CN111994079A - A non-cooperative game lane change assistant decision-making system and method considering driving style characteristics - Google Patents

Abstract

The invention discloses a non-cooperative game lane change auxiliary decision-making system and method considering driving style characteristics. Driving environment data set; the lane-changing motivation judgment module is used to receive the above-mentioned driving environment data set, and analyze and calculate it to determine whether the driver has a lane-changing motivation; the lane-changing income calculation and judgment module is used to receive the above-mentioned lane-changing motivation instruction , carry out game profit calculation and driving style calculation, obtain the result of lane change income and judge the level of income; the lane change assistant decision prompt module prompts the driver to perform lane change or give up lane change according to the above income level information. The invention deeply analyzes the game lane changing scene and the influence of the driving style and combines the two, so that the proposed model decision method makes the lane change decision efficiency higher than the simple game model-based lane change decision efficiency.

Description

A non-cooperative game lane change assistant decision-making system and method considering driving style characteristics

technical field

The invention relates to the technical field of vehicle safety, and in particular, to a non-cooperative game lane changing auxiliary decision-making system and method considering driving style characteristics.

Background technique

During the driving of the vehicle, the driver adopts different driving behaviors according to the changes of the road environment and surrounding vehicle information. In the two driving behaviors of lane-changing and car-following, due to the complexity of the lane-changing process, it is more likely to cause traffic accidents due to wrong lane-changing decisions. In the networked environment, the driver assistance system can provide effective decision assistance for the driver's lane changing behavior through a comprehensive perception of the surrounding environment and the vehicle.

Most of the lane-changing decision-making models used in the existing lane-changing driving assistance systems are established based on the laws of vehicle kinematics, which can only give an early warning of the danger of lane-changing before the vehicle changes lanes, and cannot accurately reflect the perceptual decision-making that is affected by the driver's factors. The process of realization is difficult to reflect the change of the driver's driving behavior during the lane change process.

SUMMARY OF THE INVENTION

In view of the above-mentioned deficiencies of the prior art, the purpose of the present invention is to provide a non-cooperative game lane-changing assistance decision-making system and method considering driving style characteristics, so as to solve the problem that the lane-changing driving assistance system in the prior art cannot accurately reflect the driver The factors that affect the realization process of perceptual decision-making are difficult to reflect the change of the driver's driving behavior during the lane changing process. By combining the idea of non-cooperative game theory with driver characteristics, the invention makes the lane-changing assistant decision-making method more flexible and applicable.

For achieving the above object, the technical scheme adopted in the present invention is as follows:

A non-cooperative game lane change auxiliary decision-making system considering driving style characteristics of the present invention includes:

The driving environment data collection module is used to collect the position data and motion data of the surrounding mobile units and stationary units of the connected vehicle to obtain the driving environment data set;

The lane-changing motivation judgment module is used to receive the above-mentioned driving environment data set, and analyze and calculate it to determine whether the driver has a lane-changing motivation, and if "yes", generate a lane-changing motivation instruction;

The lane-changing income calculation and judgment module is used to receive the above-mentioned lane-changing motivation instruction, perform game income calculation and driving style calculation, obtain the lane-changing income result and judge the level of income;

The lane change assistant decision prompt module, according to the above-mentioned income level information, prompts the driver to perform lane change or give up lane change.

Further, the method for judging whether the lane-changing motive judging module has a lane-changing motive is as follows:

v_i+1分别表示目标车辆的期望速度、实际速度以及前导车辆的实际速度；v^bar为影响目标车辆继续行驶的物体的移动速度。Among them, Δx _i represents the distance between the target vehicle and the obstacle; T ^safe represents the safe time distance, and T ^min represents the minimum reaction time;

v _i+1 respectively represent the expected speed, actual speed of the target vehicle and the actual speed of the leading vehicle; v ^bar is the moving speed of the object that affects the target vehicle to continue driving.

Further, the method for calculating the lane change income and the judging module calculation is as follows:

The formula for calculating the profit function is as follows:

G or g=α ₁ *ε+α ₂ *δ

Among them, ε and δ are the safety benefit and time-effectiveness benefit obtained by the current vehicle when the lane change decision is made, respectively, and their values are calculated by the following formula:

In the formula, S _min is the minimum safe distance required to make a lane change decision; t is the time required to reach the target location, t ₀ is the time required to reach the destination in the original state, S is the distance to the destination, and the benefits The calculation of the function is converted into the minimum safe distance required to make a lane change decision and the time it takes to reach the destination;

Among them, α ₁ and α ₂ are weight coefficients. Different styles of drivers have different requirements for efficiency and safety, and the weights will be different. The initial value is arbitrary, and α ₁ , α ₂ ∈ {α ₁ +α ₂ =1, 0<α ₁ , α ₂ <1}, the value rules for α ₁ and α ₂ are as follows:

α ₂ =1-α ₁

In the formula, R _driver is the driver style recognition coefficient, and the calculation formula is as follows:

分别为识别域中冲击度J(t)的标准差和标准型驾驶员在当前工况下的平均值；冲击度J(t)以车辆行驶速度v(t)衡量，其定义为：In the formula, R _J ,

are the standard deviation of the shock degree J(t) in the recognition domain and the average value of the standard driver under the current working conditions; the shock degree J(t) is measured by the vehicle speed v(t), which is defined as:

In the formula, the average value of J(t) is 0.59, 0.31, 0.26, and 0.25 in crowded conditions, urban conditions, suburban conditions, and expressway conditions, respectively.

A non-cooperative game lane change auxiliary decision-making method considering driving style characteristics of the present invention, the steps are as follows:

Step 1) judge whether the driver has a lane-changing motive according to the collected driving environment data; if there is no lane-changing motive, then enter step 4), if there is a lane-changing motive, then enter step 2);

Step 2) make a game lane change decision, and construct a game profit matrix;

Step 3) According to the income matrix obtained in the above-mentioned step 2), solve all Nash-equilibrium mixed strategy combinations, and determine the level of income; if the income is low, the driver is prompted to give up changing lanes, and enter step 4); if the income is high, the driver is prompted. It is recommended to change lanes and go to step 5);

Step 4) It is recommended to give up changing lanes and go to step 6);

Step 5) It is recommended to perform a lane change, and enter step 6);

Step 6) The decision is over.

Further, whether the driver in the described step 1) has a method for judging whether the motive for changing lanes is as follows:

v_i+1分别表示目标车辆的期望速度、实际速度以及前导车辆的实际速度；v^bar为影响目标车辆继续行驶的物体的移动速度。In the formula, Δx _i represents the distance between the target vehicle and the obstacle; T ^safe represents the safe time distance, and T ^min represents the minimum reaction time;

v _i+1 respectively represent the expected speed, actual speed of the target vehicle and the actual speed of the leading vehicle; v ^bar is the moving speed of the object that affects the target vehicle to continue driving.

Further, the step 2) specifically includes: under a static non-cooperative game with complete information, then there is, defining the strategy set of the target vehicle M as Φ ₁ ={C,N}, where C represents lane change, and N represents no Lane changing, the corresponding probabilities of the strategies are p, 1-p respectively; the strategy set of the following vehicle B ₁ in the adjacent lane is Φ ₂ ={D,R}, where D is the lane change allowed, R is the refusal to change the lane, and its The corresponding probability of the strategy is q, 1-q; the income of M and B ₁ is represented by G _ij and g _ij , then the game income matrix is obtained;

The target vehicle _M changes lanes, and the following vehicle B1 in the adjacent lane is allowed to change lanes. _At this time, the gains of M and B1 are _G11 and _g11 respectively; the target vehicle _M does not change lanes, and the following vehicle B1 in the adjacent lane is allowed to change lanes, At this time, the gains of M and B1 are G ₁₂ and g ₁₂ respectively; the target vehicle M changes lanes, and the following vehicle B ₁ is not allowed to change lanes, at this time, the gains of M and B ₁ are G ₂₁ and g ₂₁ respectively _; the target vehicle M does not change lanes, and the following vehicle B ₁ in the adjacent lane is not allowed to change lanes. At this time, the benefits of M and B ₁ are G ₂₂ and g ₂₂ respectively.

Further, the step 3) specifically includes: calculating each profit function of the mixed strategy, and the calculation formula of the profit function is as follows:

G=α ₁ *ε+α ₂ *δ (1)

Among them, ε and δ are the safety benefits and time-effectiveness benefits obtained by the current vehicle in the lane-changing decision, respectively; α ₁ , α ₂ are the weight coefficients, and drivers of different styles have different requirements for efficiency and safety, and the weights will be different. The initial value Any, and satisfy α ₁ , α ₂ ∈ {α ₁ +α ₂ =1, 0<α ₁ , α ₂ <1};

The driving style is quantified, and the impact degree R _driver generated during driving represents the driver's style recognition coefficient. The calculation formula is as follows:

分别为识别域中冲击度J(t)的标准差和标准型驾驶员在当前工况下的平均值；冲击度J(t)以车辆行驶速度v(t)衡量，其定义为：where R _J ,

are the standard deviation of the shock degree J(t) in the recognition domain and the average value of the standard driver under the current working conditions; the shock degree J(t) is measured by the vehicle speed v(t), which is defined as:

Among them, the average value of J(t) is 0.59, 0.31, 0.26, and 0.25 in crowded conditions, urban conditions, suburban conditions, and expressway conditions, respectively;

Calculate the impact degree R _driver generated during driving according to the vehicle speed information during the monitoring time, and compare the results with the constants _Ragg and R _norm ; R _norm represents the threshold value of normal driving style, which is 0.5, and _Ragg represents aggressive driving The style threshold, which takes a value of 1; if R _driver < R _norm , the driver's driving style is cautious; if R _driver > R _agg , the driver's style is aggressive, and if it is between the two room, it belongs to the ordinary type.

When the driver's driving style is different, the requirements for safety and time benefit are different, that is, the values of α ₁ and α ₂ are different:

α ₂ =1-α ₁ (5)

The calculation method of security benefit and aging benefit is as follows:

In the formula, S _min is the minimum safe distance required for decision-making; t is the current time required to reach the target location, t ₀ is the time required to reach the destination in the original state, and S is the distance to the destination; the benefit function The calculation is converted into the minimum safe distance required to make a decision and the time it takes to reach the destination;

In the car-following state, the leading car decides the reaction of the car-following car, then the minimum safe distance model is:

Among them, V _F (t) is the speed of the preceding vehicle, _VL (t) is the speed of the following vehicle, β ₁ =1/w is the reaction time of the current vehicle to the leading vehicle, w is the width of the vehicle, and β ₂ is the maximum speed of the current vehicle The reciprocal of twice the deceleration, according to the parameter calibration, there is γ=-β ₁ ;

When changing lanes, the possible collisions of vehicles changing lanes are oblique collisions. The minimum safe distance for lane changing is modeled. The conditions for _M and B1 to avoid collisions are:

and

Let the distance between the two cars in the direction of travel be:

Among them, θ is the angle between the tangential direction of the vehicle M's travel trajectory and the longitudinal direction of the road;

but:

a_M分别是B₁和M车的纵向加速度；Equation (12) is the necessary and sufficient condition for non-collision lane change, where,

a _M are the longitudinal accelerations of B ₁ and M cars, respectively;

As long as h>0 is guaranteed during lane changing, no collision will occur, thus we can get:

In the same way, the minimum safe distance for changing lanes is:

The time taken to reach the destination is obtained by the distance to the destination and the initial speed of the vehicle.

Further, the step 3) specifically also includes: under the Nash equilibrium strategy combination, the decision made by any game participant is the optimal decision for other participants; The described game profit matrix, the probability of M and B ₁ 's decision-making is recorded as vector x = (x ₁ , x ₂ ), y = (y ₁ , y ₂ ) ^T , then the solution of the mixed strategy Nash equilibrium is obtained. for:

Equivalent to:

Satisfying (p ^* , q ^* ) in the payoff calculation in the game payoff matrix is the solution of a mixed-strategy Nash equilibrium combination of a non-cooperative game lane change.

Further, according to the solution obtained in step 3), the profit of the target vehicle changing lanes is determined. When ε>1000, the safety benefit is high; otherwise, it is low; when δ>800, the aging benefit is high, otherwise it is low; then the benefit The high-low judgment rules are as follows:

When G>1000*α ₁ +800*α ₂ , the target vehicle lane-changing profit is high, otherwise it is low.

Beneficial effects of the present invention:

The present invention can assist the driver to make a lane change decision in a closer real scene, and introduce the style coefficient into the income calculation, so that the decision-making method is closer to the actual driving situation, reflecting the influence of different styles on the expected income; at the same time, it is beneficial to the driving assistance system. development and lane change decisions as a connected car. The existing lane-changing decision based on cellular automata model only analyzes the lane-changing vehicles, ignoring the existence of the dynamic interaction of related vehicles, and its lane-changing rules are relatively strict, which makes the lane-changing efficiency of the target vehicle low. However, the present invention deeply analyzes the game lane changing scene and the influence of the driving style and combines the two, so that the proposed model decision method makes the lane change decision more efficient than the simple game model-based lane change decision.

Description of drawings

Fig. 1 is the principle block diagram of the system of the present invention.

Figure 2 is a flow chart of the method of the present invention.

Detailed ways

In order to facilitate the understanding of those skilled in the art, the present invention will be further described below with reference to the embodiments and the accompanying drawings, and the contents mentioned in the embodiments are not intended to limit the present invention.

Referring to Fig. 1, a non-cooperative game lane change auxiliary decision-making system considering driving style characteristics of the present invention includes:

The driving environment data collection module is used to collect the position data and motion data of the surrounding mobile units and stationary units of the connected vehicle to obtain the driving environment data set;

The lane-changing motivation judgment module is used to receive the above-mentioned driving environment data set, and analyze and calculate it to determine whether the driver has a lane-changing motivation, and if "yes", generate a lane-changing motivation instruction;

The lane-changing income calculation and judgment module is used to receive the above-mentioned lane-changing motivation instruction, perform game income calculation and driving style calculation, obtain the lane-changing income result and judge the level of income;

The lane change assistant decision prompt module, according to the above-mentioned income level information, prompts the driver to perform lane change or give up lane change.

Wherein, the method for judging whether the lane-changing motive judging module has a lane-changing motive is as follows:

v_i+1分别表示目标车辆的期望速度、实际速度以及前导车辆的实际速度；v^bar为影响目标车辆继续行驶的物体的移动速度。Among them, Δx _i represents the distance between the target vehicle and the obstacle; T ^safe represents the safe time distance, and T ^min represents the minimum reaction time;

v _i+1 respectively represent the expected speed, actual speed of the target vehicle and the actual speed of the leading vehicle; v ^bar is the moving speed of the object that affects the target vehicle to continue driving.

Wherein, the method for calculating the lane change income and the judging module calculation is as follows:

The formula for calculating the profit function is as follows:

G or g=α ₁ *ε+α ₂ *δ

Among them, ε and δ are the safety benefit and time-effectiveness benefit obtained by the current vehicle when the lane change decision is made, respectively, and their values are calculated by the following formula:

In the formula, S _min is the minimum safe distance required to make a lane change decision; t is the time required to reach the target location, t ₀ is the time required to reach the destination in the original state, S is the distance to the destination, and the benefits The calculation of the function is converted into the minimum safe distance required to make a lane change decision and the time it takes to reach the destination;

Among them, α ₁ and α ₂ are weight coefficients. Different styles of drivers have different requirements for efficiency and safety, and the weights will be different. The initial value is arbitrary, and α ₁ , α ₂ ∈ {α ₁ +α ₂ =1, 0<α ₁ , α ₂ <1}, the value rules for α ₁ and α ₂ are as follows:

α ₂ =1-α ₁

In the formula, R _driver is the driver style recognition coefficient, and the calculation formula is as follows:

分别为识别域中冲击度J(t)的标准差和标准型驾驶员在当前工况下的平均值；冲击度J(t)以车辆行驶速度v(t)衡量，其定义为：In the formula, R _J ,

are the standard deviation of the shock degree J(t) in the recognition domain and the average value of the standard driver under the current working conditions; the shock degree J(t) is measured by the vehicle speed v(t), which is defined as:

In the formula, the average value of J(t) is 0.59, 0.31, 0.26, and 0.25 in crowded conditions, urban conditions, suburban conditions, and expressway conditions, respectively.

Referring to Fig. 2, a non-cooperative game lane change auxiliary decision-making method considering driving style characteristics of the present invention, the steps are as follows:

Step 1) judge whether the driver has a lane-changing motive according to the collected driving environment data; if there is no lane-changing motive, then enter step 4), if there is a lane-changing motive, then enter step 2);

The method for judging whether the driver has the motive to change lanes is as follows:

v_i+1分别表示目标车辆的期望速度、实际速度以及前导车辆的实际速度；v^bar为影响目标车辆继续行驶的物体的移动速度。In the formula, Δx _i represents the distance between the target vehicle and the obstacle; T ^safe represents the safe time distance, and T ^min represents the minimum reaction time;

v _i+1 respectively represent the expected speed, actual speed of the target vehicle and the actual speed of the leading vehicle; v ^bar is the moving speed of the object that affects the target vehicle to continue driving.

Step 2) make a game lane change decision, and construct a game profit matrix;

Under the static non-cooperative game with complete information, then, the strategy set of the target vehicle M is defined as Φ ₁ ={C,N}, where C means lane change, N means no lane change, and the corresponding probabilities of the strategies are p, 1-p; the strategy set of the following vehicle B ₁ in the adjacent lane is Φ ₂ ={D,R}, where D is the allowed lane change, R is the refusal to change the lane, and the corresponding probability of the policy is q, 1-q; M And B ₁ income is represented by G _ij , g _ij , then the game income matrix is obtained;

The target vehicle _M changes lanes, and the following vehicle B1 in the adjacent lane is allowed to change lanes. _At this time, the gains of M and B1 are _G11 and _g11 respectively; the target vehicle _M does not change lanes, and the following vehicle B1 in the adjacent lane is allowed to change lanes, At this time, the gains of M and B1 are G ₁₂ and g ₁₂ respectively; the target vehicle M changes lanes, and the following vehicle B ₁ is not allowed to change lanes, at this time, the gains of M and B ₁ are G ₂₁ and g ₂₁ respectively _; the target vehicle M does not change lanes, and the following vehicle B ₁ in the adjacent lane is not allowed to change lanes. At this time, the benefits of M and B ₁ are G ₂₂ and g ₂₂ respectively.

Step 3) According to the income matrix obtained in the above-mentioned step 2), solve all Nash-equilibrium mixed strategy combinations, and determine the level of income; if the income is low, the driver is prompted to give up changing lanes, and enter step 4); if the income is high, the driver is prompted. It is recommended to change lanes and go to step 5);

Calculate each profit function of the mixed strategy, and the calculation formula of the profit function is as follows:

G=α ₁ *ε+α ₂ *δ (1)

Among them, ε and δ are the safety benefits and time-effectiveness benefits obtained by the current vehicle in the lane-changing decision, respectively; α ₁ , α ₂ are the weight coefficients, and drivers of different styles have different requirements for efficiency and safety, and the weights will be different. The initial value Any, and satisfy α ₁ , α ₂ ∈ {α ₁ +α ₂ =1, 0<α ₁ , α ₂ <1};

The driving style is quantified, and the impact degree R _driver generated during driving represents the driver's style recognition coefficient. The calculation formula is as follows:

分别为识别域中冲击度J(t)的标准差和标准型驾驶员在当前工况下的平均值；冲击度J(t)以车辆行驶速度v(t)衡量，其定义为：where R _J ,

are the standard deviation of the shock degree J(t) in the recognition domain and the average value of the standard driver under the current working conditions; the shock degree J(t) is measured by the vehicle speed v(t), which is defined as:

Among them, the average value of J(t) is 0.59, 0.31, 0.26, and 0.25 in crowded conditions, urban conditions, suburban conditions, and expressway conditions, respectively;

Calculate the impact degree R _driver generated during driving according to the vehicle speed information during the monitoring time, and compare the results with the constants _Ragg and R _norm ; R _norm represents the threshold value of normal driving style, which is 0.5, and _Ragg represents aggressive driving The style threshold, which takes a value of 1; if R _driver < R _norm , the driver's driving style is cautious; if R _driver > R _agg , the driver's style is aggressive, and if it is between the two room, it belongs to the ordinary type.

When the driver's driving style is different, the requirements for safety and time benefit are different, that is, the values of α ₁ and α ₂ are different:

α ₂ =1-α ₁ (5)

The calculation method of security benefit and aging benefit is as follows:

In the formula, S _min is the minimum safe distance required for decision-making; t is the current time required to reach the target location, t ₀ is the time required to reach the destination in the original state, and S is the distance to the destination; the benefit function The calculation is converted into the minimum safe distance required to make a decision and the time it takes to reach the destination;

In the car-following state, the leading car decides the reaction of the car-following car, then the minimum safe distance model is:

Among them, V _F (t) is the speed of the preceding vehicle, _VL (t) is the speed of the following vehicle, β ₁ =1/w is the reaction time of the current vehicle to the leading vehicle, w is the width of the vehicle, and β ₂ is the maximum speed of the current vehicle The reciprocal of twice the deceleration, according to the parameter calibration, there is γ=-β ₁ ;

When changing lanes, the possible collisions of vehicles changing lanes are oblique collisions. The minimum safe distance for lane changing is modeled. The conditions for _M and B1 to avoid collisions are:

and

Let the distance between the two cars in the direction of travel be:

Among them, θ is the angle between the tangential direction of the vehicle M's travel trajectory and the longitudinal direction of the road;

but:

a_M分别是B₁和M车的纵向加速度；Equation (12) is the necessary and sufficient condition for non-collision lane change, where,

a _M are the longitudinal accelerations of B ₁ and M cars, respectively;

As long as h>0 is guaranteed during lane changing, no collision will occur, thus we can get:

In the same way, the minimum safe distance for changing lanes is:

The time taken to reach the destination is obtained by the distance to the destination and the initial speed of the vehicle.

The step 3) specifically further includes: under the Nash equilibrium strategy combination, the decision made by any game participant is the optimal decision for other participants; on this basis, compare the game described in step 2). Profit matrix, the probability of M and B ₁ selected decisions are recorded as vectors x = (x ₁ , x ₂ ), y = (y ₁ , y ₂ ) ^T , then the solution of the mixed strategy Nash equilibrium is:

Equivalent to:

Satisfying (p ^* , q ^* ) in the payoff calculation in the game payoff matrix is the solution of a mixed-strategy Nash equilibrium combination of a non-cooperative game lane change.

According to the solution obtained in step 3), determine the income level of the target vehicle changing lanes, when ε>1000, the safety income is high; otherwise, it is low; when δ>800, the aging income is high, otherwise it is low; then the income level judgment rule as follows:

When G>1000*α ₁ +800*α ₂ , the target vehicle lane-changing profit is high, otherwise it is low.

Step 4) It is recommended to give up changing lanes and go to step 6);

Step 5) It is recommended to perform a lane change, and enter step 6);

Step 6) The decision is over.

There are many specific application ways of the present invention, and the above are only the preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, without departing from the principles of the present invention, several improvements can be made. These Improvements should also be considered as the protection scope of the present invention.

Claims (9)

1. a non-cooperative game lane change auxiliary decision-making system considering driving style characteristics, is characterized in that, comprises: The driving environment data collection module is used to collect the position data and motion data of the surrounding mobile units and stationary units of the connected vehicle to obtain the driving environment data set; a lane-changing motive judgment module, used for receiving the above-mentioned driving environment data set, and analyzing and calculating it, judging whether the driver has a lane-changing motive, and if so, generating a lane-changing motive instruction; The lane-changing income calculation and judgment module is used to receive the above-mentioned lane-changing motivation instruction, perform game income calculation and driving style calculation, obtain the lane-changing income result and judge the level of income; The lane change assistant decision prompt module, according to the above-mentioned income level information, prompts the driver to perform lane change or give up lane change. 2. the non-cooperative game lane-changing auxiliary decision-making system considering driving style characteristics according to claim 1, is characterized in that, the judging method that described lane-changing motive judging module judges whether to have lane-changing motive is as follows:

Among them, Δx _i represents the distance between the target vehicle and the obstacle; T ^safe represents the safe time distance, and T ^min represents the minimum reaction time;

v _i+1 respectively represent the expected speed, actual speed of the target vehicle and the actual speed of the leading vehicle; v ^bar is the moving speed of the object that affects the target vehicle to continue driving. 3. the non-cooperative game lane-changing auxiliary decision-making system considering driving style characteristics according to claim 1, is characterized in that, the method that described lane-changing income is calculated and judged module calculates is as follows: The formula for calculating the profit function is as follows: G or g=α ₁ *ε+α ₂ *δ Among them, ε and δ are the safety benefit and time-effectiveness benefit obtained by the current vehicle when the lane change decision is made, respectively, and their values are calculated by the following formula:

In the formula, S _min is the minimum safe distance required to make a lane change decision; t is the time required to reach the target location, t ₀ is the time required to reach the destination in the original state, S is the distance to the destination, and the benefits The calculation of the function is converted into the minimum safe distance required to make a lane change decision and the time it takes to reach the destination; Among them, α ₁ and α ₂ are weight coefficients. Different styles of drivers have different requirements for efficiency and safety, and the weights will be different. The initial value is arbitrary, and α ₁ , α ₂ ∈ {α ₁ +α ₂ =1, 0<α ₁ , α ₂ <1}, the value rules for α ₁ and α ₂ are as follows:

α ₂ =1-α ₁ In the formula, R _driver is the driver style recognition coefficient, and the calculation formula is as follows:

In the formula, R _J ,

are the standard deviation of the shock degree J(t) in the recognition domain and the average value of the standard driver under the current working conditions; the shock degree J(t) is measured by the vehicle speed v(t), which is defined as:

In the formula, the average value of J(t) is 0.59, 0.31, 0.26, and 0.25 in crowded conditions, urban conditions, suburban conditions, and expressway conditions, respectively. 4. A non-cooperative game lane-changing auxiliary decision-making method considering driving style characteristics, characterized in that the steps are as follows: Step 1) judge whether the driver has a lane-changing motive according to the collected driving environment data; if there is no lane-changing motive, then enter step 4), if there is a lane-changing motive, then enter step 2); Step 2) make a game lane change decision, and construct a game profit matrix; Step 3) According to the income matrix obtained in the above-mentioned step 2), solve all Nash-equilibrium mixed strategy combinations, and determine the level of income; if the income is low, the driver is prompted to give up changing lanes, and enter step 4); if the income is high, the driver is prompted. It is recommended to change lanes and go to step 5); Step 4) It is recommended to give up changing lanes and go to step 6); Step 5) It is recommended to perform a lane change, and enter step 6); Step 6) The decision is over. 5. the non-cooperative game lane-changing auxiliary decision-making method considering driving style characteristics according to claim 1, is characterized in that, whether the driver in described step 1) has the judging method of lane-changing motivation as follows:

In the formula, Δx _i represents the distance between the target vehicle and the obstacle; T ^safe represents the safe time distance, and T ^min represents the minimum reaction time;

v _i+1 respectively represent the expected speed, actual speed of the target vehicle and the actual speed of the leading vehicle; v ^bar is the moving speed of the object that affects the target vehicle to continue driving. 6. The non-cooperative game lane-changing auxiliary decision-making method considering driving style characteristics according to claim 1, wherein the step 2) specifically comprises: under the static non-cooperative game of complete information, then there is, defining a target vehicle The strategy set of M is Φ ₁ ={C,N}, C means changing lanes, N means not changing lanes, the corresponding probabilities of the strategies are p, 1-p respectively; the strategy set of the following vehicle B ₁ in the adjacent lane is Φ ₂ ={D,R}, D is the permission to change lanes, R is the refusal to change lanes, and the corresponding probability of the strategy is q, 1-q; M and B ₁ income are represented by G _ij , g _ij , then the game income matrix is obtained;

The target vehicle _M changes lanes, and the following vehicle B1 in the adjacent lane is allowed to change lanes. _At this time, the gains of M and B1 are _G11 and _g11 respectively; the target vehicle _M does not change lanes, and the following vehicle B1 in the adjacent lane is allowed to change lanes, At this time, the gains of M and B1 are G ₁₂ and g ₁₂ respectively; the target vehicle M changes lanes, and the following vehicle B ₁ is not allowed to change lanes, at this time, the gains of M and B ₁ are G ₂₁ and g ₂₁ respectively _; the target vehicle M does not change lanes, and the following vehicle B ₁ in the adjacent lane is not allowed to change lanes. At this time, the benefits of M and B ₁ are G ₂₂ and g ₂₂ respectively. 7. The non-cooperative game lane-changing auxiliary decision-making method considering driving style characteristics according to claim 1, wherein the step 3) specifically comprises: calculating each profit function of the mixed strategy, and the profit function calculation formula is as follows: G=α ₁ *ε+α ₂ *δ (1) In the formula, ε and δ are the safety and timeliness benefits obtained by the current vehicle in the lane-changing decision, respectively; α ₁ and α ₂ are the weight coefficients, and drivers of different styles have different needs for efficiency and safety, and the weights will be different. The value is arbitrary and satisfies α ₁ , α ₂ ∈ {α ₁ +α ₂ =1, 0<α ₁ , α ₂ <1}; The driving style is quantified, and the impact degree R _driver generated during driving represents the driver's style recognition coefficient. The calculation formula is as follows:

In the formula, R _J ,

are the standard deviation of the shock degree J(t) in the recognition domain and the average value of the standard driver under the current working conditions; the shock degree J(t) is measured by the vehicle speed v(t), which is defined as:

In the formula, the average value of J(t) is 0.59, 0.31, 0.26, and 0.25 in crowded conditions, urban conditions, suburban conditions, and expressway conditions, respectively; Calculate the impact degree R _driver generated during driving according to the vehicle speed information during the monitoring time, and compare the results with the constants _Ragg and R _norm ; R _norm represents the threshold value of normal driving style, which is 0.5, and _Ragg represents aggressive driving The style threshold, which takes a value of 1; if R _driver < R _norm , the driver's driving style is cautious; if R _driver > R _agg , the driver's style is aggressive, and if it is between the two room, it belongs to the ordinary type; When the driver's driving style is different, the requirements for safety and time benefit are different, that is, the values of α ₁ and α ₂ are different:

α ₂ =1-α ₁ (5) The calculation method of security benefit and aging benefit is as follows:

In the formula, S _min is the minimum safe distance required for decision-making; t is the current time required to reach the target location, t ₀ is the time required to reach the destination in the original state, and S is the distance to the destination; the benefit function The calculation is converted into the minimum safe distance required to make a decision and the time it takes to reach the destination; In the car-following state, the leading car decides the reaction of the car-following car, then the minimum safe distance model is:

In the formula, V _F (t) is the speed of the preceding vehicle, _VL (t) is the speed of the following vehicle, β ₁ =1/w is the reaction time of the current vehicle to the leading vehicle, w is the width of the vehicle, and β ₂ is the current vehicle’s speed. The reciprocal of twice the maximum deceleration, according to parameter calibration, there is γ=-β ₁ ; When changing lanes, the possible collisions of vehicles changing lanes are oblique collisions. The minimum safe distance for lane changing is modeled. The conditions for _M and B1 to avoid collisions are:

and

Let the distance between the two cars in the direction of travel be:

Among them, θ is the angle between the tangential direction of the vehicle M's travel trajectory and the longitudinal direction of the road; but:

Equation (12) is the necessary and sufficient condition for non-collision lane change, where,

a _M are the longitudinal accelerations of B ₁ and M cars, respectively; As long as h>0 is guaranteed during lane changing, no collision will occur, thus we can get:

In the same way, the minimum safe distance for changing lanes is:

The time taken to reach the destination is obtained by the distance to the destination and the initial speed of the vehicle. 8. the non-cooperative game lane-changing auxiliary decision-making method considering driving style characteristics according to claim 7, is characterized in that, described step 3) specifically also comprises: under Nash equilibrium strategy combination, any one game participant's action. The decision made is the optimal decision for other participants; on this basis, according to the game profit matrix described in step 2), the probability of the decision selected by M and B ₁ is recorded as a vector x=(x ₁ , x ₂ ), y=(y ₁ , y ₂ ) ^T , then the solution of the mixed strategy Nash equilibrium is:

Equivalent to:

Satisfying (p ^* , q ^* ) in the payoff calculation in the game payoff matrix is the solution of a mixed-strategy Nash equilibrium combination of a non-cooperative game lane change. 9. The non-cooperative game lane-changing auxiliary decision-making method considering driving style characteristics according to claim 1, characterized in that, according to the solution obtained in step 3), the income level of judging the target vehicle lane-changing, when ε>1000, the safety income is high; otherwise, it is low; when δ > 800, the aging benefit is high, otherwise it is low; the rules for judging the benefit are as follows: When G>1000*α ₁ +800*α ₂ , the target vehicle lane-changing profit is high, otherwise it is low.

Images