficient of the main vehicle. ω2 is the magnitude of the influence of the degree of aggression on the weight coefficient of the following vehicle.

[0349] β′1=β1-ω1ρ

[0350] β′2=β2+ω2ρ (28)

[0351] In summary, this invention divides the lane-changing process into lane-changing execution stage 1, lane-changing execution stage 2, and lane-changing preparation stage. It also incorporates overtaking and waiting lane-changing driving schemes into the lane-changing decision-making system, and proposes conditions for selecting overtaking and waiting lane-changing based on the current traffic scenario, providing different driving schemes for autonomous driving lane-changing. The timing of lane-changing is determined through event triggering. Finally, by identifying potential conflict areas, a dynamic game model is established with the following vehicle to ensure the safe and efficient interaction between the two vehicles.

[0352] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A human-like lane-changing decision-making method, characterized in that, Includes the following steps: Step 1: Select the optimal lane; Step 2: If a lane-changing intention is generated, select one of the three behaviors: waiting to change lanes, overtaking, or changing lanes, and generate the path and speed of the main vehicle based on the vehicle's current state. Step 3: Evaluate the benefit indicators, which include speed benefits and security benefits; The specific method for step two is as follows: When an intention to change lanes is made, if there is insufficient space to change lanes, then wait to change lanes; if there is sufficient space to change lanes, then change lanes; if the lead vehicle is behind other interacting traffic participants, then overtake. When the lane change action is selected, the lane change preparation and lane change execution actions are performed. During the lane-changing preparation phase, the vehicle's path is along the center line of its current lane; the space required for the lane change is... In the formula, The speed of the main vehicle; Preparation time for changing lanes; When executing the lane change action, first calculate the distance required for the lane change, and then execute the lane change action once the lane change distance of the vehicle meets the required distance. The distance required for a lane change is calculated using the following formula; In the formula, The amount of space required for lane changing; This refers to the speed of the vehicle. The lane-changing time for this vehicle is set to [specify]. ; The minimum space required for lane changing; The global coordinate system for the start and end of a lane change is obtained using the following formula: In the formula, This is the starting position for lane changing; This is the projection of the vehicle's position onto the road coordinate system; This is the end position of the lane change; Determine the starting position of the lane change. and the end position Then, the lane-changing path is generated using a cubic polynomial, obtained through the following formula: In the formula, The x-coordinate of the starting position of the lane change; The vertical coordinate represents the starting position of the lane change; The heading angle at the starting position of the lane change; The x-coordinate of the lane change endpoint; The ordinate of the lane change endpoint; The heading angle at the end of the lane change; The lane change execution phase includes lane change execution phase 1 and lane change execution phase 2; During lane-changing execution phase 1, before the primary vehicle enters the desired lane 1, it engages in a game of strategy with the following vehicle B1 in lane 1 to determine whether to contest or yield the right-of-way. If the primary vehicle is in lane-changing execution phase 1 and the outcome of the game is yielding, it will return to its original lane, thus avoiding a collision. The primary vehicle's entry into the adjacent lane requires... In the formula, The distance between the main vehicle's corner point A and the lane center reference line; The width of the lane currently occupied by the main vehicle; The distance required to return to the original lane is In the formula, This is the projected position of the vehicle on the road coordinate system; Time to return to the original lane; Return to the starting position of the original lane and the end position of returning to the original lane as follows: In the formula, The x-coordinate of the current position of the main vehicle; The vertical coordinate of the current position of the main vehicle; The heading angle of the main vehicle's current position; When returning to the original lane, the speed is determined based on the position and movement of traffic participants in the original lane; specifically as follows: In the formula, This is the distance to the vehicle in front in this lane; Speed ​​planning for returning to the original lane; In the second phase of lane change execution, the lead vehicle has entered the potential conflict zone and no longer engages in a game with the following vehicle. The vehicle speed is calculated using the following formula: In the formula, This is the distance between this vehicle and the vehicle in front; The speed planned for the waiting phase of lane changing; This refers to the driver's braking reaction time; This is the acceleration of the vehicle during braking; The speed of the vehicle being overtaken; The maximum braking acceleration of the vehicle being overtaken; The distance between the two vehicles after braking ends; Speeds regulated by law.

2. The anthropomorphic lane-changing decision-making method according to claim 1, characterized in that, The specific method for step one is as follows: 11) Select lanes based on the distance between the vehicle and other road users in each lane ahead and the speed of the vehicles ahead; The lane assessment is obtained by the following formula: In the formula, The speed evaluation function for candidate lanes; The speed of the nearest road user in front of this vehicle in the waiting lane; Speed ​​limits are set by traffic regulations; Candidate lane ID; For the spatial evaluation function of candidate lanes; The distance between this vehicle and the nearest traffic participant in the candidate lane located in front of it; The distance within the sensing range; The driving burden function for the candidate lanes; The candidate lane number is used for evaluation; This is the lane number where the vehicle is located; This is the sum of the cost functions of the candidate lanes; , , These are the weight coefficients for each cost function; 12) The candidate lane with the largest sum of substitution valence functions is the target lane. And it is used as input in the decision-making process for driving behavior.

3. The anthropomorphic lane-changing decision-making method according to claim 1, characterized in that, When choosing to wait for a lane change, the vehicle's path is the center line of the current lane, and the vehicle's speed is calculated using the following formula: In the formula, This is the distance between this vehicle and the vehicle in front; The speed planned for the waiting phase of lane changing; This refers to the driver's braking reaction time; This is the acceleration of the vehicle during braking; The speed of the vehicle being overtaken; The maximum braking acceleration of the vehicle being overtaken; The distance between the two vehicles after braking ends; Speeds regulated by law.

4. The anthropomorphic lane-changing decision-making method according to claim 1, characterized in that, When choosing to overtake, the center line of the driver's lane is used as a reference path; a target position is first set. ; Reach the target location Then, the process moves on to lane change preparation and lane change execution. The target location The following formula can be used to obtain: In the formula, The safe distance between the main vehicle's current speed and the vehicle being overtaken; The vehicle outline length of the overtaken vehicle; The safe distance between the main vehicle's current speed and the vehicle being overtaken; The speed selected for this vehicle to overtake; The projection of the overtaken vehicle onto the road coordinate system; Speed ​​planning during the overtaking phase is obtained using the following formula: In the formula, This is the distance between this vehicle and the nearest vehicle in front in this lane; The speed planned for overtaking; This is the speed of the nearest vehicle ahead of this vehicle in this lane.

5. The anthropomorphic lane-changing decision-making method according to claim 4, characterized in that, The method for determining the overtaking behavior is as follows: If the lead vehicle intends to change lanes to the left and the traffic situation ahead allows for overtaking, it will attempt to overtake and will not choose to change lanes to follow behind vehicle F1. When the lead vehicle is behind an interacting traffic participant, it will attempt to overtake. If there are vehicles F2 and F3 in the main lane and the overtaking lane, the speed of vehicles F2 and F3 and the space of the target position will limit the possibility of reaching the target overtaking position, so overtaking is not chosen. Overtaking acceleration constraint Within the range, and for the acceleration within the constraint range... The system is divided into discrete steps to determine if there is a feasible solution within the range of overtaking acceleration. If a solution exists, overtaking is performed; otherwise, overtaking is not performed. The following formula can be used to calculate the time required for the main vehicle to reach the target position under different accelerations for overtaking; In the formula, The speed at which the main vehicle reaches the target location; The speed of the main vehicle; The overtaking acceleration selected for the main vehicle; This refers to the time for overtaking; The longitudinal acceleration of the overtaken vehicle F1; Due to the constraints imposed by vehicles F2 and F3 ahead, the main vehicle was able to [achieve its intended purpose] within time. Upon reaching the target position, overtake the vehicle that was overtaken (F1), but still need to maintain a safe distance from the vehicle in front; if the distance from the vehicle in front is not met, it indicates that the selected acceleration... It is not feasible; only when the feasible acceleration space is traversed and feasible acceleration exists can the overtaking behavior be determined, and the feasibility of the overtaking behavior is judged by Algorithm 1. The judgment method of Algorithm 1 is as follows: 11) Calculate the target location ; 12) If the projection of the position of vehicle F3 onto the road coordinate system Then proceed to step 13). 13) By selecting the acceleration for overtaking. Calculate the choice Overtaking time under overtaking acceleration ,in, 14) Calculation ;in, This represents the projection of the current vehicle's position onto the road coordinate system. This represents the longitudinal speed of the main vehicle at the current moment; For selection Overtaking time under overtaking acceleration; This represents the longitudinal acceleration of the main vehicle at the current moment; 15) Through get ; 16) Through get ;in, For traffic vehicle F2 during overtaking time Then, the projection of the location onto the road coordinate system; In order to overtake Then, the projection of the position of the main vehicle onto the road coordinate system; 17) Through get ;in, For traffic vehicle F3 during overtaking time Then, the projection of the location onto the road coordinate system; 18) Through get ;in, In order to overtake Then, the distance between vehicle F2 and the main vehicle; 19) Through get ;in, In order to overtake Then, the distance between vehicle F3 and the main vehicle; 110) If Then proceed to step 111). 111) If If the condition is met, proceed to step 112); otherwise, proceed to step 113). 112) Overtaking; 113) Do not overtake.

6. The anthropomorphic lane-changing decision-making method according to claim 5, characterized in that, The method for determining the lane-changing behavior is as follows: When the target lane is not the current lane, a lane change intention is generated; however, the driving plan for lane change is determined by Algorithm 2; first, it is determined whether the lane can be changed by overtaking; if the lane cannot be changed by overtaking, it is determined whether there is enough space for lane change; if the space for lane change can meet the lane change needs of the main vehicle, then the lane change intention is entered. If there is not enough space to change lanes, then the intention to change lanes will be made when there is enough space in the main lane; determining the timing of lane changing is also a way of triggering lane changes through conditions. The judgment method of Algorithm 2 is as follows: 21) Determine whether to overtake; 22) Predict the vehicles ahead in the adjacent lane using the following method: ; 23) If the result of step 21) is to choose to overtake, then overtake; otherwise, proceed to step 24). 24) If the conditions are met If the current behavior is not specified, proceed to step 25; otherwise, proceed to step 25. 25) Intending to change lanes.

7. The anthropomorphic lane-changing decision-making method according to claim 6, characterized in that, When changing lanes, the potential conflict zone is determined by the position of the lead vehicle during the interaction between the lead vehicle and the following vehicle. The speed planning of the lead vehicle during lane changing is determined by the traffic participant ahead of the target lane, i.e., vehicle F1. The time for the following vehicle B1 to reach the potential conflict zone is calculated by the following formula. In the formula, The time it takes for the following vehicle to reach the potential conflict zone; The projection of the main vehicle's position onto the road coordinate system; This is the projection of the rear vehicle's position B1 onto the road coordinates; The longitudinal speed of the following vehicle B1; The longitudinal speed of the main vehicle; No game occurs when the speed of the following vehicle B1 is less than that of the main vehicle; if the time it takes for the following vehicle to reach the conflict zone is greater than... At that time, it is also a game that does not occur; it only occurs when the speed of the following vehicle is greater than that of the leading vehicle and the time to reach the conflict zone is less than At that time, the main vehicle and the following vehicle will engage in a game, and the calculation process is shown in Algorithm 3; The judgment method of Algorithm 3 is as follows: 31) Calculate the time it takes for vehicles behind in the adjacent lane to reach the conflict zone. ; 32) If the time of arrival at the conflict zone meets the condition. Then, a game is played between the two vehicles, and the result of the game is output as the intended outcome; otherwise, step 33 is executed. 33) The driver's intention was not to yield; When two vehicles engage in a game, the master vehicle calculates the aggressiveness of each vehicle based on its motion state towards the conflict point; the master vehicle's set of actions is... The set of actions of the following vehicle is ,in, The action of yielding to the main vehicle; This vehicle failed to yield to the other vehicle. The action of yielding to the vehicle behind; The action of the following vehicle not yielding.

8. The anthropomorphic lane-changing decision-making method according to claim 1, characterized in that, The specific method for step three is as follows: Two benefit metrics were evaluated, including speed benefit and safety benefit: The speed gains of the main vehicle are as follows: When the main vehicle does not change lanes: When the main vehicle chooses not to change lanes, it will continue to travel in the original lane and enter the return to the original lane phase; the speed planning result of the return to the original lane phase is taken as the speed gain of choosing not to overtake, which is obtained by the following formula. In the formula, The benefit of a determined speed under the condition of not overtaking; In order to be in The speed gain of the main vehicle if it switches from the overtaking phase to the return-to-original-lane phase at a given moment; The legal speed limit in this traffic scenario; When the main vehicle selects a lane-changing strategy: When the main vehicle chooses the lane-changing strategy, the speed gain is obtained by the following formula: in, The benefit of a determined speed under the condition of not overtaking; The speed obtained; the legal speed limit in this traffic scenario. ; The speed gains of the following vehicle are as follows: When the following vehicle is swerving to avoid a collision: After the main vehicle arrives at the conflict zone, the following vehicle also arrives at the conflict zone; therefore, it is necessary to calculate the speed gain based on the time it takes for the main vehicle to arrive at the conflict zone, which can be obtained by the following formula. In the formula, The vehicle in the target lane that is closest to this vehicle and is behind it. The speed of the vehicle behind when yielding to it; This refers to the driver's braking reaction time; Let be the payoff function for the following vehicle when it tries to cut in front of another vehicle; When the following vehicle forces its way through: When the car behind cuts in, the speed the car behind will gain is The following vehicle's speed is determined by the distance between itself and the vehicle ahead (B1) and the speeds of both vehicles. The following vehicle will gain a greater speed, therefore its speed gain is calculated using the following formula. In the formula, The shortest distance from the following vehicle to the next road user in the lane in front of it; Let be the payoff function for the following vehicle when it tries to cut in front of another vehicle; The safety benefits of this vehicle are as follows: When the intention of this vehicle is to overtake, and the intention of the vehicle behind is also to overtake, the benefit of safety is obtained by the following formula; In the formula, Let be the revenue function of this vehicle when it cuts in front of another vehicle and the vehicle behind it does not cut in front of another vehicle; The time it took for the following vehicle to arrive at the conflict zone; The time threshold for the interaction between the following vehicle and this vehicle; Furthermore, the risk and benefit of the vehicle need to be normalized; if the time it takes for the main vehicle to reach the conflict zone is longer than the time it takes for the following vehicle to reach the conflict zone, the safety benefit obtained will be lower. When the driver's intention is to change lanes and the following vehicle's intention is to yield, the safety benefit is 1, which is the maximum value. When the driver's intention is to yield, there is no conflict with the following vehicle, so the safety benefit is also 1. The safety benefit is then derived using the following formula: In the formula, The benefit of the following vehicle yielding when the main vehicle does not yield; The benefit the primary vehicle receives when yielding to another vehicle; The safety benefits for the following vehicle are as follows: When the following vehicle intends not to yield, and the leading vehicle intends to overtake, the safety benefit is obtained by the following formula: In the formula, The payoff function for the following vehicle when the following vehicle does not yield and the leading vehicle does not yield; When the following vehicle intends to overtake, and the lead vehicle intends to yield, the safety benefit is 1, which is the maximum value. When the following vehicle intends to yield, there is no conflict with the lead vehicle, so the safety benefit is also 1, as shown by the following formula: In the formula, The payoff function for the following vehicle when the following vehicle does not yield and the leading vehicle yields; The benefit function of the following vehicle when yielding to another vehicle; The payoff matrix for the two vehicles is as follows: When the leading vehicle yields and the following vehicle also yields, the benefit is When the main vehicle yields or does not yield, the benefit is... When the main vehicle changes lanes and the following vehicle yields, the benefit is... When the main vehicle changes lanes and does not yield, the benefit is... ; In strategy combination Below, among them, A combination of strategies when the main vehicle yields and the following vehicle yields; The action of yielding to the main vehicle; The action of yielding to the vehicle behind; The total revenue of the main vehicle and the following vehicle is shown below; In the formula, Let be the revenue function of the lead vehicle when both the lead vehicle and the following vehicle yield. The payoff function of the following vehicle when both the lead vehicle and the following vehicle yield; , The weighting parameter between speed and safety factor for the main vehicle; , Weighting parameters of the main vehicle between speed and safety factor The sum of the two weighting coefficients is 1; for income; The safety benefits of the primary vehicle when yielding to the primary vehicle; The speed gain of the following vehicle when yielding to it; The safety benefits to the vehicle behind when yielding to it; In strategy combination Below, among them, A combination of strategies when the main vehicle yields but the following vehicle does not yield; The action of yielding to the main vehicle; The action of the following vehicle not yielding; The total revenue of the main vehicle and the following vehicle is obtained by the following formula: In the formula, Let be the revenue function of the lead vehicle when the lead vehicle yields and the following vehicle does not yield; The payoff function of the following vehicle when the main vehicle yields but the following vehicle does not; Speed ​​gain when yielding to the main vehicle; Safety benefits when yielding to the main vehicle; Let be the speed gain function when the following vehicle does not yield; The safety benefit function for the following vehicle when the following vehicle does not yield and the leading vehicle yields; In strategy combination Below, among them, Strategy combinations when the lead vehicle does not yield and the following vehicle yields; The action of not yielding to the main vehicle; The action of yielding to the vehicle behind; The total revenue of the main vehicle and the following vehicle is obtained by the following formula: In the formula, The benefit function of this vehicle when the main vehicle does not yield and the following vehicle yields; The benefit function of the following vehicle when the main vehicle does not yield and the following vehicle yields; The speed benefit function when the main vehicle does not yield; When the vehicle behind fails to yield, the following vehicle benefits from the safety of the vehicle that yields. The speed gain function of the following vehicle when yielding to another vehicle; The safety benefit function for the following vehicle when yielding to it; In strategy combination Below, among them, Strategy combinations when the lead vehicle does not yield and the following vehicle does not yield. The action of not yielding to the main vehicle; The action of the following vehicle not yielding; The total revenue of the main vehicle and the following vehicle is obtained by the following formula: In the formula, The benefit function of this vehicle when it does not yield and the vehicle behind it does not yield; The payoff function for the following vehicle when the following vehicle does not yield to the first vehicle; The speed gain function of the main vehicle when the main vehicle does not yield; The safety benefits of the lead vehicle when both the lead vehicle and the following vehicle fail to yield; Let be the speed gain function of the following vehicle when the following vehicle does not yield; Let be the safety benefit function for the following vehicle when the vehicle does not yield and the following vehicle does not yield.

Images