CN111081065A - Intelligent vehicle collaborative lane change decision model under road section mixed traveling condition - Google Patents

Abstract

The invention discloses an intelligent vehicle collaborative lane change model under a road section mixed-traveling condition, which is used for establishing a vehicle lane change interactive relation judgment model based on a fuzzy logic method so as to analyze vehicle-vehicle interactive behaviors under the mixed-traveling condition; establishing a cooperative lane-changing game model of the manned vehicle and the unmanned vehicle; variable cooperation coefficients are introduced to establish a cooperative lane changing game model of the manned vehicle and the unmanned vehicle, and a Lemke-Howson algorithm is adopted to carry out Nash equilibrium solving on the game model to obtain the optimal strategy combination of whether the lane of the vehicle is changed or not. According to the invention, by establishing the cooperative lane-changing game model of manned vehicles and unmanned vehicles, the traffic flow operation efficiency can be effectively improved, the driving experience of passengers can be greatly improved, and the positive effect on traffic safety is achieved.

Description

Intelligent vehicle collaborative lane change decision model under road section mixed traveling condition

Technical Field

The invention belongs to the technical field of intelligent traffic systems and intelligent vehicle research, and relates to driving interaction behavior classification.

Background

The development of the unmanned technology has become a global consensus, and the intelligent networked automobile has great potential in solving traffic safety and controlling traffic jam in China, and the policies related to internet of vehicles and unmanned driving are taken out one after another, so that the intelligent networked automobile can be further developed and applied in the directions of networked collaborative perception, networked collaborative decision and control, conditional automatic driving, full automatic driving and the like.

Although the research on the unmanned vehicle related technology has made great progress, enabling low-speed driving in a limited area of an urban road, and autonomous driving in a simple environment of an expressway, the application of unmanned in the automotive field is also becoming problematic, so that people are still in worry and apprehension about the unmanned technology. According to statistics of road safety records of manned and unmanned vehicles by the university of michigan in traffic research, research shows that unmanned vehicles are more prone to accidents than manned vehicles, the accident ratio of the unmanned vehicles to the manned vehicles is 9.1:1.9, and the rear-end collision accident ratio of the unmanned vehicles is 50% higher than that of the manned vehicles. For example, in 2016, a traffic accident occurred when Tesla Model S had initiated autonomous driving in California, USA, which was the earliest reported traffic accident due to an error in the autonomous driving program. In 2018, in 3 months, a Uber unmanned test vehicle crashes into a pedestrian in arizona to cause death, which is the death accident caused by the pedestrian being crashed together with the automatic driving vehicle, and after the accident, the Uber suspends the automatic driving test work. Therefore, the automatic driving function is not mature, and the decision control system still has potential safety hazards.

In addition, in the united states DARPA urban challenge race, "chinese intelligent vehicle future challenge race (IVFC)", "chinese intelligent vehicle race (CIVC)", "world intelligent driving challenge race (WIDC)", when facing the interference of the manned vehicles, the racing vehicles mostly adopt the conservative driving behaviors of deceleration or waiting to avoid conflict, the influence of the dynamic interactive behaviors of other vehicles is not considered, the difference influence of the interactive behaviors of different types of drivers is ignored, and the passing potential of the unmanned vehicles is greatly limited.

At present, the intelligent degree of an automobile has realized full automatic driving under a limited condition, the automation level basically reaches the level of L3, but the unmanned level at the level of L4 and L5 has a certain distance, so that the unmanned vehicle needs to realize real road traveling in the future and faces a mixed traffic environment coexisting with a manned vehicle, and the unmanned vehicle needs to have an effective interaction mechanism, realize cooperative driving of the manned vehicle and the unmanned vehicle in the mixed traveling environment, and fully exert the advantages of accurate control, internet communication and the like.

Aiming at the problems, when a person and an unmanned automobile share the road resources, the road resources can be effectively utilized only by realizing the interactive cooperation of the person and the unmanned automobile, and traffic accidents are prevented. The research combines the research of the national key research and development plan subject 'quantitative evaluation technology research of environmental adaptability of the automatic driving electric automobile', and establishes a road section vehicle lane changing interactive relation judgment model by taking a vehicle-vehicle interactive relation under the condition that people and unmanned vehicles are mixed to run as a key link of safe interactive driving, and the road section vehicle lane changing interactive relation judgment model is expressed by vehicle cooperation competition degree to help the unmanned vehicles to understand and judge the behavior of people and vehicles around.

The unmanned vehicle is required to have an effective interaction mechanism, realize cooperative driving of the unmanned vehicle and the manned vehicle in a hybrid driving environment, and fully exert the advantages of accurate control, internet communication and the like.

Aiming at the problems, when a person and an unmanned automobile share the road resources, the road resources can be effectively utilized only by realizing the interactive cooperation of the person and the unmanned automobile, and traffic accidents are prevented. The research combines the research of the national key research and development plan topic 'quantitative evaluation technology research of environmental adaptability of the automatic driving electric automobile', establishes a vehicle cooperative lane changing model under the road section mixed condition by introducing variable cooperation coefficients according to the vehicle-vehicle interaction behavior under the mixed condition, optimizes the unmanned vehicle action generation result, and completes the efficiency simulation experiment under the road section interaction mode. The model accords with the interactive process of the passing process of human drivers, so that the unmanned vehicle shows social cooperation behaviors, and has important significance for improving the autonomous driving level of the unmanned vehicle.

The method verifies the effectiveness of the vehicle lane change interactive relation judgment model through Chinese intelligent automobile competition and real road real vehicle data, helps the unmanned vehicle to understand and judge the behavior of the surrounding vehicles, and has important significance for improving the autonomous driving level of the unmanned vehicle

Disclosure of Invention

The invention aims to establish a vehicle lane change interactive relationship judgment model based on a fuzzy logic method, which comprehensively considers the lane change intention of two interactive parties and the types of drivers of other vehicles and fuzzily infers the cooperative competition degree among the vehicles, establishes a mapping relationship between a fuzzy inference output value and an interactive relationship prediction result, so as to analyze vehicle-vehicle interactive behaviors under mixed conditions, help unmanned vehicles to understand and judge the behaviors of peripheral vehicles with people, and lay a foundation for realizing multi-vehicle cooperative driving.

The invention also aims to provide a cooperative lane change game model of the unmanned and manned vehicles by introducing variable cooperation coefficients based on a vehicle lane change interactive relation judgment model and an expected lane change decision. The model aims at maximizing the combined expected effectiveness, the cost of the strategy is evaluated by adjusting the cooperation coefficient, Nash equilibrium solving is carried out on the game model by adopting a Lemke-Howson algorithm, and the efficiency simulation experiment under different combined strategies is completed.

The invention adopts the following technical scheme:

the invention discloses an intelligent vehicle collaborative lane change model under a road section mixed-traffic condition, which is characterized in that a vehicle lane change interactive relation judgment model is established based on a fuzzy logic method, so that vehicle-vehicle interactive behaviors under the mixed-traffic condition are analyzed, lane change will is conducted on a vehicle, driving benefits increased after the vehicle changes lanes and the degree of conflict with a vehicle behind a target lane are used as model input quantity, and the degree of the lane change will is used as output variable; aiming at the driving type of the vehicle behind the target lane, the vehicle speed and the acceleration are used as model input quantities, and the driving acceleration degree is used as a model output quantity;

according to the lane change intention of a target vehicle and the driving type of a vehicle behind a target lane, the lane change intention and the type of drivers of other vehicles are used as input variables, the output is the vehicle cooperation competition degree, the vehicle cooperation degree is expressed by language variables with high, medium and low values, the interaction relations of the other vehicles are 3 results of cooperation relations, ambiguous and competition relations, fuzzy implication relations of fuzzy reasoning adopt a Mamdani rule, ambiguity resolution adopts a gravity center method, and the corresponding relations between the vehicle cooperation competition degree, the lane change intention and the driving type are obtained by establishing a cooperative lane change game model of an unmanned vehicle; variable cooperation coefficients are introduced to establish a cooperative lane changing game model of the manned vehicle and the unmanned vehicle, and a Lemke-Howson algorithm is adopted to carry out Nash equilibrium solving on the game model to obtain the optimal strategy combination of whether the lane of the vehicle is changed or not.

Furthermore, the vehicle lane changing interactive relationship model is divided into two steps:

the first step is to judge the level of the lane change intention of the target vehicle on the original lane;

the lane change will of the target vehicle on the original lane comprises the following steps:

(1) a driving benefit; the driving benefit of the original lane takes the distance L between the target vehicle and the front vehicle of the original lane into consideration_i(t) and velocity difference Δ v_i(t) the target lane driving benefit takes into account the distance L between the target vehicle and the vehicle ahead of the target lane_i(t) and velocity difference Δ v_i(t)

(2) Severity of conflict T_c

TTC (Time to Collision) is used for representing the intensity of Collision with the rear vehicle of the target lane and is marked as T_cThe length of the vehicle body is considered when calculating the TTC;

judging the driving type of the vehicle behind the target lane; the interactive relation between the vehicles is deduced by integrating the two steps, and the cooperation degree between the vehicles is represented; the driver is divided into an aggressive type, a common type and a conservative type, the speed and the acceleration of the vehicle are used as model input quantity, the driving aggressive degree is used as model output quantity, and a fuzzy logic rule is constructed;

when the lane driving benefit value is larger and the collision between the vehicle and the vehicle behind the target lane is smaller, the vehicle is more likely to select lane change to pursue better driving conditions; conversely, the smaller the lane driving benefit value is, the greater the collision with the vehicle behind the target lane is, and the lower the possibility of lane change of the driver is.

Further, the driving benefits are:

the reason that the target lane and the original lane have poor driving benefits causes the generation of lane changing intention, and the larger the benefit difference is, the stronger the lane changing intention of a driver is; the driving benefit of the original lane takes the distance L between the target vehicle and the front vehicle of the original lane into consideration_i(t) and velocity difference Δ v_i(t) the target lane driving benefit takes into account the distance L between the target vehicle and the vehicle ahead of the target lane_i(t) and velocity difference Δ v_i(t), in addition to this, a correction of the vehicle type is taken into account;

delta D for poor driving benefit_bExpressed, its formula is as follows:

ΔD_b＝D_bi-D_bj

D_bi＝(v_PV(t)-v_SV(t)+L_i(t))·α₁

D_bj＝(v_TPV(t)-v_SV(t)+L_j(t))·α₂

in the formula: d_biIs a benefit of driving on the original lane; d_bjAs a driving benefit for the target lane; v. of_SV(t) is a target vehicle speed (m/s); v. of_PV(t) is the speed of the target vehicle ahead of the original lane; v. of_PV(t) is the speed (m/s) of the target vehicle in front of the original lane; v. of_TPV(t) is the speed (m/s) of the vehicle ahead of the target lane; l is_i(t) is the distance (m) between the target vehicle and the vehicle in front of the original lane; l is_j(t) is the distance (m) between the target vehicle and the vehicle ahead of the target lane, α₁α is the correction coefficient of the front vehicle of the original lane₂The correction coefficient of the front vehicle of the target lane is obtained;

and (3) constructing a driving benefit difference membership function by combining measured data and the existing research, setting the domain scope of the lane driving benefit difference Db as {0, 5, 10, 15 and 20}, and setting the fuzzy set as { Very Small (VS), small (S), medium (M), large (L) and Very Large (VL) }.

Further, said collision severity T_c：

TTC is used for representing the intensity of the rear vehicle collision with the target lane and is recorded as T_cWhen the TTC is calculated, the length of the vehicle body should be considered, and the calculation formula is as follows:

in the formula: x is the number of_SV(t) is the current position (m) of the target vehicle SV; x is the number of_TFV(t) is the current position (m) of the vehicle TFV behind the target lane; v. of_SV(t) is the speed (m/s) of the target vehicle FV; v. of_TFV(t) TFV speed (m/s) of the vehicle behind the target lane; l is the body length (m).

Further, the driving type of the vehicle behind the target lane is as follows:

when the target vehicle generates the lane change intention, the intention of the vehicle to accept or reject the lane change request is different for different driver types and vehicles behind the target lane. The driver is divided into an aggressive type, a common type and a conservative type, the speed and the acceleration of the vehicle are used as model input quantity, the driving aggressive degree is used as model output quantity, and a fuzzy logic rule is constructed.

The speed and acceleration input values are obtained by the following formula:

wherein v is_i、a_iRespectively representing the velocity, acceleration, N at time i_vN is respectively the time frequency corresponding to the received speed and acceleration information; the driving motivation degree is respectively expressed by language variable motivation, common and conservative.

Furthermore, the variable cooperation coefficient is represented by p according to the lane change will and the driver type of other vehicles to carry out fuzzy reasoning on the vehicle cooperation competition degree; and p takes a value between 0 and 1, the lower the quantized value of p is, the stronger the competition among the vehicles is, and conversely, the higher the quantized value of p is, the higher the cooperation degree among the vehicles is.

Further, judging lane changing requirements, namely judging the lane changing requirements according to expected lane changing decision conditions, and selecting a target lane according to an expected lane changing decision;

selecting and adjusting the cooperation coefficient according to a lane changing mode, and if the target gap of the target lane is larger than the critical lane changing gap, selecting a free lane changing method to execute a lane changing decision; otherwise, if the target gap of the target lane is smaller than the critical lane change gap, selecting a cooperative lane change decision and establishing a driving game model; adjusting the cooperation coefficient according to the vehicle cooperation competition degree;

executing a lane change decision, solving a Nash equilibrium solution of the game model in each cycle to obtain an optimal probability combination of a cooperative lane change strategy or a competitive lane change strategy, and executing lane change according to the optimal strategy combination; and if the vehicle has the condition of changing the target lane or the target gap, the vehicle is considered to enter a next new lane changing decision process, the target lane or the target gap is reselected, and a new expected lane changing decision is made.

Further, the expectation-lane-change decision

(1) Safety guidelines

The acceleration value of the following vehicle estimated by the IDM model needs to be larger than the maximum safe deceleration; the acceleration of the target vehicle after lane changing is mainly influenced by a front guide vehicle on the target lane, and the acceleration is also larger than the maximum safe deceleration; therefore, after the target vehicle changes lanes, the target vehicle and the rear-mounted acceleration on the target lane should respectively satisfy the following formulas:

in the formula, b_safeFor a given maximum safe deceleration (m/s)²)。

(2) Criteria for benefits

On the basis of the IDM following model, the acceleration is used as the lane changing benefit, and whether the vehicle obtains a better driving expectation through the lane changing behavior is judged. The total lane change benefit is composed of the self benefit of the lane change vehicle and the benefit of the affected following vehicles, and when the total lane change benefit is larger than a given threshold value under the condition of meeting the safety constraint, the decision result of the model is lane change. The benefit criterion can thus be expressed by the following formula:

in the formula, p is a cooperative coefficient and takes a value of [ 0-1%]. When p is 0, the complete competition relationship is expressed, and when p is 1, the complete cooperation relationship is expressed; Δ a_thA benefit threshold for lane change;

the complete expected lane change decision model is to select all the candidate target lanes, namely to select the lane with the maximum total benefit from the candidate lanes meeting the requirements of the safety criterion and the benefit criterion as the lane change target lane. Therefore, the complete expectation-lane-change decision model is a constrained optimization model with the goal of maximizing the expected lane-change benefit, and the expression is as follows:

TL′＝argmax_TLN(SV)u_SV，TL∈N(SV)

subject to

u_SV(TL)＞Δa_th

in the formula, TL^*Represents the optimal target lane, u_SV(TL) Total benefit is expected for the lane change.

Therefore, the decision is not to change the lane when the optimization model has no solution, and the lane change of the vehicle to the corresponding target lane is represented when the optimization model has a reasonable optimal solution.

Further, multi-vehicle collaborative lane change decision modeling:

the target vehicle and the following vehicle of the target lane in a lane change situation, may create a lateral disturbance or conflict,

(1) the participators: the lane change behavior may be that of a vehicle that is creating a lateral disturbance or collision, the target vehicle SV and the trailing vehicle TFV of the target lane participating in the man-made lane change situation.

(2) Strategy: target vehicle SV has two pure strategies, lane change or no lane change, i.e. S₁＝{a₁ ⁽¹⁾， a₂ ⁽¹⁾-LC, LK }; the TFV of the following vehicle of the target lane can choose to accept the merging request or reject the request, adopt cooperation or competition, and the strategy is set as S₂＝{a₁ ⁽²⁾，a₂ ⁽²⁾If CO-operation is chosen, it is possible to slow down or change to another lane.

(3) And (4) payment: the joint earnings of the game-playing vehicles under different strategy combinations are respectively expressed as U₁＝(S₁,S₂)，U₂＝(S₁,S₂)。

The expected yield ESV and ETFV of the mixed probability of the target vehicle SV and the following vehicle TFV are the sum of the products of the yield and the corresponding probability of each mixed strategy:

wherein, theta₁＝θ,θ₂＝1-θ，λ₁＝λ,λ₂＝1-λ。

From the above, it can be seen that the optimal solution, i.e. solving nash equilibrium, can be obtained by maximizing the expected gains of SV and TFV of the vehicle;

the collaborative lane change decision model is an optimization model aiming at maximizing the expected joint action space benefit, and the expression is as follows:

in the formula (I), the compound is shown in the specification,

representing the optimal candidate action, E (U (a)) represents the expected benefit of the joint action space, p (a)_SV|a_TFV) Denotes a given a_SVUnder the condition that the vehicle selects a_TFVA probability of an action;

to combine the benefits of the action space

Can be expressed by the following formula:

in the formula (I), the compound is shown in the specification,

the benefit of the joint action space;

is given by a_TFVUnder the conditions, the benefits of the vehicle SV;

is given by a_SVUnder conditions, vehicle TFV benefits; p is a cooperative coefficient and takes a value of [ 0-1%]. When the vehicles are in complete competition relationship, p is equal to0; when the vehicle-to-vehicle cooperation relationship is complete, p is 1; and calculating the driving income according to the benefit criterion, and then judging whether the vehicle executes a lane change decision or not by combining the vehicle cooperation and the competition degree prediction result.

Further, the adjustment of the cooperation coefficient:

fuzzy reasoning vehicle cooperation competition degree according to the lane change will and the types of other drivers, and is represented by p; and p takes a value between 0 and 1, the lower the quantized value of p is, the more fierce the competition among the vehicles is, and on the contrary, the higher the quantized value of p is, the higher the cooperation degree among the vehicles is, and the result influences the value of the cooperation coefficient in the benefit formula. Redefining lane changing rules according to the vehicle cooperation competition degree;

(1) free lane changing

When the lane change target Gap TG is larger than the critical lane change Gap Gap0, the target vehicle SV has no influence on the TFV of the rear vehicle of the target lane, and when the benefit of the lane change vehicle is larger than the benefit of not changing the lane, the target vehicle SV changes the lane; the TFV of the rear vehicle of the target lane keeps the original speed to follow the front vehicle, so the speed and the position of the SV and the TFV are updated as follows in the free lane changing process:

the vehicle SV and TFV speeds are updated as:

the vehicle SV and TFV position updates are:

(2) collaborative lane changing

When there is a full cooperation relationship between vehicles, p is 1, then the benefit of the joint action space is expressed as:

the lane change decision simulation experiment is characterized in that A, B, C three groups of comparison items are respectively set, a group A of vehicles adopts a traditional lane change model based on a gap acceptance theory, a group B of vehicles adopts a lane change expecting model without considering coordination, and a group C of vehicles adopts a lane change game model with considering coordination.

To be defined as the benefit of the joint action space;

given a_TFVUnder the conditions, the benefits of the vehicle SV;

given a_SVUnder conditions, vehicle TFV benefits; when the lane change target Gap TG is smaller than the critical lane change Gap Gap0, if the TFV of the rear vehicle of the target lane decelerates and cooperates with pm probability, assuming that the deceleration of the TFV of the rear vehicle of the target lane at the next moment meets the lane change condition, the target vehicle SV performs lane change when the following formula is met;

v_SV(t+1)-v_TFV(t+1)+d_SV，TFV＞d_safeand E (U (LC) > E (U (LK))

In the formula: d_SV,TFVIs expressed as the distance (m) between the vehicle SV and the TFV;^a^ TFV is the deceleration (m/s) required for the lane change condition²)；^dsafeThe track spacing (m) is changed safely.

The vehicle SV and TFV speeds are updated as:

the vehicle SV and TFV position updates are:

(3) competitive lane change

When the vehicle is represented as a competitive relationship, p is 0, then the benefit of the joint action space is expressed as:

in the formula:

to be defined as the benefit of the joint action space;

the benefit of the vehicle SV under given conditions;

when the lane change target Gap TG is smaller than the critical lane change Gap0, if the rear vehicle TFV of the target lane is in probability acceleration competition, the speeds of the vehicles SV and TFV in the lane change process are respectively:

minimum distance d for no collision between SV and TFV at next moment_minComprises the following steps:

the target vehicle SV makes a lane change when the following equation is satisfied,

d_SV，TFV＞d_minand E (U (LC) > E (U (LK))

In the formula (d)_SV，TFVRepresents the distance (m) between the vehicle SV and TFV.

The invention has the beneficial effects that: a vehicle lane change interactive relation judgment model is established based on a fuzzy logic method so as to analyze vehicle-vehicle interactive behaviors under mixed-driving conditions, help the unmanned vehicle to understand and judge the behavior of the vehicles with people around; and several groups of road section driving data under the mixed running condition in the Chinese intelligent automobile competition are selected as input parameters of the model, the effectiveness of the vehicle lane change interactive relation judgment model is verified, and the result shows that the accuracy rate of the fuzzy reasoning method for predicting the cooperative competition degree in the vehicle lane change process exceeds 87.3%, the more extreme the vehicle cooperation and competition degree is, the higher the accuracy rate is, the unmanned vehicle has the capability of understanding human behaviors, and the foundation is laid for realizing multi-vehicle cooperative driving. By establishing the cooperative lane-changing game model of the manned vehicle and the unmanned vehicle, the traffic flow operation efficiency can be effectively improved, the driving experience of passengers can be greatly improved, and the traffic safety is positively influenced.

Drawings

FIG. 1 is a vehicle lane change interaction concept; (ii) a

FIG. 2 is a schematic diagram illustrating a lane change interaction relationship determination for a vehicle;

FIG. 3 is a diagram of a road section driving vehicle interaction relationship prediction result;

FIG. 4 is a flow chart of a vehicle lane change process;

FIG. 5 is a diagram of a lane change game scenario;

FIG. 6 is a distribution diagram of lane change times of vehicles under different traffic flow densities;

FIG. 7 is a graph comparing the simulated average traffic flow at different densities;

fig. 8 is a graph of the average transit time of the vehicle under different density conditions.

Detailed Description

A fuzzy logic method is applied to establish a vehicle lane change interactive relationship judgment model, the model comprehensively considers the lane change intention of both interactive parties and the types of drivers of other vehicles, the cooperation competition degree between vehicles is subjected to fuzzy reasoning, and the mapping relationship between the output value of the fuzzy reasoning and the interactive relationship prediction result is established to represent the cooperation and competition relationship of other vehicles in the interactive process. By analyzing the vehicle-vehicle interaction behavior under the mixed-driving condition, the unmanned vehicle is helped to understand and judge the behavior of the people around the unmanned vehicle.

The invention divides the lane changing interactive behavior into three modes of free lane changing behavior, cooperation lane changing behavior and competition lane changing behavior.

(1) Free lane change behavior

In the free lane changing process with stable traffic flow running state and good road conditions, the target vehicle is a better driving expectation, and directly changes lanes under the condition of meeting lane changing conditions without communication and negotiation with the rear vehicle of the target lane and interactive behaviors.

(2) Collaborative lane change behavior

And under the condition that the lane changing condition is insufficient, the target vehicle sends a lane changing request to the rear vehicle of the target lane, and the rear vehicle of the target lane allows the lane changing request to be decelerated at the same time to increase the lane changing gap so as to provide a condition for changing the lane of the target vehicle. Therefore, the interactive behavior among the vehicles in the cooperative lane changing process is very obvious, and the result is finished by the deceleration and avoidance of the vehicles behind the target lane in cooperation with the successful lane changing of the lane changing vehicles.

(3) Competitive lane change behavior

When the lane change clearance in the driving scene is smaller than the critical clearance, the target vehicle sends a lane change request to the rear vehicle of the target lane under the condition that the lane change condition is insufficient, and the rear vehicle of the target lane refuses the lane change request. At the moment, the TFV and the SV of the rear vehicle of the target lane compete for road resources, the game is usually carried out for a plurality of times, the lane change behavior can be completed only when two vehicles are stiff near a lane line for a plurality of seconds, and the process is the competitive lane change behavior. Therefore, in the process of competing lane changing, the target vehicle SV adheres to the lane changing requirement under the condition of insufficient lane changing conditions, and the target vehicle SV mutually 'squeeze' to complete lane changing under the competing conditions, so that great interference is generated on a target lane.

First, vehicle lane changing interactive relation model

In combination with understanding of lane change behavior interaction characteristics, when a vehicle lane change interaction relation determination model is established, the method comprises the following two steps: the first step is to judge the level of the lane change intention of the target vehicle on the original lane; and the second step is to judge the driving type of the vehicle behind the target lane. And the interactive relation between the vehicles is deduced by combining the two steps, so that the degree of cooperation between the vehicles is represented.

1. The level of the intention of the target vehicle to change the lane of the original lane

The main judgment criterion for lane changing of the vehicle is whether a better driving space can be obtained under the condition of meeting the safety. Thus, the increased Driving benefits (Driving Benefit) and Δ D of the vehicle following a lane change in this chapter_bAnd the degree of conflict with the rear vehicle of the target lane is used as a model input quantity, and the degree of lane change will is used as an output variable.

1) Model variables and membership functions

The level of the lane change will of the target vehicle is generally related to the following two factors:

(1) benefits of driving

The purpose of the lane change behavior is to improve the driving benefit of the vehicle, and the greater the driving benefit increased after the lane change of the target vehicle, the stronger the motivation of the lane change, and the stronger the lane change requirement of the vehicle, as shown in fig. 3. The reason that the target lane and the original lane have poor driving benefits causes the lane changing intention, and the greater the benefit difference is, the stronger the lane changing intention of the driver is. The driving benefit of the original lane takes the distance L between the target vehicle and the front vehicle of the original lane into consideration_i(t) and velocity difference Δ v_i(t) the interest of the target lane driving takes into account the distance L between the target vehicle and the vehicle ahead of the target lane_i(t) and velocity difference Δ v_i(t), in addition to this, correction of the vehicle type is also taken into account.

Delta D for poor driving benefit_bExpressed, its formula is as follows:

ΔD_b＝D_bi-D_bj

D_bi＝(v_PV(t)-v_SV(t)+L_i(t))·α₁

D_bj＝(v_TPV(t)-v_SV(t)+L_j(t))·α₂

in the formula: d_biIs a benefit of driving on the original lane; d_bj-a target lane driving benefit; v. of_SV(t) -target vehicle speed (m/s); v. of_PV(t) -the speed of the vehicle ahead of the target vehicle on the original lane; v. of_PV(t) -speed (m/s) ahead of the original lane target vehicle; v. of_TPV(t) -target lane front speed (m/s); l is_i(t) -the distance (m) of the target vehicle from the vehicle ahead of the original lane; l is_j(t) distance (m) of target vehicle from the front of target lane α₁Before the original laneCorrection factor α₂-correction factor of the vehicle ahead of the target lane.

And (3) constructing a driving benefit difference membership function by combining measured data and the existing research, setting the domain scope of the lane driving benefit difference Db as {0, 5, 10, 15 and 20}, and setting the fuzzy set as { Very Small (VS), small (S), medium (M), large (L) and Very Large (VL) }.

(2) Severity of conflict T_c

When the lane change intention is generated by the vehicle, the lane change operation is not executed immediately, and whether the lane change safety condition is met needs to be further judged. TTC (Time to Collision) is used for representing the intensity of Collision with the rear vehicle of the target lane and is marked as T_c. In the rear-end collision, the collision point is that the rear vehicle head contacts with the front vehicle tail, and the vehicle body length is considered when calculating the TTC, so the calculation formula is as follows:

in the formula: x is the number of_SV(t) is the current position (m) of the target vehicle SV; x is the number of_TFV(t) is the current position (m) of the vehicle TFV behind the target lane; x is the number of_SV(t) is the speed (m/s) of the target vehicle FV; v. of_TFV(t) TFV speed (m/s) of the vehicle behind the target lane; l is the body length (m).

The domain scope of the severity c T of the collision between the target vehicle and the rear vehicle of the target lane is taken as {0, 3, 5, 7, 10} by combining the measured data and the existing research, and the fuzzy set is { Very Large (VL), large (L), medium (M), small (S) and Very Small (VS) }.

(3) Fuzzy control rule

The lane change will level is expressed by language variables of extremely low (PL), low (L), medium (M), high (H) and extremely high (PH), and thus the fuzzy rule of the lane change will of the vehicle is determined as shown in the table 3.2.

TABLE 3.2 fuzzy logic rule Table for lane change willingness

From the above analysis, when the lane is onDriving benefit value Δ D_bThe larger the collision T between the vehicle and the target lane_cThe smaller, the more likely the vehicle is to choose a lane change in pursuit of better driving conditions; conversely, the value of interest Δ D for lane driving_bThe smaller the collision T with the rear vehicle of the target lane_cThe larger the lane change probability of the driver.

Secondly, driving type of vehicle carried behind target lane

When the target vehicle generates the lane change intention, the intention of the vehicle to accept or reject the lane change request is different for different driver types and vehicles behind the target lane. Such as more conservative drivers being more likely to give way or compromise, and more aggressive drivers being more likely to give way to other vehicles, the driver type of the other vehicle is therefore closely related to the vehicle interaction.

The human driver can be divided into an aggressive type, a common type and a conservative type according to the driving aggressive degree, the speed and the acceleration of the vehicle are used as model input quantity, the driving aggressive degree is used as model output quantity, a fuzzy logic rule is constructed, and the speed and the acceleration input value are obtained through the following formula.

Wherein v is_i、a_iRespectively representing the velocity, acceleration, N at time i_vAnd N are the time times corresponding to the received speed and acceleration information respectively.

Combining measured data and the existing research, taking the domain range of the speed of the target vehicle behind as {5, 10, 15, 20, 25}, and taking the fuzzy set as { Very Small (VS), small (S), medium (M), large (L), Very Large (VL) }; the domain range of the absolute value of the acceleration is taken as {1, 2, 3}, and the fuzzy set is { small (S), medium (M), large (L) }.

And determining the fuzzy rules of the types of drivers of other vehicles according to an expert experience method and an observation method, wherein the driving motivation degrees are respectively expressed by linguistic variables of motivation (A), common (N) and conservation (C), and are shown in a table 3.3.

TABLE 3.3 fuzzy logic rules Table

Fuzzy inference

According to the level of the lane change intention of the target vehicle and the driving type of the vehicle behind the target lane, the level of the lane change intention and the type of the driver of the other vehicle are used as input variables, the output is the level of the vehicle cooperation competition degree, the values of the vehicle cooperation degree are expressed by language variables of high (H), medium (M) and low (L), the fuzzy logic rules of the vehicle lane change interaction relationship are respectively corresponding to 3 results of the other vehicle interaction relationship, namely the cooperation relationship, the ambiguity and the competition relationship, and are shown in a table 3.4.

TABLE 3.4 fuzzy logic rule table of interaction relation for changing lane of vehicle

The fuzzy implication relations of the partial fuzzy reasoning adopt a Mamdani rule, the ambiguity resolution adopts a gravity center method, and the corresponding relation between the vehicle cooperation competition degree, the lane change will and the driving type is obtained.

Establishing a vehicle lane change interactive relation judgment model by using a fuzzy logic method, introducing a variable cooperation coefficient on the basis of an expected lane change decision model, providing a cooperative lane change game model of the manned and unmanned vehicles, and performing Nash equilibrium solution on the game model by using a Lemke-Howson algorithm to obtain an optimal strategy combination for judging whether the vehicle lane change is performed or not; and (3) carrying out simulation experiment verification on the vehicle lane change decision under the mixed-running condition through SUMO simulation software, and setting comparison items of a lane change model (A group), an expected lane change decision model (B group) and a collaborative lane change game model (C group) which are acceptable based on gaps in the experiment respectively. According to the traffic flow simulated by the three groups of experiments, the traffic flow simulation system analyzes and evaluates the indexes such as traffic efficiency, stability and safety, and the result shows that the lane changing game model (group C) with multiple vehicles in cooperation has better overall performance, higher safety guarantee and reduced loss of the traffic flow operation efficiency. Compared with other systems, the method considers all possible optimization schemes and considers real-time prediction results, and the robustness of the system is improved.

The lane change interactive relationship judgment model of the vehicle judges the lane change intention of the vehicle and the driving types of other vehicles by fuzzy reasoning, and integrates the lane change intention and the driving types of other vehicles to show the degree of cooperation between the vehicles; aiming at the lane change intention of the vehicle, the Driving Benefit (Driving Benefit) increased after the lane change of the vehicle and the degree of conflict with the vehicle behind the target lane are used as model input quantities, and the degree of the lane change intention is used as an output variable; and aiming at the driving type of the vehicle behind the target lane, the vehicle speed and the acceleration are used as model input quantities, and the driving acceleration degree is used as a model output quantity.

Three-way and cooperative lane changing game model

An action is selected for the vehicle to maximize the combined expected utility E (u (a)), and thus the collaborative lane-change decision model is an optimization model aimed at maximizing the expected joint action spatial benefit.

The variable cooperation coefficient is represented by p according to the lane change will and the driver type of other vehicles to carry out fuzzy reasoning on the vehicle cooperation competition degree. And p takes a value between 0 and 1, the lower the quantized value of p is, the stronger the competition among the vehicles is, and conversely, the higher the quantized value of p is, the higher the cooperation degree among the vehicles is.

The method comprises the following steps: gathering lane change data sets

In the road section driving state, the system starts to acquire the conditions of the target vehicle and other vehicles in the communication range.

Step two: collaborative lane change requirement determination

Firstly, judging lane change requirements according to expected lane change decision conditions, and selecting a target lane according to an expected lane change decision.

Step three: lane change mode selection and cooperative coefficient adjustment

If the target Gap of the target lane is larger than the critical lane change Gap, TG is more than or equal to Gap₀Then a free lane change method is selected to execute the lane change decision. Otherwise, if the target gap of the target lane is smaller than the critical lane change gapI.e. TG < Gap₀And selecting a cooperative lane change decision to establish a driving game model. And adjusting the cooperation coefficient according to the vehicle cooperation competition degree.

Step four: performing lane change decisions

And (3) solving a Nash equilibrium solution of the game model in each cycle, which is detailed in section 4.4.3, obtaining the optimal probability combination of the cooperative channel changing strategy or the competitive channel changing strategy, and executing channel changing according to the optimal strategy combination. And if the vehicle has the condition of changing the target lane or the target gap, the vehicle is considered to enter a next new lane changing decision process, the target lane or the target gap is reselected, and a new expected lane changing decision is made.

Fourth, expectation lane change decision

(1) Safety guidelines

From the following theory, it is known that the acceleration of the following of a vehicle is not substantially influenced by the following vehicle. The acceleration of the trailing vehicle on the target lane is therefore mainly influenced by the target vehicle. The speed is mainly affected by the target vehicle. The acceleration value of the following vehicle estimated by the IDM model needs to be larger than the maximum safe deceleration, so that the gap is larger than the critical gap at the moment, and no collision accident occurs. Deceleration, which ensures that the gap is greater than the critical gap, and no collision accident occurs. Similarly, the acceleration of the target vehicle after changing lane is mainly influenced by the leading vehicle on the target lane, and the acceleration is also larger than the maximum safe deceleration. Therefore, after the target vehicle changes lanes, the target vehicle and the rear-mounted acceleration on the target lane should respectively satisfy the following formulas:

in the formula, b_safeFor a given maximum safe deceleration (m/s)²)。

(2) Criteria for benefits

On the basis of the IDM following model, the acceleration is used as the lane changing benefit, and whether the vehicle obtains a better driving expectation through the lane changing behavior is judged. The total lane change benefit is composed of the self benefit of the lane change vehicle and the benefit of the affected following vehicles, and when the total lane change benefit is larger than a given threshold value under the condition of meeting the safety constraint, the decision result of the model is lane change. The benefit criterion can thus be expressed by the following formula:

in the formula, p is a cooperative coefficient and takes a value of [ 0-1%]. When p is 0, the complete competition relationship is expressed, and when p is 1, the complete cooperation relationship is expressed; Δ a_thFor lane change benefit thresholds.

The complete expected lane change decision model is to select all the candidate target lanes, namely to select the lane with the maximum total benefit from the candidate lanes meeting the requirements of the safety criterion and the benefit criterion as the lane change target lane. Therefore, the complete expectation-lane-change decision model is a constrained optimization model with the goal of maximizing the expected lane-change benefit, and the expression is as follows:

TL＝argmax_TLeN(SV)u_SV，TL∈N(SV)

subject to

u_SV(TL)＞Δa_th

in the formula, TL^*Represents the optimal target lane, u_SV(TL) Total benefit is expected for the lane change.

Therefore, the decision is not to change the lane when the optimization model has no solution, and the lane change of the vehicle to the corresponding target lane is represented when the optimization model has a reasonable optimal solution. The model integrates the generation of lane change requirements and the judgment of lane change feasibility into one model, and can effectively simulate the vehicle micro-driving behaviors of road section scenes. The model is designed to set the total benefit threshold value of lane changing so as to prevent the traffic flow from being influenced by random lane changing and frequent lane changing of vehicles.

Five-vehicle and multi-vehicle collaborative lane change decision modeling

The target vehicle and the following vehicle of the target lane in a lane change situation, may create a lateral disturbance or conflict,

(1) the participators: the lane change behavior may be that of a vehicle that is creating a lateral disturbance or collision, the target vehicle SV and the trailing vehicle TFV of the target lane participating in the man-made lane change situation.

(2) Strategy: target vehicle SV has two pure strategies, lane change or no lane change, i.e. S₁＝{a₁ ⁽¹⁾， a₂ ⁽¹⁾-LC, LK }; the TFV of the following vehicle of the target lane can choose to accept the merging request or reject the request, adopt cooperation or competition, and the strategy is set as S₂＝{a₁ ⁽²⁾，a₂ ⁽²⁾If CO-operation is chosen, it is possible to slow down or change to another lane.

(3) And (4) payment: the joint earnings of the game-playing vehicles under different strategy combinations are respectively expressed as U₁＝(S₁,S₂)，U₂＝(S₁,S₂)。

Therefore, the game model accords with the condition of double-matrix game and can be expressed as G ═ S₁，S₂；U₁， U₂}. Then this dual matrix game can be as shown in table 4.2:

TABLE 4.2 Driving strategy double matrix Game model

The expected yield ESV and ETFV of the mixed probability of the target vehicle SV and the following vehicle TFV are the sum of the products of the yield and the corresponding probability of each mixed strategy:

wherein, theta₁＝θ,θ₂＝1-θ，λ₁＝λ,λ₂＝1-λ。

From the above equations, it can be seen that the optimal solution, i.e., solving for Nash equilibrium, can be obtained by maximizing the expected revenue for the vehicles SV and TFV.

The collaborative lane change decision model is an optimization model aiming at maximizing the expected joint action space benefit, and the expression is as follows:

in the formula (I), the compound is shown in the specification,

representing the optimal candidate action, E (U (a)) represents the expected benefit of the joint action space, p (a)_SV|a_TFV) Denotes a given a_SVUnder the condition that the vehicle selects a_TFVA probability of an action;

to combine the benefits of the action space

Can be expressed by the following formula:

in the formula (I), the compound is shown in the specification,

the benefit of the joint action space;

is given by a_TFVUnder the conditions, the benefits of the vehicle SV;

is given by a_SVUnder conditions, vehicle TFV benefits; p is a cooperative coefficient and takes a value of [ 0-1%]. When the vehicles are in complete competition relationship, p is 0; when there is a full cooperation relationship between the vehicles, p is 1.

And calculating the driving income according to the benefit criterion, and then judging whether the vehicle executes a lane change decision or not by combining the vehicle cooperation and the competition degree prediction result.

Sixthly, adjustment of cooperation coefficient

And (4) fuzzy reasoning the vehicle cooperation competition degree according to the lane change will and the types of drivers of other vehicles, and expressing the vehicle cooperation competition degree by p. p takes a value between 0 and 1, the lower the quantized value of p is, the more fierce the competition among the vehicles is, on the contrary, the higher the quantized value of p is, the higher the cooperation degree among the vehicles is, and the result influences the cooperation coefficient value in the benefit formula (4.15). The lane change rule is redefined according to the degree of vehicle cooperation competition.

(1) Free lane changing

When the lane change target Gap TG is larger than the critical lane change Gap Gap0, the target vehicle SV has no influence on the TFV of the rear vehicle of the target lane, and when the benefit of the lane change vehicle is larger than the benefit of not changing the lane, the target vehicle SV changes the lane. The TFV of the rear vehicle of the target lane keeps the original speed to follow the front vehicle, so the speed and the position of the SV and the TFV are updated as follows in the free lane changing process:

the vehicle SV and TFV speeds are updated as:

the vehicle SV and TFV position updates are:

(2) collaborative lane changing

When there is a full cooperation relationship between vehicles, p is 1, then the benefit of the joint action space is expressed as:

the lane change decision simulation experiment is characterized in that A, B, C three groups of comparison items are respectively set, a group A of vehicles adopt a traditional lane change model based on a gap acceptance theory, a group B of vehicles adopt a lane change expecting model without considering coordination, and a group C of vehicles adopt a lane change game model considering coordination.

To be defined as the benefit of the joint action space;

given a_TFVUnder the conditions, the benefits of the vehicle SV;

given a_SVVehicle TFV benefits under the conditions. When the lane change target Gap TG is smaller than the critical lane change Gap0, if the target lane rear vehicle TFV decelerates and cooperates with pm probability, assuming that the deceleration of the target lane rear vehicle TFV satisfies the lane change condition at the next moment, the target vehicle SV changes lanes when the following formula is satisfied.

V_SV(t+1)-v_TFV(t+1)+d_SV，TFV＞d_safeAnd E (U (LC) > E (U (LK))

In the formula: d_SV，TFVIs expressed as the distance (m) between the vehicle SV and the TFV;^a^ TFV is the deceleration (m/s) required for the lane change condition²)；^dsafeThe track spacing (m) is changed safely.

The vehicle SV and TFV speeds are updated as: the vehicle SV and TFV position updates are:

(3) competitive lane change

When the vehicle is represented as a competitive relationship, p is 0, then the benefit of the joint action space is expressed as:

in the formula:

to be defined as the benefit of the joint action space;

the benefit of the vehicle SV under given conditions.

When the lane change target Gap TG is smaller than the critical lane change Gap0, if the rear vehicle TFV of the target lane is in probability acceleration competition, the speeds of the vehicles SV and TFV in the lane change process are respectively:

minimum distance d for no collision between SV and TFV at next moment_minComprises the following steps:

therefore, when the following expression is satisfied, the target vehicle SV makes a lane change, and the speeds and positions of the vehicle SV and TFV are updated by the same expression (4.24) and expression (4.25).

d_SV，TFV＞d_minAnd E (U (LC) > E (U (LK))

In the formula (d)_SV，TFVRepresents the distance (m) between the vehicle SV and TFV.

Examples

As shown in fig. 1, the driving states of the unmanned vehicle and the manned vehicle are acquired based on the environment sensing module, a vehicle interaction relation determination model is established by using fuzzy reasoning, and the unmanned vehicle decision is made according to the prediction result obtained by the fuzzy reasoning.

As shown in fig. 2, in combination with understanding of lane change behavior interaction characteristics, when a vehicle lane change interaction relationship determination model is established, the method is divided into two steps: the first step is to judge the level of the lane change intention of the target vehicle on the original lane; and the second step is to judge the driving type of the vehicle behind the target lane. And the interactive relation between the vehicles is deduced by combining the two steps, so that the degree of cooperation between the vehicles is represented.

As shown in fig. 3, 150 sets of real road segment lane change data samples are selected for vehicle cooperation and competition degree classification, wherein 3 is cooperation relation and represents the yielding intention of other vehicles, 2 is ambiguous and represents the uncertain state of other vehicles, and 1 is competition relation and represents the failing intention of other vehicles. From the prediction results in the figure, it can be seen that the model correctly identifies 48 samples of the cooperative relationship, 38 samples of the ambiguous state, 45 samples of the competitive relationship, and a total of 131 correctly identified samples, and the prediction accuracy is 87.3%. The more extreme the vehicle cooperation competition degree is, the higher the accuracy is, the accuracy of the prediction result 1 or 3 exceeds 90%, which shows that the unmanned vehicle has the capability of understanding human behaviors, and lays a foundation for the cooperation of the unmanned vehicle and the manned vehicle.

As shown in fig. 4, the lane change decision process is divided into four steps, and a lane change data set is collected: under the driving state of the road section, the system starts to acquire the conditions of the target vehicle and other vehicles in the communication range; judging the cooperative lane change requirement: firstly, judging lane change requirements according to an expected lane change decision condition, and selecting a target lane according to an expected lane change decision; lane changing mode selection and cooperation coefficient adjustment: if the target Gap of the target lane is larger than the critical lane change Gap, TG is more than or equal to Gap₀Then a free lane change method is selected to execute the lane change decision. On the contrary, if the target Gap of the target lane is smaller than the critical lane change Gap, namely TG < Gap₀And selecting a cooperative lane change decision to establish a driving game model. Adjusting the cooperation coefficient according to the vehicle cooperation competition degree; executing a lane change decision: and solving a Nash equilibrium solution of the game model in each cycle to obtain an optimal probability combination of the cooperative channel changing strategy or the competitive channel changing strategy, and executing channel changing according to the optimal strategy combination. If the vehicle has the situation of changing the target lane or the target gap, the vehicle is considered to enter the next new lane changing decision process, and the target lane or the target gap is reselectedTarget gap, a new desired lane change decision is made.

As shown in fig. 5, a target vehicle SV in a lane change situation and a following vehicle TFV of the target lane may generate a lateral disturbance or a collision.

As shown in fig. 6, the distribution of the lane change times of vehicles under different traffic flow densities in three lane scenes is described, and under the condition that the traffic flow density is low, the lane change times of vehicles in the simulated traffic flow of the group a are not greatly different from those of the group B and the group C, which indicates that when the traffic flow density is low, the vehicle clearance is increased, and the lane change action is overall low; under the condition of high traffic flow density, the number of lane changing times of vehicles in the simulated traffic flow of the group B and the group C is higher than that of the group A, but the number of lane changing times of the group C is reduced compared with that of the group B, which shows that when the traffic flow density is high, the vehicle clearance is reduced, the driving space is blocked, the vehicles expect to change lanes to obtain a better driving space, and because of the lane changing conflict of a certain degree between the vehicles, the cooperative lane changing decision in the group C aims at the overall system benefit, so that frequent and inappropriate lane changing behaviors are avoided, and the number of lane changing times is reduced compared with that of the group B.

As shown in fig. 7, when the density is less than 20Veh/km, the simulated traffic flow of the group a, the group B and the group C is equivalent, and when the density is greater than 25Veh/km, the traffic flow of the group C is significantly higher than that of the group a and the group B, because frequent and inappropriate lane changing behavior in the simulated traffic flow of the group B causes influence on other vehicles around, disturbance to the traffic flow is increased, and the stability of the traffic flow is reduced, while in the simulated traffic flow of the group C, the overall benefit of the system is optimized through vehicle-vehicle cooperation, which indicates that the invention can improve the road traffic capacity.

As shown in fig. 8, in the simulated traffic flow of group C, the average passing time of vehicles is better than that of groups a and B, and it can be seen that the game lane change model has a higher effective utilization rate of road resources.

The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (10)

1. The utility model provides an intelligent vehicle under highway section thoughtlessly moves condition trades way model in coordination which characterized in that: establishing a vehicle lane change interactive relation judgment model based on a fuzzy logic method, analyzing vehicle-vehicle interactive behaviors under mixed traffic conditions, taking driving benefits increased after the vehicle lane change and the degree of conflict with a vehicle behind a target lane as model input quantities for the lane change intention of the vehicle, and taking the degree of the lane change intention as output variables; aiming at the driving type of the vehicle behind the target lane, the vehicle speed and the acceleration are used as model input quantities, and the driving acceleration degree is used as a model output quantity;

according to the lane change intention of a target vehicle and the driving type of a vehicle behind a target lane, the lane change intention and the type of drivers of other vehicles are used as input variables, the output is the vehicle cooperation competition degree, the vehicle cooperation degree is expressed by language variables with high, medium and low values, the interaction relations of the other vehicles are 3 results of cooperation relations, ambiguous and competition relations, fuzzy implication relations of fuzzy reasoning adopt a Mamdani rule, ambiguity resolution adopts a gravity center method, and the corresponding relations between the vehicle cooperation competition degree, the lane change intention and the driving type are obtained by establishing a cooperative lane change game model of an unmanned vehicle; variable cooperation coefficients are introduced to establish a cooperative lane changing game model of the manned vehicle and the unmanned vehicle, and a Lemke-Howson algorithm is adopted to carry out Nash equilibrium solving on the game model to obtain the optimal strategy combination of whether the lane of the vehicle is changed or not.

2. The intelligent vehicle collaborative lane changing model under the road section mixed traveling condition according to claim 1, wherein the vehicle lane changing interactive relationship model is divided into two steps:

the first step is to judge the level of the lane change intention of the target vehicle on the original lane;

the lane change will of the target vehicle on the original lane comprises the following steps:

(1) a driving benefit; the driving benefit of the original lane takes into account the distance between the target vehicle and the front vehicle of the original laneIs far from L_i(t) and velocity difference Δ v_i(t) the target lane driving benefit takes into account the distance L between the target vehicle and the vehicle ahead of the target lane_i(t) and velocity difference Δ v_i(t)

(2) Severity of conflict T_c

TTC (Time to Collision) is used for representing the intensity of Collision with the rear vehicle of the target lane and is marked as T_cThe length of the vehicle body is considered when calculating the TTC;

judging the driving type of the vehicle behind the target lane; the interactive relation between the vehicles is deduced by integrating the two steps, and the cooperation degree between the vehicles is represented; the driver is divided into an aggressive type, a common type and a conservative type, the speed and the acceleration of the vehicle are used as model input quantity, the driving aggressive degree is used as model output quantity, and a fuzzy logic rule is constructed;

when the lane driving benefit value is larger and the collision between the vehicle and the vehicle behind the target lane is smaller, the vehicle is more likely to select lane change to pursue better driving conditions; conversely, the smaller the lane driving benefit value is, the greater the collision with the vehicle behind the target lane is, and the lower the possibility of lane change of the driver is.

3. The intelligent vehicle collaborative lane changing model under the road section mixing condition according to claim 2, wherein the driving benefits are as follows:

the reason that the target lane and the original lane have poor driving benefits causes the generation of lane changing intention, and the larger the benefit difference is, the stronger the lane changing intention of a driver is; the driving benefit of the original lane takes the distance L between the target vehicle and the front vehicle of the original lane into consideration_i(t) and velocity difference Δ v_i(t) the target lane driving benefit takes into account the distance L between the target vehicle and the vehicle ahead of the target lane_i(t) and velocity difference Δ v_i(t), in addition to this, a correction of the vehicle type is taken into account;

delta D for poor driving benefit_bExpressed, its formula is as follows:

ΔD_b＝D_bi-D_bj

D_bi＝(v_PV(t)-v_SV(t)+L_i(t))·α₁

D_bj＝(v_TPV(t)-v_SV(t)+L_j(t))·α₂

in the formula: d_biIs a benefit of driving on the original lane; d_bjAs a driving benefit for the target lane; v. of_SV(t) is a target vehicle speed (m/s); v. of_PV(t) is the speed of the target vehicle ahead of the original lane; v. of_PV(t) is the speed (m/s) of the target vehicle in front of the original lane; v. of_TPV(t) is the speed (m/s) of the vehicle ahead of the target lane; l is_i(t) is the distance (m) between the target vehicle and the vehicle in front of the original lane; l is_j(t) is the distance (m) between the target vehicle and the vehicle ahead of the target lane, α₁α is the correction coefficient of the front vehicle of the original lane₂The correction coefficient of the front vehicle of the target lane is obtained;

and (3) constructing a driving benefit difference membership function by combining measured data and the existing research, setting the domain scope of the lane driving benefit difference Db as {0, 5, 10, 15 and 20}, and setting the fuzzy set as { Very Small (VS), small (S), medium (M), large (L) and Very Large (VL) }.

4. The intelligent vehicle collaborative lane changing model under the road section mixing condition according to claim 2, wherein the conflict severity degree T_c：

TTC is used for representing the intensity of the rear vehicle collision with the target lane and is recorded as T_cWhen the TTC is calculated, the length of the vehicle body should be considered, and the calculation formula is as follows:

in the formula: x is the number of_SV(t) is the current position (m) of the target vehicle SV; x is the number of_TFV(t) is the current position (m) of the vehicle TFV behind the target lane; v. of_SV(t) is the speed (m/s) of the target vehicle FV; v. of_TFV(t) TFV speed (m/s) of the vehicle behind the target lane; l is the body length (m).

5. The intelligent vehicle collaborative lane changing model under the road section mixed-traveling condition according to claim 1, wherein the driving type of the vehicle behind the target lane is:

when the target vehicle generates the lane change intention, the intention of the vehicle to accept or reject the lane change request is different for the vehicles behind the target lane and different driver types; the method is divided into three types, namely an aggressive type, a common type and a conservative type, the vehicle speed and the acceleration are used as model input quantities, the driving aggressive degree is used as a model output quantity, and a fuzzy logic rule is constructed;

the speed and acceleration input values are obtained by the following formula:

wherein v is_i、a_iRespectively representing the velocity, acceleration, N at time i_vN is respectively the time frequency corresponding to the received speed and acceleration information; the driving motivation degree is respectively expressed by language variable motivation, common and conservative.

6. The intelligent vehicle collaborative lane changing model under the road section mixed traveling condition according to claim 1, wherein: the variable cooperation coefficient is represented by p according to the lane change will and the driver type of other vehicles to carry out fuzzy reasoning on the vehicle cooperation competition degree; and p takes a value between 0 and 1, the lower the quantized value of p is, the stronger the competition among the vehicles is, and conversely, the higher the quantized value of p is, the higher the cooperation degree among the vehicles is.

7. The intelligent vehicle collaborative lane changing model under the road section mixed traveling condition according to claim 1, wherein: judging lane changing requirements, namely judging the lane changing requirements according to expected lane changing decision conditions, and selecting a target lane according to an expected lane changing decision;

selecting and adjusting the cooperation coefficient according to a lane changing mode, and if the target gap of the target lane is larger than the critical lane changing gap, selecting a free lane changing method to execute a lane changing decision; otherwise, if the target gap of the target lane is smaller than the critical lane change gap, selecting a cooperative lane change decision and establishing a driving game model; adjusting the cooperation coefficient according to the vehicle cooperation competition degree;

executing a lane change decision, solving a Nash equilibrium solution of the game model in each cycle to obtain an optimal probability combination of a cooperative lane change strategy or a competitive lane change strategy, and executing lane change according to the optimal strategy combination; and if the vehicle has the condition of changing the target lane or the target gap, the vehicle is considered to enter a next new lane changing decision process, the target lane or the target gap is reselected, and a new expected lane changing decision is made.

8. The intelligent vehicle collaborative lane changing method under the road section mixed traveling condition according to claim 7, wherein the method comprises the following steps: the expectation lane change decision

(1) Safety guidelines

The acceleration value of the following vehicle estimated by the IDM model needs to be larger than the maximum safe deceleration; the acceleration of the target vehicle after lane changing is mainly influenced by a front guide vehicle on the target lane, and the acceleration is also larger than the maximum safe deceleration; therefore, after the target vehicle changes lanes, the target vehicle and the rear-mounted acceleration on the target lane should respectively satisfy the following formulas:

in the formula, b_safeFor a given maximum safe deceleration (m/s)²)；

(2) Criteria for benefits

On the basis of the IDM following model, the acceleration is used as lane changing benefit, and whether the vehicle obtains better driving expectation through lane changing behavior is judged; the lane change total benefit is composed of the self benefit of the lane change vehicle and the benefit of the affected following vehicles, and when the lane change total benefit is larger than a given threshold value under the condition of meeting the safety constraint, the model decision result is lane change; the benefit criterion can thus be expressed by the following formula:

in the formula, p is a cooperative coefficient and takes a value of [ 0-1%](ii) a When p is 0, the complete competition relationship is expressed, and when p is 1, the complete cooperation relationship is expressed; Δ a_thA benefit threshold for lane change;

the complete expected lane change decision model is to select all candidate target lanes, namely to select the lane with the maximum total benefit from the candidate lanes meeting the requirements of the safety criterion and the benefit criterion as a lane change target lane; therefore, the complete expectation-lane-change decision model is a constrained optimization model with the goal of maximizing the expected lane-change benefit, and the expression is as follows:

TL^*＝arg max_TLeN(SV)u_SV，TL∈N(SV)

subject to

u_SV(TL)＞Δa_th

in the formula, TL^*Represents the optimal target lane, u_SV(TL) total expected benefit for lane change;

therefore, the decision is not to change the lane when the optimization model has no solution, and the lane change of the vehicle to the corresponding target lane is represented when the optimization model has a reasonable optimal solution.

9. The intelligent vehicle collaborative lane change method under the road section mixed traveling condition according to claim 1, characterized in that multi-vehicle collaborative lane change decision modeling:

the target vehicle and the following vehicle of the target lane in a lane change situation, may create a lateral disturbance or conflict,

(1) the participators: vehicles with possibly generated transverse interference or conflict in lane changing behavior participate in target vehicles SV and trailing vehicles TFV of a target lane in the situation of artificial lane changing;

(2) strategy: target vehicle SV has two pure strategies, lane change or no lane change, i.e. S₁＝{a₁ ⁽¹⁾，a₂ ⁽¹⁾-LC, LK }; the TFV of the following vehicle of the target lane can choose to accept the merging request or reject the request, adopt cooperation or competition, and the strategy is set as S₂＝{a₁ ⁽²⁾，a₂ ⁽²⁾If cooperation is selected, the speed can be reduced or other lanes can be changed;

(3) and (4) payment: the joint earnings of the game-playing vehicles under different strategy combinations are respectively expressed as U₁＝(S₁,S₂)，U₂＝(S₁,S₂)；

The expected yield ESV and ETFV of the mixed probability of the target vehicle SV and the following vehicle TFV are the sum of the products of the yield and the corresponding probability of each mixed strategy:

wherein, theta₁＝θ,θ₂＝1-θ，λ₁＝λ,λ₂＝1-λ；

From the above, it can be seen that the optimal solution, i.e. solving nash equilibrium, can be obtained by maximizing the expected gains of SV and TFV of the vehicle; the collaborative lane change decision model is an optimization model aiming at maximizing the expected joint action space benefit, and the expression is as follows:

in the formula (I), the compound is shown in the specification,

representing the optimal candidate action, E (U (a)) represents the expected benefit of the joint action space, a (a)_SV|a_TFV) Denotes a given a_SVUnder the condition that the vehicle selects a_TFVA probability of an action;

to combine the benefits of the action space

Can be expressed by the following formula:

in the formula (I), the compound is shown in the specification,

the benefit of the joint action space;

is given by a_TFVUnder the conditions, the benefits of the vehicle SV;

is given by a_SVUnder conditions, vehicle TFV benefits; p is a cooperative coefficient and takes a value of [ 0-1%](ii) a When the vehicles are in complete competition relationship, p is 0; when the vehicle-to-vehicle cooperation relationship is complete, p is 1; and calculating the driving income according to the benefit criterion, and then judging whether the vehicle executes a lane change decision or not by combining the vehicle cooperation and the competition degree prediction result.

10. The intelligent vehicle collaborative lane changing method under the road section mixing condition according to claim 1, wherein the adjustment of the collaborative coefficient:

fuzzy reasoning vehicle cooperation competition degree according to the lane change will and the types of other drivers, and is represented by p; p takes a value between 0 and 1, the lower the quantized value of p is, the more fierce the competition among the vehicles is, otherwise, the higher the quantized value of p is, the higher the cooperation degree among the vehicles is, and the result can influence the value of the cooperation coefficient in the benefit formula; redefining lane changing rules according to the vehicle cooperation competition degree;

(1) free lane changing

When the lane change target Gap TG is larger than the critical lane change Gap Gap0, the target vehicle SV has no influence on the TFV of the rear vehicle of the target lane, and when the benefit of the lane change vehicle is larger than the benefit of not changing the lane, the target vehicle SV changes the lane; the TFV of the rear vehicle of the target lane keeps the original speed to follow the front vehicle, so the speed and the position of the SV and the TFV are updated as follows in the free lane changing process:

the vehicle SV and TFV speeds are updated as:

the vehicle SV and TFV position updates are:

(2) collaborative lane changing

When there is a full cooperation relationship between vehicles, p is 1, then the benefit of the joint action space is expressed as:

the lane change decision simulation experiment is characterized in that A, B, C three groups of comparison items are respectively set, wherein the A group of vehicles adopt a traditional lane change model based on a gap acceptance theory, the B group of vehicles adopt a single expected lane change model without considering coordination, and the C group of vehicles adopt a lane change game model with considering coordination;

to be defined as the benefit of the joint action space;

given a_TFVUnder the conditions, the benefits of the vehicle SV;

given a_SVUnder conditions, vehicle TFV benefits; when the lane change target Gap TG is smaller than the critical lane change Gap Gap0, if the TFV of the rear vehicle of the target lane decelerates and cooperates with pm probability, assuming that the deceleration of the TFV of the rear vehicle of the target lane at the next moment meets the lane change condition, the target vehicle SV performs lane change when the following formula is met;

v_SV(t+1)-v_TFV(t+1)+d_SV，TFV＞d_safeand E (U (LC) > E (U (LK))

In the formula: d_SV,TFVIs expressed as the distance (m) between the vehicle SV and the TFV;^a^ TFV is the deceleration (m/s) required for the lane change condition²) (ii) a dsafe is the safe lane change distance (m);

the vehicle SV and TFV speeds are updated as:

the vehicle SV and TFV position updates are:

(3) competitive lane change

When the vehicle is represented as a competitive relationship, p is 0, then the benefit of the joint action space is expressed as:

in the formula:

to be defined as the benefit of the joint action space;

under given conditions, the vehicleThe benefit of the vehicle SV;

when the lane change target Gap TG is smaller than the critical lane change Gap0, if the rear vehicle TFV of the target lane is in probability acceleration competition, the speeds of the vehicles SV and TFV in the lane change process are respectively:

minimum distance d for no collision between SV and TFV at next moment_minComprises the following steps:

the target vehicle SV makes a lane change when the following equation is satisfied,

d_SV，TFV＞d_minand E (U (LC) > E (U (LK))

In the formula (d)_SV，TFVRepresents the distance (m) between the vehicle SV and TFV.

Images