Intelligent automobile stability control method based on driver characteristics

Abstract

The invention discloses an intelligent automobile stability control method based on driver characteristics. A vehicle state information acquisition module, a driver style identification module, a vehicle additional yaw moment calculation module and a moment distribution execution module are included. The vehicle state information acquisition module is used for acquiring various dynamic parameters such as vehicle slip rate, yaw velocity, roll angle, tire force and adhesion coefficient; the driver style identification module is combined with the control behavior factor, the environmental adaptability factor and the risk perception and decision factor to identify the driver style category based on a neural network model; the vehicle additional yaw moment calculation module calculates additional yaw moment for controlling the yaw velocity, the side slip angle and the transverse load transfer rate according to different styles; the torque distribution execution module judges an execution mode according to the vehicle state coefficient and calculates the output torque of each wheel.

Description

Technical Field

[0001] The invention belongs to the field of automobile safety, relates to the field of electric vehicle stability control, and specifically relates to an intelligent automobile stability control method based on driver characteristics. Background Art

[0002] With the continuous development of intelligent connected vehicles and autonomous driving technology, vehicle stability control on complex roads and extreme conditions has become an important research direction in vehicle safety technology. Traditional vehicle stability control systems are mainly based on vehicle dynamics models, which maintain the vehicle's driving stability by performing feedback control on parameters such as vehicle lateral acceleration, yaw rate, and tire force. However, such methods generally adopt a unified control logic, lack the ability to adapt to individual differences and operating characteristics of drivers, and are unable to adjust dynamic control strategies for different driving styles and behavior patterns, resulting in insufficient control accuracy and handling comfort.

[0003] In actual driving, the driver's control behavior, environmental adaptability, and risk perception and decision-making level directly affect the vehicle's dynamic response and safety performance. Drivers of different styles have significantly different characteristics in throttle, steering and brake operation, and the existing stability control system fails to fully integrate driver behavior information to optimize the control strategy. At the same time, under extreme driving conditions, the vehicle's demand for torque distribution and posture adjustment is highly dependent on the driver's behavior pattern, and traditional methods are difficult to achieve personalized application of additional yaw torque. Therefore, it is urgent to build an intelligent vehicle stability control method based on driver characteristics, which can integrate multi-source driving behavior feature information, adaptively adjust the vehicle control strategy, and improve the vehicle's handling stability and active safety performance. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent automobile stability control method based on driver characteristics to solve the problems faced in the above-mentioned background technology.

[0005] In order to achieve the above object, the present invention provides the following solution:

[0006] An intelligent vehicle stability control method based on driver characteristics, comprising a vehicle state information acquisition module, a driver style recognition module, a vehicle additional yaw moment calculation module, and a moment distribution execution module;

[0007] The vehicle status information acquisition module is used to obtain the vehicle's actual center of mass sideslip angle β, the vehicle's actual yaw angular velocity ω, and the vehicle's actual lateral acceleration a y , front wheel slip rate λ f , rear wheel slip rate λ r , vehicle longitudinal speed v x , front axle vertical load Fzf , rear axle vertical load F zr , body roll angle Road adhesion coefficient μ, vehicle wheelbase W, center of mass height h CG , lateral force F on the front wheel yf ; Lateral force F on the rear wheel yr 、The longitudinal force F on the vehicle x , the lateral force F on the vehicle y ;

[0008] The driver style recognition module is used to calculate the control behavior factor, environmental adaptability factor, risk perception and decision-making factor, so as to calculate the driver's comprehensive evaluation factor based on multiple evaluation factors and judge the driver's driving style;

[0009] The manipulation behavior factor S c The calculation formula is as follows:

[0010]

[0011] In the formula, is the average value of the steering wheel angular velocity; δ s,max is the maximum value of the steering wheel angular velocity; is the average value of the steering wheel angular acceleration; a s,max is the maximum value of the steering wheel angular acceleration; is the average value of the throttle opening change rate; ΔT a,max is the maximum value of the throttle opening change rate; is the average value of the brake pedal pressure change rate; ΔP b,max is the maximum value of the brake pedal pressure change rate; N acc N is the number of rapid accelerations per kilometer; acc,max N is the maximum number of rapid acceleration times per kilometer; brk N is the number of emergency brakes per kilometer; brk,max The maximum number of emergency brakes per kilometer;

[0012] The environmental adaptability factor S e The calculation formula is as follows:

[0013]

[0014] In the formula, is the average value of the following distance change rate; Δd f,max The maximum value of the following distance change rate; N l is the number of lane changes per kilometer; N l,max is the maximum number of lane changes per kilometer; is the average value of the speed adjustment on the curve; Δv c,maxis the maximum value of the speed adjustment range of the curve; t rg The traffic light response time; is the average value of traffic light response time; σ rg is the standard deviation of traffic light response time;

[0015] The risk perception and decision factor S r The calculation formula is as follows:

[0016]

[0017] In the formula, is the average value of the following distance; d f,max is the maximum value of the following distance; t b is the brake advance time; t b,max The maximum value of the brake advance time; R s R is the overspeed time ratio; s,max is the maximum value of the overspeed time ratio; R close is the triggering frequency of forward collision warning and lane departure warning; T dis Distract the driver's attention; dis,max The maximum time the driver's attention is distracted;

[0018] The driver style recognition module uses a fully connected neural network algorithm to determine the driver style category based on the driver's control behavior score, environmental adaptability score, risk perception and decision-making score. The specific structure design is as follows:

[0019] The input layer includes 3 nodes, corresponding to the control behavior factor, environmental adaptability factor, and risk perception and decision-making factor respectively; the hidden layer adopts a 3-layer structure, including the first hidden layer, the second hidden layer, and the third hidden layer; the first hidden layer, the second hidden layer, and the third hidden layer include 64, 32, and 16 neurons respectively; the output layer includes 1 node, and the output is the driver's style;

[0020] The calculation formula for the driver style judgment is:

[0021]

[0022] In the formula, S c , S e , S r They are the manipulation behavior factor, environmental adaptability factor, risk perception and decision-making factor; Z (1) ,Z (2) ,Z (3) ,Z (4) is the output of the first hidden layer, the second hidden layer, the third hidden layer and the output layer after linear transformation; W (1) ,W (2),W (3) ,W (4) are the weight matrices from the input layer to the first hidden layer, from the first hidden layer to the second hidden layer, from the second hidden layer to the third hidden layer, and from the third hidden layer to the output layer; b (1) ,b (2) ,b (3) ,b (4) are the bias items of the first hidden layer, the second hidden layer, the third hidden layer and the output layer respectively; ReLU is the activation function, which is defined as ReLU(x)=max(0,x) to enhance the nonlinear expression ability of the network; P(y=k|X) represents the probability that the input data X is classified as category k, and its calculation process is Softmax normalization to ensure that the output is a probability distribution. The three neurons correspond to the probabilities of radical, balanced and conservative types respectively; is the exponential transformation score of the kth category; The index sum of scores for all categories; It means that among all possible categories k, the k that maximizes the expression on the right is selected. A, B, and C represent the driver style as aggressive, balanced, and conservative respectively;

[0023] The vehicle additional yaw moment calculation module is used to receive the driver style information output by the driver style recognition module, and apply different additional yaw moments to conservative, balanced and aggressive drivers respectively, wherein the additional yaw moment is calculated by combining the additional yaw moment for controlling the yaw angular velocity, the additional yaw moment for controlling the sideslip angle of the center of mass and the additional yaw moment for controlling the lateral load transfer rate;

[0024] The torque distribution execution module calculates the vehicle driving state coefficient by considering the yaw instability factor and the roll instability factor, determines the vehicle execution mode according to the vehicle driving state coefficient, and calculates the torque of each wheel for the four-wheel independent drive mode and the four-wheel independent braking mode respectively.

[0025] The vehicle additional yaw moment calculation module is used to calculate the vehicle additional yaw moment according to the driver's style;

[0026] When the driver's style is conservative, the calculation formula of the vehicle additional yaw moment ΔM is:

[0027]

[0028] Where, β is the actual side slip angle of the vehicle's center of mass; β * is the ideal vehicle center of mass sideslip angle; ΔM ω The additional yaw moment to control the yaw rate; ΔM β is the additional yaw moment to control the sideslip angle of the center of mass; s β is the coefficient that controls the steepness of the transition;

[0029] When the driver's style is balanced, the calculation formula of the vehicle additional yaw moment ΔM is:

[0030]

[0031] Where, LTR is the actual lateral load transfer rate of the vehicle; LTR * is the ideal lateral load transfer rate of the vehicle; β is the actual side slip angle of the vehicle's center of mass; β * is the ideal vehicle center of mass sideslip angle; E β is the threshold of the vehicle's center of mass sideslip angle; ΔM ω The additional yaw moment to control the yaw rate; ΔM β The additional yaw moment to control the sideslip angle of the center of mass; ΔM LTR Additional yaw moment to control the lateral load transfer rate;

[0032] When the driver's style is aggressive, the calculation formula of the vehicle's additional yaw moment ΔM is:

[0033] ΔM=(1-lnS yaw )ΔM β +lnS lateral ΔM LTR

[0034] In the formula, ΔM β The additional yaw moment to control the sideslip angle of the center of mass; ΔM LTR The additional yaw moment for controlling the lateral load transfer rate; S yaw is the yaw instability factor; S lateral is the roll instability factor;

[0035] The yaw instability factor S yaw The calculation formula is as follows:

[0036]

[0037] Where, β is the actual side slip angle of the vehicle's center of mass; β * is the ideal sideslip angle of the vehicle's center of mass; ω is the actual yaw rate of the vehicle; ω * is the ideal yaw rate of the vehicle; v x is the longitudinal speed of the vehicle; μ is the road adhesion coefficient; is the roll angular velocity; LTR is the actual lateral load transfer rate of the vehicle; LTR * is the ideal lateral load transfer rate of the vehicle; F yf is the lateral force on the front wheel; F yr is the lateral force on the rear wheel;

[0038] The roll instability factor Slateral The calculation formula is as follows:

[0039]

[0040] In the formula, The maximum roll angle allowed for the vehicle body; is the body roll angle; LTR is the actual lateral load transfer rate of the vehicle; LTR * is the ideal lateral load transfer rate of the vehicle; v x is the longitudinal speed of the vehicle; for suspension roll damping; is the body roll inertia moment; is the suspension roll stiffness; F zf is the vertical load on the front axle; F zr is the vertical load on the rear axle.

[0041] The vehicle additional yaw moment calculation module controls the additional yaw moment ΔM of the yaw angular velocity ω The calculation formula is as follows:

[0042]

[0043] Where ω is the actual yaw rate of the vehicle; ω * is the desired yaw rate of the vehicle; δ f is the actual steering wheel angle of the vehicle; is the ideal steering wheel angle of the vehicle; F yf is the lateral force on the front wheel; F yr is the lateral force on the rear wheel; a y is the actual lateral acceleration of the vehicle; is the ideal lateral acceleration of the vehicle; LTR is the actual lateral load transfer rate of the vehicle;

[0044] The additional yaw moment ΔM of the control mass center sideslip angle β The calculation formula is as follows:

[0045]

[0046] Where, β is the actual side slip angle of the vehicle's center of mass; β * is the ideal sideslip angle of the vehicle's center of mass; a y is the actual lateral acceleration of the vehicle; is the ideal lateral acceleration of the vehicle; F yf is the lateral force on the front wheel; F yr is the lateral force on the rear wheel; v x is the longitudinal speed of the vehicle; LTR is the actual lateral load transfer rate of the vehicle;

[0047] The additional yaw moment ΔM that controls the lateral load transfer rate LTR The calculation formula is as follows:

[0048]

[0049] Where ω is the actual yaw rate of the vehicle; ω * is the desired yaw rate of the vehicle; LTR * is the ideal lateral load transfer rate of the vehicle; is the vehicle roll angular velocity; β is the vehicle's actual center of mass sideslip angle; β * is the ideal sideslip angle of the vehicle's center of mass; a y is the actual lateral acceleration of the vehicle; is the ideal lateral acceleration of the vehicle; v x is the longitudinal speed of the vehicle; δ f is the actual steering wheel angle of the vehicle; is the ideal steering wheel angle of the vehicle;

[0050] The calculation formula of the actual lateral load transfer rate LTR of the vehicle is as follows:

[0051]

[0052] In the formula, a y is the actual lateral acceleration of the vehicle; h CG is the height of the center of mass; W is the wheelbase of the vehicle; F zf is the vertical load on the front axle; F zr is the vertical load on the rear axle; is the body roll angle; is the roll angular velocity; is the roll angular acceleration; is the suspension roll stiffness; is the suspension roll damping; μ is the road adhesion coefficient; ω is the actual yaw angular velocity of the vehicle.

[0053] The torque distribution execution module is used to distribute the torque of the four wheels, and the vehicle execution mode is determined by the vehicle state driving coefficient;

[0054] The vehicle state driving coefficient calculation formula is as follows:

[0055]

[0056] In the formula, S yaw is the yaw instability factor; S lateral is the roll instability factor; v x F is the longitudinal speed of the vehicle; x is the longitudinal force on the vehicle; F y is the lateral force on the vehicle; a yis the actual lateral acceleration of the vehicle; is the actual yaw angular velocity of the vehicle; LTR is the actual lateral load transfer rate of the vehicle; h CG is the height of the center of mass; W is the wheelbase of the vehicle;

[0057] When the driving state driving coefficient 0 ≤ S driver When <1, the vehicle execution mode selects the four-wheel independent drive mode;

[0058] In the four-wheel independent drive mode, the calculation formula for the torque applied by each wheel is as follows:

[0059]

[0060]

[0061] Where, T fl , T fr , T rl , T rr are the driving torques of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively; T d is the total driving torque; f , r are the slip rates of the front and rear wheels respectively; F zf is the vertical load on the front axle; F zr is the vertical load on the rear axle; v x is the longitudinal speed of the vehicle; is the vehicle body roll angle; ω is the actual yaw rate of the vehicle; μ is the road adhesion coefficient; ΔM is the vehicle additional yaw moment; W is the vehicle wheelbase; β is the actual center of mass sideslip angle of the vehicle; LTR is the actual lateral load transfer rate of the vehicle; a y is the actual lateral acceleration of the vehicle; h CG is the height of the centroid;

[0062] When the driving state driving coefficient S driver When ≥1, the vehicle execution mode selects the four-wheel independent braking mode;

[0063] In the four-wheel independent braking mode, the calculation formula for the torque applied by each wheel is as follows:

[0064]

[0065] In the formula, B fl , B fr , B rl , B rr are the braking torques of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively; B d is the total braking torque; β is the actual side slip angle of the vehicle's center of mass; F x is the longitudinal force acting on the vehicle; is the roll angular velocity; ΔM is the additional yaw moment of the vehicle; W is the wheelbase of the vehicle; δ f is the front wheel turning angle; μ is the road adhesion coefficient; v x is the longitudinal speed of the vehicle; ω is the actual yaw rate of the vehicle; LTR is the actual lateral load transfer rate of the vehicle; a x is the longitudinal acceleration of the vehicle; F zf is the vertical load on the front axle; F zr is the vertical load on the rear axle; h CG is the centroid height.

[0066] The beneficial effects of the present invention are:

[0067] 1. The present invention can collect the vehicle dynamic state and the driver's control behavior, risk perception and environmental adaptation characteristics in real time, and use a multi-factor evaluation model and a neural network algorithm to identify the driving style, thereby realizing personalized vehicle stability control strategy selection and enhancing the system's adaptability to differences in driver behavior.

[0068] 2. The present invention constructs additional yaw moment models for three types of driving styles, and respectively adjusts key indicators such as yaw angular velocity, center of mass sideslip angle and lateral load transfer rate, so as to effectively improve the dynamic response accuracy and driving stability of the vehicle under different driving styles and avoid the degradation of control performance caused by a unified control strategy.

[0069] 3. The present invention further combines the vehicle state coefficient to intelligently select the four-wheel independent drive or four-wheel independent braking mode, and dynamically adjusts the output torque in combination with the torque distribution model of each wheel, thereby achieving refined control of the vehicle posture under extreme working conditions, improving the vehicle's yaw stability and roll suppression capabilities, and enhancing active safety performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] The present invention will be further described below in conjunction with the accompanying drawings:

[0071] Figure 1 This is a framework diagram of an intelligent vehicle stability control method based on driver characteristics proposed in the present invention. DETAILED DESCRIPTION

[0072] The present invention is further described in detail below based on the accompanying drawings and specific embodiments.

[0073] See also Figure 1 , the intelligent automobile stability control method based on driver characteristics described in the present invention comprises a vehicle state information acquisition module, a driver style recognition module, a vehicle additional yaw moment calculation module, and a moment distribution execution module;

[0074] The vehicle status information acquisition module is used to obtain the vehicle's actual center of mass sideslip angle β, the vehicle's actual yaw angular velocity ω, and the vehicle's actual lateral acceleration a y , front wheel slip rate λ f , rear wheel slip rate λ r , vehicle longitudinal speed v x , front axle vertical load F zf , rear axle vertical load F zr , body roll angle , road adhesion coefficient μ, vehicle wheelbase W, center of mass height h CG , lateral force F on the front wheel yf ; Lateral force F on the rear wheel yr 、The longitudinal force F on the vehicle x , the lateral force F on the vehicle y ;

[0075] The driver style recognition module is used to calculate the control behavior factor, environmental adaptability factor, risk perception and decision-making factor, so as to calculate the driver's comprehensive evaluation factor based on multiple evaluation factors and judge the driver's driving style;

[0076] The manipulation behavior factor S c The calculation formula is as follows:

[0077]

[0078] In the formula, is the average value of the steering wheel angular velocity; δ s,max is the maximum value of the steering wheel angular velocity; is the average value of the steering wheel angular acceleration; a s,max is the maximum value of the steering wheel angular acceleration; is the average value of the throttle opening change rate; ΔT a,max is the maximum value of the throttle opening change rate; is the average value of the brake pedal pressure change rate; ΔP b,max is the maximum value of the brake pedal pressure change rate; N acc N is the number of rapid accelerations per kilometer; acc,max N is the maximum number of rapid acceleration times per kilometer; brk N is the number of emergency brakes per kilometer; brk,max The maximum number of emergency brakes per kilometer;

[0079] The environmental adaptability factor S e The calculation formula is as follows:

[0080]

[0081] In the formula, is the average value of the following distance change rate; Δd f,max The maximum value of the following distance change rate; N l is the number of lane changes per kilometer; N l,max is the maximum number of lane changes per kilometer; is the average value of the speed adjustment on the curve; Δv c,max is the maximum value of the speed adjustment range of the curve; t rg The traffic light response time; is the average value of traffic light response time; σ rg is the standard deviation of traffic light response time;

[0082] The risk perception and decision factor S r The calculation formula is as follows:

[0083]

[0084] Where, d f is the average value of the following distance; d f,max is the maximum value of the following distance; t b is the brake advance time; t b,max The maximum value of the brake advance time; R s R is the overspeed time ratio; s,max is the maximum value of the overspeed time ratio; R close is the triggering frequency of forward collision warning and lane departure warning; T dis Distract the driver's attention; dis,max The maximum time the driver's attention is distracted;

[0085] The driver style recognition module uses a fully connected neural network algorithm to determine the driver style category based on the driver's control behavior score, environmental adaptability score, risk perception and decision-making score. The specific structure design is as follows:

[0086] The input layer includes 3 nodes, corresponding to the control behavior factor, environmental adaptability factor, and risk perception and decision-making factor respectively; the hidden layer adopts a 3-layer structure, including the first hidden layer, the second hidden layer, and the third hidden layer; the first hidden layer, the second hidden layer, and the third hidden layer include 64, 32, and 16 neurons respectively; the output layer includes 1 node, and the output is the driver's style;

[0087] The calculation formula for the driver style judgment is:

[0088]

[0089] In the formula, S c , S e , S rThey are the manipulation behavior factor, environmental adaptability factor, risk perception and decision-making factor; Z (1) ,Z (2) ,Z (3) ,Z (4) is the output of the first hidden layer, the second hidden layer, the third hidden layer and the output layer after linear transformation; W (1) ,W (2) ,W (3) ,W (4) are the weight matrices from the input layer to the first hidden layer, from the first hidden layer to the second hidden layer, from the second hidden layer to the third hidden layer, and from the third hidden layer to the output layer; b (1) ,b (2) ,b (3) ,b (4) are the bias items of the first hidden layer, the second hidden layer, the third hidden layer and the output layer respectively; ReLU is the activation function, which is defined as ReLU(x)=max(0,x) to enhance the nonlinear expression ability of the network; P(y=k|X) represents the probability that the input data X is classified as category k, and its calculation process is Softmax normalization to ensure that the output is a probability distribution. The three neurons correspond to the probabilities of radical, balanced and conservative types respectively; is the exponential transformation score of the kth category; The index sum of scores for all categories; It means that among all possible categories k, the k that maximizes the expression on the right is selected. A, B, and C represent the driver style as aggressive, balanced, and conservative respectively;

[0090] The vehicle additional yaw moment calculation module is used to receive the driver style information output by the driver style recognition module, and apply different additional yaw moments to conservative, balanced and aggressive drivers respectively, wherein the additional yaw moment is calculated by combining the additional yaw moment for controlling the yaw angular velocity, the additional yaw moment for controlling the sideslip angle of the center of mass and the additional yaw moment for controlling the lateral load transfer rate;

[0091] The torque distribution execution module calculates the vehicle driving state coefficient by considering the yaw instability factor and the roll instability factor, determines the vehicle execution mode according to the vehicle driving state coefficient, and calculates the torque of each wheel for the four-wheel independent drive mode and the four-wheel independent braking mode respectively.

[0092] The vehicle additional yaw moment calculation module is used to calculate the vehicle additional yaw moment according to the driver's style;

[0093] When the driver's style is conservative, the calculation formula of the vehicle additional yaw moment ΔM is:

[0094]

[0095] Where, β is the actual side slip angle of the vehicle's center of mass; β * is the ideal vehicle center of mass sideslip angle; ΔM ω The additional yaw moment to control the yaw rate; ΔM β is the additional yaw moment to control the sideslip angle of the center of mass; s β is the coefficient that controls the steepness of the transition;

[0096] When the driver's style is balanced, the calculation formula of the vehicle additional yaw moment ΔM is:

[0097]

[0098] Where, LTR is the actual lateral load transfer rate of the vehicle; LTR * is the ideal lateral load transfer rate of the vehicle; β is the actual side slip angle of the vehicle's center of mass; β * is the ideal vehicle center of mass sideslip angle; E β is the threshold of the vehicle's center of mass sideslip angle; ΔM ω The additional yaw moment to control the yaw rate; ΔM β The additional yaw moment to control the sideslip angle of the center of mass; ΔM LTR Additional yaw moment to control the lateral load transfer rate;

[0099] When the driver's style is aggressive, the calculation formula of the vehicle's additional yaw moment ΔM is:

[0100] ΔM=(1-lnS yaw )ΔM β +lnS lateral ΔM LTR

[0101] Where, ΔM β The additional yaw moment to control the sideslip angle of the center of mass; ΔM LTR The additional yaw moment for controlling the lateral load transfer rate; S yaw is the yaw instability factor; S lateral is the roll instability factor;

[0102] The yaw instability factor S yaw The calculation formula is as follows:

[0103]

[0104] Where, β is the actual side slip angle of the vehicle's center of mass; β * is the ideal sideslip angle of the vehicle's center of mass; ω is the actual yaw rate of the vehicle; ω * is the ideal yaw rate of the vehicle; v x is the longitudinal speed of the vehicle; μ is the road adhesion coefficient; is the roll angular velocity; LTR is the actual lateral load transfer rate of the vehicle; LTR * is the ideal lateral load transfer rate of the vehicle; F yf is the lateral force on the front wheel; F yr is the lateral force on the rear wheel;

[0105] The roll instability factor S lateral The calculation formula is as follows:

[0106]

[0107] In the formula, The maximum roll angle allowed for the vehicle body; is the body roll angle; LTR is the actual lateral load transfer rate of the vehicle; LTR * is the ideal lateral load transfer rate of the vehicle; v x is the longitudinal speed of the vehicle; for suspension roll damping; is the body roll inertia moment; is the suspension roll stiffness; F zf is the vertical load on the front axle; F zr is the vertical load on the rear axle.

[0108] The vehicle additional yaw moment calculation module controls the additional yaw moment ΔM of the yaw angular velocity ω The calculation formula is as follows:

[0109]

[0110] Where ω is the actual yaw rate of the vehicle; ω * is the desired yaw rate of the vehicle; δ f is the actual steering wheel angle of the vehicle; δ f * is the ideal steering wheel angle of the vehicle; F yf is the lateral force on the front wheel; F yr is the lateral force on the rear wheel; a y is the actual lateral acceleration of the vehicle; a y * is the ideal lateral acceleration of the vehicle; LTR is the actual lateral load transfer rate of the vehicle;

[0111] The additional yaw moment ΔM of the control mass center sideslip angle β The calculation formula is as follows:

[0112]

[0113] Where, β is the actual side slip angle of the vehicle's center of mass; β * is the ideal sideslip angle of the vehicle's center of mass; a yis the actual lateral acceleration of the vehicle; is the ideal lateral acceleration of the vehicle; F yf is the lateral force on the front wheel; F yr is the lateral force on the rear wheel; v x is the longitudinal speed of the vehicle; LTR is the actual lateral load transfer rate of the vehicle;

[0114] The additional yaw moment ΔM that controls the lateral load transfer rate LTR The calculation formula is as follows:

[0115]

[0116] Where ω is the actual yaw rate of the vehicle; ω * is the desired yaw rate of the vehicle; LTR * is the ideal lateral load transfer rate of the vehicle; is the vehicle roll angular velocity; β is the vehicle's actual center of mass sideslip angle; β * is the ideal sideslip angle of the vehicle's center of mass; a y is the actual lateral acceleration of the vehicle; is the ideal lateral acceleration of the vehicle; v x is the longitudinal speed of the vehicle; f is the actual steering wheel angle of the vehicle; is the ideal steering wheel angle of the vehicle;

[0117] The calculation formula of the actual lateral load transfer rate LTR of the vehicle is as follows:

[0118]

[0119] In the formula, a y is the actual lateral acceleration of the vehicle; h CG is the height of the center of mass; W is the wheelbase of the vehicle; F zf is the vertical load on the front axle; F zr is the vertical load on the rear axle; is the body roll angle; is the roll angular velocity; is the roll angular acceleration; is the suspension roll stiffness; is the suspension roll damping; μ is the road adhesion coefficient; ω is the actual yaw angular velocity of the vehicle.

[0120] The torque distribution execution module is used to distribute the torque of the four wheels, and the vehicle execution mode is determined by the vehicle state driving coefficient;

[0121] The vehicle state driving coefficient calculation formula is as follows:

[0122]

[0123] In the formula, S yaw is the yaw instability factor; S lateral is the roll instability factor; v x F is the longitudinal speed of the vehicle; x is the longitudinal force on the vehicle; F y is the lateral force on the vehicle; a y is the actual lateral acceleration of the vehicle; is the actual yaw angular velocity of the vehicle; LTR is the actual lateral load transfer rate of the vehicle; h CG is the height of the center of mass; W is the wheelbase of the vehicle;

[0124] When the driving state driving coefficient 0 ≤ S driver When <1, the vehicle execution mode selects the four-wheel independent drive mode;

[0125] In the four-wheel independent drive mode, the calculation formula for the torque applied by each wheel is as follows:

[0126]

[0127] Where, T fl , T fr , T rl , T rr are the driving torques of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively; T d is the total driving torque; f , r are the slip rates of the front and rear wheels respectively; F zf is the vertical load on the front axle; F zr is the vertical load on the rear axle; v x is the longitudinal speed of the vehicle; is the vehicle body roll angle; ω is the actual yaw rate of the vehicle; μ is the road adhesion coefficient; ΔM is the vehicle additional yaw moment; W is the vehicle wheelbase; β is the actual center of mass sideslip angle of the vehicle; LTR is the actual lateral load transfer rate of the vehicle; a y is the actual lateral acceleration of the vehicle; h CG is the height of the centroid;

[0128] When the driving state driving coefficient S driver When ≥1, the vehicle execution mode selects the four-wheel independent braking mode;

[0129] In the four-wheel independent braking mode, the calculation formula for the torque applied by each wheel is as follows:

[0130]

[0131]

[0132] In the formula, B fl , B fr , B rl , B rr are the braking torques of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively; B d is the total braking torque; β is the actual side slip angle of the vehicle's center of mass; F x is the longitudinal force acting on the vehicle; is the roll angular velocity; ΔM is the additional yaw moment of the vehicle; W is the wheelbase of the vehicle; δ f is the front wheel turning angle; μ is the road adhesion coefficient; v x is the longitudinal speed of the vehicle; ω is the actual yaw rate of the vehicle; LTR is the actual lateral load transfer rate of the vehicle; a x is the longitudinal acceleration of the vehicle; F zf is the vertical load on the front axle; F zr is the vertical load on the rear axle; h CG is the centroid height.

Claims

1. An intelligent vehicle stability control method based on driver characteristics, characterized in that: It includes a vehicle status information acquisition module, a driver style recognition module, a vehicle additional yaw moment calculation module, and a moment distribution execution module; The vehicle status information acquisition module is used to obtain the vehicle's actual center of mass sideslip angle β, the vehicle's actual yaw angular velocity ω, and the vehicle's actual lateral acceleration a y , front wheel slip rate λ f , rear wheel slip rate λ r , vehicle longitudinal speed v x , front axle vertical load F zf , rear axle vertical load F zr , body roll angle Road adhesion coefficient μ, vehicle wheelbase W, center of mass height h CG , lateral force F on the front wheel yf ; Lateral force F on the rear wheel yr 、The longitudinal force F on the vehicle x , the lateral force F on the vehicle y ; The driver style recognition module is used to calculate the control behavior factor, environmental adaptability factor, risk perception and decision-making factor, so as to calculate the driver's comprehensive evaluation factor based on multiple evaluation factors and judge the driver's driving style; The manipulation behavior factor S c The calculation formula is as follows: In the formula, is the average value of the steering wheel angular velocity; δ s,max is the maximum value of the steering wheel angular velocity; is the average value of the steering wheel angular acceleration; a s,max is the maximum value of the steering wheel angular acceleration; is the average value of the throttle opening change rate; ΔT a,max is the maximum value of the throttle opening change rate; is the average value of the brake pedal pressure change rate; ΔP b,max is the maximum value of the brake pedal pressure change rate; N acc N is the number of rapid accelerations per kilometer; acc,max N is the maximum number of rapid acceleration times per kilometer; brk N is the number of emergency brakes per kilometer; brk,max The maximum number of emergency brakes per kilometer; The environmental adaptability factor S e The calculation formula is as follows: In the formula, is the average value of the following distance change rate; Δd f,max The maximum value of the following distance change rate; N l is the number of lane changes per kilometer; N l,max is the maximum number of lane changes per kilometer; is the average value of the speed adjustment on the curve; Δv c,max is the maximum value of the speed adjustment range on the curve; t rg The traffic light response time; is the average value of traffic light response time; σ rg is the standard deviation of traffic light response time; The risk perception and decision factor S r The calculation formula is as follows: In the formula, is the average value of the following distance; d f,max is the maximum value of the following distance; t b Provides time for braking; t b,max The maximum value of the brake advance time; R s R is the overspeed time ratio; s,max is the maximum value of the overspeed time ratio; R close is the triggering frequency of forward collision warning and lane departure warning; T dis Provide distraction time for the driver; T dis,max The maximum time the driver's attention is distracted; The driver style recognition module uses a fully connected neural network algorithm to determine the driver style category based on the driver's control behavior score, environmental adaptability score, risk perception and decision-making score. The specific structure design is as follows: The input layer includes 3 nodes, corresponding to the control behavior factor, environmental adaptability factor, and risk perception and decision-making factor respectively; the hidden layer adopts a 3-layer structure, including the first hidden layer, the second hidden layer, and the third hidden layer; the first hidden layer, the second hidden layer, and the third hidden layer include 64, 32, and 16 neurons respectively; the output layer includes 1 node, and the output is the driver's style; The calculation formula for the driver style judgment is: In the formula, S c , S e , S r They are the manipulation behavior factor, environmental adaptability factor, risk perception and decision-making factor; Z (1) ,Z (2) ,Z (3) ,Z (4) is the output of the first hidden layer, the second hidden layer, the third hidden layer and the output layer after linear transformation; W (1) ,W (2) ,W (3) ,W (4) are the weight matrices from the input layer to the first hidden layer, from the first hidden layer to the second hidden layer, from the second hidden layer to the third hidden layer, and from the third hidden layer to the output layer; b (1) ,b (2) ,b (3) ,b (4) are the bias items of the first hidden layer, the second hidden layer, the third hidden layer and the output layer respectively; ReLU is the activation function, which is defined as ReLU(x)=max(0,x) to enhance the nonlinear expression ability of the network; P(y=k|X) represents the probability that the input data X is classified as category k, and its calculation process is Softmax normalization to ensure that the output is a probability distribution. The three neurons correspond to the radical, balanced and conservative types respectively. The probability of type; is the exponential transformation score of the kth category; The index sum of scores for all categories; It means that among all possible categories k, the k that maximizes the expression on the right is selected. A, B, and C represent the driver style as aggressive, balanced, and conservative respectively; The vehicle additional yaw moment calculation module is used to receive the driver style information output by the driver style recognition module, and apply different additional yaw moments to conservative, balanced and aggressive drivers respectively, wherein the additional yaw moment is calculated by combining the additional yaw moment for controlling the yaw angular velocity, the additional yaw moment for controlling the sideslip angle of the center of mass and the additional yaw moment for controlling the lateral load transfer rate; The torque distribution execution module calculates the vehicle driving state coefficient by considering the yaw instability factor and the roll instability factor, determines the vehicle execution mode according to the vehicle driving state coefficient, and calculates the torque of each wheel for the four-wheel independent drive mode and the four-wheel independent braking mode respectively.

2. The intelligent vehicle stability control method based on driver characteristics according to claim 1, characterized in that: The vehicle additional yaw moment calculation module is used to calculate the vehicle additional yaw moment according to the driver's style; When the driver's style is conservative, the calculation formula of the vehicle additional yaw moment ΔM is: Where, β is the actual center of mass sideslip angle of the vehicle; β * is the ideal vehicle center of mass sideslip angle; ΔM ω The additional yaw moment to control the yaw rate; ΔM β is the additional yaw moment to control the sideslip angle of the center of mass; s β is the coefficient that controls the steepness of the transition; When the driver's style is balanced, the calculation formula of the vehicle additional yaw moment ΔM is: Where, LTR is the actual lateral load transfer rate of the vehicle; LTR * is the ideal lateral load transfer rate of the vehicle; β is the actual side slip angle of the vehicle's center of mass; β * is the ideal vehicle center of mass sideslip angle; E β is the threshold of the vehicle's center of mass sideslip angle; ΔM ω The additional yaw moment to control the yaw rate; ΔM β The additional yaw moment to control the sideslip angle of the center of mass; ΔM LTR Additional yaw moment to control the lateral load transfer rate; When the driver's style is aggressive, the calculation formula of the vehicle's additional yaw moment ΔM is: ΔM=(1-lnS yaw )ΔM β +lnS lateral ΔM LTR Where, ΔM β The additional yaw moment to control the sideslip angle of the center of mass; ΔM LTR The additional yaw moment for controlling the lateral load transfer rate; S yaw is the yaw instability factor; S lateral is the roll instability factor; The yaw instability factor S yaw The calculation formula is as follows: Where, β is the actual center of mass sideslip angle of the vehicle; β * is the ideal sideslip angle of the vehicle's center of mass; ω is the actual yaw rate of the vehicle; ω * is the ideal yaw rate of the vehicle; v x is the longitudinal speed of the vehicle; μ is the road adhesion coefficient; is the roll angular velocity; LTR is the actual lateral load transfer rate of the vehicle; LTR * is the ideal lateral load transfer rate of the vehicle; F yf is the lateral force on the front wheel; F yr is the lateral force on the rear wheel; The roll instability factor S lateral The calculation formula is as follows: In the formula, The maximum roll angle allowed for the vehicle body; is the body roll angle; LTR is the actual lateral load transfer rate of the vehicle; LTR * is the ideal lateral load transfer rate of the vehicle; v x is the longitudinal speed of the vehicle; for suspension roll damping; is the body roll inertia moment; is the suspension roll stiffness; F zf is the vertical load on the front axle; F zr is the vertical load on the rear axle.

3. The intelligent vehicle stability control method based on driver characteristics according to claim 1, characterized in that: The vehicle additional yaw moment calculation module controls the additional yaw moment ΔM of the yaw angular velocity ω The calculation formula is as follows: Where ω is the actual yaw rate of the vehicle; ω * is the desired yaw rate of the vehicle; δ f is the actual steering wheel angle of the vehicle; is the ideal steering wheel angle of the vehicle; F yf is the lateral force on the front wheel; F yr is the lateral force on the rear wheel; a y is the actual lateral acceleration of the vehicle; is the ideal lateral acceleration of the vehicle; LTR is the actual lateral load transfer rate of the vehicle; The additional yaw moment ΔM of the control mass center sideslip angle β The calculation formula is as follows: Where, β is the actual center of mass sideslip angle of the vehicle; β * is the ideal sideslip angle of the vehicle's center of mass; a y is the actual lateral acceleration of the vehicle; is the ideal lateral acceleration of the vehicle; F yf is the lateral force on the front wheel; F yr is the lateral force on the rear wheel; v x is the longitudinal speed of the vehicle; LTR is the actual lateral load transfer rate of the vehicle; The additional yaw moment ΔM that controls the lateral load transfer rate LTR The calculation formula is as follows: Where ω is the actual yaw rate of the vehicle; ω * is the desired yaw rate of the vehicle; LTR * is the ideal lateral load transfer rate of the vehicle; is the vehicle roll angular velocity; β is the actual vehicle center of mass sideslip angle; β * is the ideal sideslip angle of the vehicle's center of mass; a y is the actual lateral acceleration of the vehicle; is the ideal lateral acceleration of the vehicle; v x is the longitudinal speed of the vehicle; δ f is the actual steering wheel angle of the vehicle; is the ideal steering wheel angle of the vehicle; The calculation formula of the actual lateral load transfer rate LTR of the vehicle is as follows: In the formula, a y is the actual lateral acceleration of the vehicle; h CG is the height of the center of mass; W is the wheelbase of the vehicle; F zf is the vertical load on the front axle; F zr is the vertical load on the rear axle; is the body roll angle; is the roll angular velocity; is the roll angular acceleration; is the suspension roll stiffness; is the suspension roll damping; μ is the road adhesion coefficient; ω is the actual yaw angular velocity of the vehicle.

4. The intelligent vehicle stability control method based on driver characteristics according to claim 1, characterized in that: The torque distribution execution module is used to distribute the torque of the four wheels, and the vehicle execution mode is determined by the vehicle state driving coefficient; The vehicle state driving coefficient calculation formula is as follows: In the formula, S yaw is the yaw instability factor; S lateral is the roll instability factor; v x F is the longitudinal speed of the vehicle; x is the longitudinal force on the vehicle; F y is the lateral force on the vehicle; a y is the actual lateral acceleration of the vehicle; is the actual yaw angular velocity of the vehicle; LTR is the actual lateral load transfer rate of the vehicle; h CG is the height of the center of mass; W is the wheelbase of the vehicle; When the driving state driving coefficient 0 ≤ S driver When <1, the vehicle execution mode selects the four-wheel independent drive mode; In the four-wheel independent drive mode, the calculation formula for the torque applied by each wheel is as follows: Where, T fl , T fr , T rl , T rr are the driving torques of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively; T d is the total driving torque; f , r are the slip rates of the front and rear wheels respectively; F zf is the vertical load on the front axle; F zr is the vertical load on the rear axle; v x is the longitudinal speed of the vehicle; is the vehicle body roll angle; ω is the actual yaw rate of the vehicle; μ is the road adhesion coefficient; ΔM is the vehicle additional yaw moment; W is the vehicle wheelbase; β is the actual center of mass sideslip angle of the vehicle; LTR is the actual lateral load transfer rate of the vehicle; a y is the actual lateral acceleration of the vehicle; h CG is the height of the centroid; When the driving state driving coefficient S driver When ≥1, the vehicle execution mode selects the four-wheel independent braking mode; In the four-wheel independent braking mode, the calculation formula for the torque applied by each wheel is as follows: In the formula, B fl , B fr , B rl , B rr are the braking torques of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively; B d is the total braking torque; β is the actual side slip angle of the vehicle's center of mass; F x is the longitudinal force acting on the vehicle; is the roll angular velocity; ΔM is the additional yaw moment of the vehicle; W is the wheelbase of the vehicle; δ f is the front wheel turning angle; μ is the road adhesion coefficient; v x is the longitudinal speed of the vehicle; ω is the actual yaw rate of the vehicle; LTR is the actual lateral load transfer rate of the vehicle; a x is the longitudinal acceleration of the vehicle; F zf is the vertical load on the front axle; F zr is the vertical load on the rear axle; h CG is the centroid height.

Images