A coordinated AFWS / DYC lateral stability control method considering driving style

Abstract

This application relates to an AFWS / DYC lateral stability coordinated control method that takes driving style into consideration. This application obtains the driver's driving style through a constructed online driving style identification model, obtains a reference yaw rate based on the driver's driving style, and determines an instability index based on the reference yaw rate and the ideal center of mass sideslip angle. A cooperative game and a non-cooperative game are performed based on the reference yaw rate, the ideal center of mass sideslip angle, and the instability index. The final direct yaw torque affects the preliminary front wheel angle increment and the preliminary rear wheel angle increment to obtain the final front wheel angle increment and the final rear wheel angle increment. This application applies driving style to the coordinated control of vehicle lateral stability and integrates cooperative and non-cooperative games, enabling vehicle lateral stability control to meet the personalized needs of different drivers.

Description

Technical Field

[0001] The present application relates to the field of vehicle control technology, and in particular to an AFWS / DYC lateral stability coordinated control method considering driving style. Background Art

[0002] Vehicle safety has always been a major concern for road safety authorities worldwide. When a vehicle travels on low-adhesion surfaces, lateral forces can easily saturate, leading to skidding and even loss of control. Vehicle lateral stability control improves vehicle stability and safety by controlling the vehicle to track an ideal yaw rate and sideslip angle. Currently, vehicle lateral dynamics control primarily focuses on stability control under extreme conditions. However, extreme instability is uncommon under normal driving conditions.

[0003] At the same time, in the study of modern vehicle dynamics, researchers are increasingly focusing on the differences in needs between individual drivers, designing tailored control strategies to enhance the driving experience for each driver. The focus of optimizing the human-vehicle relationship has shifted from meeting the general needs of ordinary drivers to addressing the specific needs of each individual driver. Controlling the vehicle only under extreme conditions not only wastes dynamic performance but also makes it difficult for the overall vehicle dynamics to meet the individual needs of different drivers. There is also a lack of personalized lateral stability control strategies that take driving style into account. Summary of the Invention

[0004] In order to overcome at least one deficiency in the prior art, the present application provides an AFWS / DYC lateral stability coordinated control method considering driving style.

[0005] In a first aspect, a method for coordinated lateral stability control of AFWS / DYC considering driving style is provided, comprising:

[0006] Build an online driving style identification model;

[0007] Obtain driver's online lane-changing data and extract characteristic indicators;

[0008] Input the characteristic index into the online driving style identification model to obtain the driver's driving style;

[0009] determining a vehicle stability factor according to a driver's driving style; determining a driving style gain coefficient according to the vehicle stability factor; and obtaining a reference yaw angular velocity according to the driving style gain coefficient;

[0010] Determine the instability index based on the reference yaw rate and the ideal center of mass sideslip angle;

[0011] Conducting cooperative and non-cooperative games based on a reference yaw rate, an ideal sideslip angle at the center of mass, and an instability index to obtain a preliminary front wheel angle increment, a preliminary rear wheel angle increment, and a final direct yaw moment; utilizing the final direct yaw moment to influence the preliminary front wheel angle increment and the preliminary rear wheel angle increment to obtain a final front wheel angle increment and a final rear wheel angle increment;

[0012] The final front wheel steering angle increment and the final rear wheel steering angle increment are applied to the steering wheel, and the final direct yaw moment is converted into tire longitudinal force and distributed to each wheel.

[0013] In one embodiment, building an online driving style identification model includes:

[0014] Obtain lane-changing process data from multiple drivers, and extract feature indicators from each driver's lane-changing process data as training samples;

[0015] Perform cluster analysis on all training samples to determine the driving style category of each training sample;

[0016] The training samples and the corresponding driving style categories constitute the driving style dataset;

[0017] The GRNN is trained based on the driving style dataset to obtain an online driving style identification model.

[0018] In one embodiment, obtaining the driver's online lane change data and extracting characteristic indicators include:

[0019] Step S21: Obtain the distance change Δdis between the vehicle and the lane line at the current time t0 lane , if Δdis lane >3, then go to step S22, otherwise, repeat step S21 at the next moment;

[0020] Step S22: Calculate the distance dis between the vehicle and the obstacle at time t0-5 front , if 40>dis front >0, then go to step S23, otherwise go to step S21;

[0021] Step S23, determining the lane change start time and end time, includes:

[0022] If str sw (t)<10° and str sw (t+0.5)>15°, t represents the time, t∈[t0-5,t0+5], str sw (t) is the steering wheel angle at time t, then the first time t1 = t;

[0023] If strsw (t)<-15° and str sw (t+0.5)>-10°, then the second moment t2=t;

[0024] Start time t start =min(t1,t2), end time t end =mac(t1,t2);

[0025] Step S24, extracting the vehicle driving data between the lane change start time and the lane change end time, i.e., the driver's online lane change data;

[0026] Step S25 , extracting characteristic indicators based on the driver's online lane change data, the characteristic indicators including speed standard deviation, maximum absolute value of lateral acceleration, standard deviation of lateral acceleration, average absolute value of steering wheel angle, and average absolute value of steering wheel angular velocity.

[0027] In one embodiment, the reference yaw rate is:

[0028]

[0029] in, is the reference yaw rate, Gain is the driving style gain coefficient, ω r_d is the ideal yaw rate.

[0030] In one embodiment, determining an instability index based on a reference yaw rate and an ideal center of mass sideslip angle includes:

[0031] The β-ω phase plane is divided into three regions: stable region I, transition region II, and unstable region III.

[0032] Determine the area to which the vehicle belongs based on the reference yaw rate and the ideal center of mass sideslip angle;

[0033] According to the area to which the vehicle belongs, the following formula is used to determine the instability index:

[0034]

[0035] Among them, S ta is the instability index, Δγ r is the deviation between the actual yaw rate and the ideal yaw rate, Δγ max is the maximum deviation between the actual yaw rate and the ideal yaw rate, Δβ r is the deviation between the actual center of mass sideslip angle and the ideal center of mass sideslip angle, Δβ max It is the maximum deviation between the actual sideslip angle and the ideal sideslip angle.

[0036] In one embodiment, the resulting direct yaw moment is:

[0037]

[0038] in, is the final direct yaw moment, k is the sampling time, L1 is the coefficient matrix of the direct yaw moment, and θ1 is the variable; for The amount of decomposition, is the weight matrix, T stands for transpose; for The amount of decomposition, is the weight matrix, Ω is the coefficient matrix; is the predicted stability control target; Λ is the coefficient matrix, X(k) is the matrix variable; Υ is the coefficient matrix.

[0039] In one embodiment, the weight matrix for:

[0040]

[0041] r1=10 -(5+4LA)

[0042]

[0043] Among them, r1 is the control output weight of the direct yaw moment, LA is the parameter, S ta is the instability index, and N represents the prediction time domain.

[0044] In one embodiment, the final front wheel steering angle increment and the final rear wheel steering angle increment are:

[0045]

[0046] in, is the final front wheel angle increment, is the final rear wheel steering angle increment, K1 and K2 are feedback gains, is the inverse of K2, E is the coefficient matrix, x(k) is the state variable, L2 is the coefficient matrix of the front wheel steering angle increment, L3 is the coefficient matrix of the rear wheel steering angle increment, is the coefficient matrix of the prediction equation, L1 is the coefficient matrix of the direct yaw moment, θ1 is the variable, and Z is the coefficient matrix.

[0047] In a second aspect, an AFWS / DYC lateral stability coordinated control device considering driving style is provided, comprising:

[0048] Model building module, used to build an online driving style identification model;

[0049] A characteristic index acquisition module is used to obtain the driver's online lane change data and extract characteristic indicators;

[0050] An identification module is used to input characteristic indicators into an online driving style identification model to obtain the driver's driving style;

[0051] A reference yaw rate acquisition module is configured to determine a vehicle stability factor based on the driver's driving style; determine a driving style gain coefficient based on the vehicle stability factor; and obtain a reference yaw rate based on the driving style gain coefficient;

[0052] An instability index determination module is used to determine an instability index based on a reference yaw rate and an ideal center of mass sideslip angle;

[0053] a control variable acquisition module for performing cooperative and non-cooperative games based on a reference yaw rate, an ideal sideslip angle at the center of mass, and an instability index to obtain a preliminary front wheel angle increment, a preliminary rear wheel angle increment, and a final direct yaw moment; and utilizing the final direct yaw moment to influence the preliminary front wheel angle increment and the preliminary rear wheel angle increment to obtain a final front wheel angle increment and a final rear wheel angle increment;

[0054] The control module is used to apply the final front wheel steering angle increment and the final rear wheel steering angle increment to the steering wheel, and convert the final direct yaw moment into tire longitudinal force and distribute it to each wheel.

[0055] In a third aspect, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the above-mentioned AFWS / DYC lateral stability coordinated control method considering driving style is implemented.

[0056] Compared to existing technologies, the present application has the following advantages: It acquires the driver's driving style through an online driving style identification model, obtains a reference yaw rate based on the driver's driving style, and determines an instability index based on the reference yaw rate and the ideal center of mass slip angle. It also conducts cooperative and non-cooperative games based on the reference yaw rate, the ideal center of mass slip angle, and the instability index, and uses the final direct yaw torque to influence the preliminary front wheel angle increment and the preliminary rear wheel angle increment to obtain the final front wheel angle increment and the final rear wheel angle increment. The present application applies driving style to the coordinated control of vehicle lateral stability and integrates cooperative and non-cooperative games, enabling vehicle lateral stability control to meet the personalized needs of different drivers. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The present application may be better understood by referring to the following description in conjunction with the accompanying drawings, which together with the following detailed description are incorporated into and form a part of this specification. In the drawings:

[0058] Figure 1 A flowchart of the AFWS / DYC lateral stability coordinated control method considering driving style is shown;

[0059] Figure 2 The figure shows the setup of the driver's test road in the loop, (a) is a small-turn road, and (b) is a large-turn road;

[0060] Figure 3 The vehicle stability factor distribution diagrams corresponding to various driving styles are shown, where (a) is the vehicle stability factor distribution diagram corresponding to the cautious type, (b) is the vehicle stability factor distribution diagram corresponding to the general type, and (c) is the vehicle stability factor distribution diagram corresponding to the aggressive type;

[0061] Figure 4 A schematic diagram of the β-ω phase plane division is shown;

[0062] Figure 5 A schematic diagram of the phase plane transition domain division is shown;

[0063] Figure 6 The block diagram of the personalized AFWS / DYC lateral stability controller based on game theory is shown;

[0064] Figure 7 The established linear two-degree-of-freedom vehicle model of a four-wheel steering vehicle is shown;

[0065] Figure 8 The structural block diagram of the AFWS / DYC lateral stability coordinated control device considering driving style is shown. DETAILED DESCRIPTION

[0066] Exemplary embodiments of the present application are described below with reference to the accompanying drawings. For the sake of clarity and conciseness, not all features of actual embodiments are described in this specification. However, it should be understood that in the process of developing any such actual embodiment, many implementation-specific decisions may be made to achieve the developer's specific goals, and these decisions may vary from one implementation to another.

[0067] It is also necessary to explain here that, in order to avoid obscuring the present application due to unnecessary details, the accompanying drawings only show the device structure closely related to the solution according to the present application, while other details that are not closely related to the present application are omitted.

[0068] It should be understood that the present application is not limited to the described embodiments due to the following description with reference to the accompanying drawings. In this document, where feasible, the embodiments may be combined with each other, features between different embodiments may be replaced or borrowed, and one or more features may be omitted in one embodiment.

[0069] The present invention provides an AFWS / DYC lateral stability coordinated control method considering driving style. Figure 1 The flowchart of the AFWS / DYC lateral stability coordinated control method considering driving style is shown in FIG. Figure 1 , methods include:

[0070] Step S1: constructing an online driving style recognition model.

[0071] Step S2: Obtain the driver's online lane-changing data and extract characteristic indicators.

[0072] Step S3: input the characteristic index into the driving style online identification model to obtain the driver's driving style.

[0073] Step S4: determining a vehicle stability factor according to the driver's driving style; determining a driving style gain coefficient according to the vehicle stability factor; and obtaining a reference yaw rate according to the driving style gain coefficient.

[0074] Step S5: determining an instability index based on a reference yaw rate and an ideal sideslip angle of the center of mass.

[0075] Step S6: performing cooperative and non-cooperative games based on the reference yaw rate, the ideal sideslip angle, and the instability index to obtain a preliminary front wheel angle increment, a preliminary rear wheel angle increment, and a final direct yaw moment (DYS); utilizing the final direct yaw moment to influence the preliminary front wheel angle increment and the preliminary rear wheel angle increment to obtain a final front wheel angle increment and a final rear wheel angle increment (AFWS);

[0076] Step S7: Apply the final front wheel steering angle increment and the final rear wheel steering angle increment to the steering wheel, and convert the final direct yaw moment into tire longitudinal force and distribute it to each wheel.

[0077] In this embodiment, the driver's driving style is acquired through a constructed online driving style identification model. A reference yaw rate is derived based on the driver's driving style, and an instability index is determined based on the reference yaw rate and the ideal center of mass slip angle. A cooperative game and a non-cooperative game are then performed based on the reference yaw rate, the ideal center of mass slip angle, and the instability index. The final direct yaw torque influences the preliminary front wheel angle increment and the preliminary rear wheel angle increment to obtain the final front wheel angle increment and the final rear wheel angle increment. This embodiment applies driving style to the coordinated control of vehicle lateral stability and integrates cooperative and non-cooperative game strategies, enabling vehicle lateral stability control to meet the personalized needs of different drivers.

[0078] In one embodiment, step S1, constructing an online driving style identification model, includes:

[0079] Step S11: Acquire lane-changing process data of multiple drivers, and extract feature indicators from the lane-changing process data of each driver as training samples.

[0080] A driving style data acquisition platform was built using Matlab / Simulink, Prescan, and the Logitech G29 driving simulator. Prescan was used to establish standardized vehicle handling stability test scenarios, and traffic scenario infrastructure components were established through a GUI interface. The vehicle model used a 10-degree-of-freedom vehicle dynamics model encompassing longitudinal, lateral, vertical, yaw, roll, pitch, and tire motion. Twenty-seven drivers (24 male and 3 female) were selected for a driving style data acquisition experiment. Data on vehicle speed, lateral acceleration, steering wheel angle, and steering wheel angular velocity were collected. Data from all lane change processes were extracted, and driving style characteristic indicators were calculated, including: X1 speed standard deviation, X2 lateral acceleration absolute maximum value, X3 lateral acceleration standard deviation, X4 steering wheel angle absolute average value, and X5 steering wheel angular velocity absolute average value. The characteristic index values ​​at the same time are combined as a single data sample x = (X1, X2, X3, X4, X5), and finally all data samples are constructed into a data set D = (x1, x2, x3, ..., xn), where n is the number of samples in the data set.

[0081] Step S12: performing cluster analysis on all training samples to determine the driving style category of each training sample.

[0082] Here, we can use the K-means algorithm to perform cluster analysis on vehicle driving status. This clustering allows us to categorize drivers into three groups, with numbers representing different driving styles. Number 1 represents a cautious driver, number 2 represents a normal driver, and number 3 represents an aggressive driver. This yields a dataset of driver style categories.

[0083] Step S13: The training samples and the corresponding driving style categories constitute a driving style dataset; the sample dataset and the driving style category dataset are merged into a driving style dataset D ds , D ds =[D,D driver ].

[0084] In step S14 , a GRNN (Generalized Regression Neural Network) is trained based on the driving style dataset to obtain an online driving style recognition model.

[0085] Here, the driving style dataset D ds The model training set is used as the input for the model training set. The standard deviation of velocity (X1), the maximum absolute value of lateral acceleration (X2), the standard deviation of lateral acceleration (X3), the average absolute value of steering wheel angle (X4), and the average absolute value of steering wheel angular velocity (X5) are used as the input for the model training set. "Careful," "Normal," and "Aggressive" are represented by "1, 0, 0," "0, 1, 0," and "0, 0, 1," respectively, and serve as the output for the model training set.

[0086] The GRNN network mainly consists of four layers: input layer, pattern layer, summation layer, and output layer.

[0087] Input layer: implements the data input process and passes the input data to the pattern layer. The number of nodes in this layer is the feature dimension of the input data. This article uses five feature indicators, so the number of nodes is 5, where x1 is the standard deviation of speed, x2 is the maximum absolute value of lateral acceleration, x3 is the standard deviation of lateral acceleration, x4 is the average absolute value of steering wheel angle, and x5 is the average absolute value of steering wheel angular velocity.

[0088] Pattern layer: Generally, Gaussian function is used to process input data, and the number of nodes is the number of training samples:

[0089] Summation layer: The summation layer mainly performs two summation calculations, one of which is the arithmetic sum of the outputs of all neurons in the pattern layer, and the other is the weighted sum of the outputs of all neurons in the pattern layer:

[0090] Output layer: The number of nodes in the output layer equals the output sample dimension. The arithmetic sum and weighted sum calculated by the summation layer are combined and output to produce the network's predicted output. If the output is (1, 0, 0), the driving style is cautious; if it is (0, 1, 0), the driving style is normal; and if it is (0, 0, 1), the driving style is aggressive.

[0091] In one embodiment, step S2, obtaining the driver's online lane change data and extracting characteristic indicators, includes:

[0092] Step S21: Obtain the distance change Δdis between the vehicle and the lane line at the current time t0 lane , if Δdis lane >3, then go to step S22, otherwise, repeat step S21 at the next moment;

[0093] Step S22: Calculate the distance dis between the vehicle and the obstacle at time t0-5 front , if 40>dis front >0, then go to step S23, otherwise go to step S21; here, the obstacle can be a vehicle or other obstacle, and the purpose of this step is to determine whether the driver changes lanes because there is a vehicle or obstacle in front.

[0094] Step S23, determining the lane change start time and end time, includes:

[0095] If str sw (t)<10° and str sw (t+0.5)>15°, t represents the time, t∈[t0-5,t0+5], str sw (t) is the steering wheel angle at time t, then the first time t1 = t;

[0096] If str sw (t)<-15° and str sw (t+0.5)>-10°, then the second moment t2=t;

[0097] Start time t start =min(t1,t2), end time t end =max(t1,t2);

[0098] This step is done by calculating the steering wheel angle str 10s before and after time t0. sw Perform cyclic judgment to determine the start and end time of lane change.

[0099] Step S24, extracting the vehicle driving data between the lane change start time and the lane change end time, i.e., the driver's online lane change data; here, the vehicle driving data may include vehicle speed, vehicle lateral acceleration, steering wheel angle, and steering wheel angular velocity data, etc.

[0100] Step S25 , extracting characteristic indicators based on the driver's online lane change data, the characteristic indicators including speed standard deviation, maximum absolute value of lateral acceleration, standard deviation of lateral acceleration, average absolute value of steering wheel angle, and average absolute value of steering wheel angular velocity.

[0101] In one embodiment, step S4, determining a vehicle stability factor according to the driver's driving style; determining a driving style gain coefficient according to the vehicle stability factor; and obtaining a reference yaw rate according to the driving style gain coefficient, includes:

[0102] First, a vehicle stability factor distribution map corresponding to each driving style is obtained.

[0103] Design two types of roads: roads with small curvature radius and roads with large curvature radius. Figure 2 The following figure shows a schematic diagram of the test road setup for drivers: (a) a small turning road and (b) a large turning road. The large curvature radius road is a curve with a length of 85m and a radius of 700m, while the small curvature radius road is a double lane change road. Drivers with different driving styles were tested under 18 different speed, road, and road adhesion conditions as shown in Table 1. The driving style gain coefficient Gain was adjusted to achieve the ideal vehicle response, and the corresponding vehicle stability factor K was calculated. style , calculate the vehicle stability factor K for each type of driving style driver style The average value is used as the control parameter of each type of controller, and the vehicle stability factor K of drivers with different driving styles is finally obtained. style data; Figure 3 The vehicle stability factor distribution diagram corresponding to each driving style is shown, where (a) is the vehicle stability factor distribution diagram corresponding to the cautious type, (b) is the vehicle stability factor distribution diagram corresponding to the general type, and (c) is the vehicle stability factor distribution diagram corresponding to the aggressive type.

[0104] Table 1 Driver in-loop test conditions

[0105]

[0106] Then, after obtaining the driver's driving style, according to the known sliding friction coefficient μ, vehicle longitudinal speed v x The vehicle stability factor K can be determined style ;

[0107] Then, the driving style gain coefficient Gain can be determined according to the following formula:

[0108]

[0109] Where K is the original stability factor.

[0110] Then, according to the following formula, the reference yaw rate is obtained

[0111]

[0112] in, is the reference yaw rate, Gain is the driving style gain coefficient, ω r_d is the ideal yaw rate.

[0113] In one embodiment, step S5, determining an instability index based on a reference yaw rate and an ideal center of mass sideslip angle, includes:

[0114] First, the β-ω phase plane is divided into three regions: stable region I, transition region II, and unstable region III. Figure 4 A schematic diagram of the β-ω phase plane division is shown. Figure 5 A schematic diagram of the phase plane transition domain division is shown.

[0115] Then, according to the reference yaw rate and the ideal center of mass sideslip angle β des , determine the area to which the vehicle belongs; here, set the ideal center of mass sideslip angle β des =0.

[0116] according to Figure 4 , given a known reference yaw rate and the ideal center of mass sideslip angle β des In this case, the area to which the vehicle belongs can be determined.

[0117] Then, based on the area to which the vehicle belongs, the following formula is used to determine the instability index:

[0118]

[0119] Among them, S ta is the instability index, Δγ r is the deviation between the actual yaw rate and the ideal yaw rate, Δγ max is the maximum deviation between the actual yaw rate and the ideal yaw rate, Δβ r is the deviation between the actual center of mass sideslip angle and the ideal center of mass sideslip angle, Δβ max It is the maximum deviation between the actual sideslip angle and the ideal sideslip angle.

[0120] In this embodiment, the instability index S is obtained by calculation ta ,The purpose is to coordinate the weight matrices of AFWS and DYC in the middle level controller in the subsequent steps.

[0121] In one embodiment, step S6 is specifically implemented by a personalized AFWS / DYC lateral stability controller based on game theory, which adopts hierarchical control and is divided into three layers: upper, middle, and lower. Figure 6 The block diagram of the personalized AFWS / DYC lateral stability controller based on game theory is shown. Here, the upper controller is used to implement steps S4 and S5, which will not be described in detail here. Figure 7 The established linear two-degree-of-freedom vehicle model of a four-wheel steering vehicle is shown.

[0122] The middle-level controller is used to implement step S6. Specifically, a cooperative game is performed based on the reference yaw rate, the ideal center of mass sideslip angle, and the instability index to obtain a preliminary front wheel steering angle increment and a preliminary rear wheel steering angle increment using the following formula:

[0123]

[0124] Among them, U′2(k) is the initial front wheel steering angle increment, U′3(k) is the initial rear wheel steering angle increment, K1 and K2 are feedback gains, is the inverse of K2, E is the coefficient matrix, x(k) is the state variable, L2 is the coefficient matrix of the front wheel steering angle increment, L3 is the coefficient matrix of the rear wheel steering angle increment, is the coefficient matrix of the prediction equation, Z is the coefficient matrix, and U1(k) is the direct yaw moment.

[0125] The specific representation of each parameter is given below:

[0126]

[0127] Among them, A (2) is the coefficient matrix of the tracker (front wheel angle increment and rear wheel angle increment);

[0128]

[0129] Wherein, i is the index of the control quantity, i=2,3, when i=2 represents the front wheel steering angle increment, i=3 represents the rear wheel steering angle increment. is the coefficient matrix of the prediction equation corresponding to the i-th control variable;

[0130] in, is the coefficient matrix of the prediction equation corresponding to the rear wheel steering angle increment;

[0131] S Q is the decomposition of Q, Q is the weight matrix;

[0132]

[0133] in, is the weight matrix, λ i is the weight coefficient of the participant cost function, and λ i >0,∑λ i =1;Q i is the weight coefficient, q β is the weight coefficient corresponding to the sideslip angle of the center of mass, is the weight coefficient corresponding to the yaw angular velocity;

[0134] R i The decomposition amount, R i is the weight matrix;

[0135]

[0136] in, is the weight matrix;

[0137]

[0138] r1=10 -(5+4LA)

[0139] r2=10+140LA

[0140] r3=r2

[0141]

[0142] Among them, r i is the control output weight of the i-th control variable, i=1,2,3, i=1 represents the direct yaw moment, LA is the parameter, S ta is the instability index, and N represents the prediction time domain.

[0143]

[0144] in, is the coefficient matrix of the prediction equation, with n = 1, 2, 3, and 4 representing the indices of the direct yaw moment, the front wheel angle increment, the rear wheel angle increment, and the front wheel angle, respectively. j = 1 and 2 represent the indices of the leader (direct yaw moment) and the follower (front wheel angle increment and rear wheel angle increment), respectively. N represents the prediction horizon. It should be noted that in this embodiment, i, j, and n are all indices representing different ways of representing the control variable indices.

[0145] C (j) For the matrix:

[0146]

[0147] Parameter matrix A = I + A0T s , I is the identity matrix, A0 is the coefficient matrix, T s is the sampling interval of the discrete model.

[0148]

[0149] Among them, k f 、k ris the tire cornering stiffness of the front and rear wheels, a and b are the distances from the center of mass to the front and rear axles, m is the vehicle mass; u is the longitudinal speed, I z is the moment of inertia of the vehicle around the z-axis.

[0150] Coefficient matrix B n =B 0n T s , B 0n is the coefficient, n=1,2,3,4.

[0151]

[0152] B 04 =B 02

[0153]

[0154] in, Predict output for followers, is the output at time k predicted at the current time k; δ f is the front wheel turning angle.

[0155] Specifically, the final direct yaw moment is:

[0156]

[0157] in, is the final direct yaw moment, k is the sampling time, L1 is the coefficient matrix of the direct yaw moment, and θ1 is the variable; for The amount of decomposition, is the weight matrix, T stands for transpose; for The amount of decomposition, is the weight matrix, Ω is the coefficient matrix; is the predicted stability control target; Λ is the coefficient matrix, X(k) is the matrix variable; Υ is the coefficient matrix.

[0158]

[0159] in, Predict output for leaders, is the output at time k predicted at the current time k. is the reference yaw rate, is the ideal center of mass sideslip angle.

[0160] Specifically, the final direct yaw moment is used to influence the preliminary front wheel steering angle increment and the preliminary rear wheel steering angle increment to obtain the final front wheel steering angle increment and the final rear wheel steering angle increment using the following formula:

[0161]

[0162] in, is the final front wheel angle increment, is the final rear wheel steering angle increment, K1 and K2 are feedback gains, is the inverse of K2, E is the coefficient matrix, x(k) is the state variable, L2 is the coefficient matrix of the front wheel steering angle increment, L3 is the coefficient matrix of the rear wheel steering angle increment, is the coefficient matrix of the prediction equation, L1 is the coefficient matrix of the direct yaw moment, θ1 is the variable, and Z is the coefficient matrix.

[0163] Based on the same inventive concept as the AFWS / DYC lateral stability coordinated control method considering driving style, this embodiment also provides a corresponding AFWS / DYC lateral stability coordinated control device considering driving style. Figure 8 The structural block diagram of the AFWS / DYC lateral stability coordinated control device considering driving style is shown, including:

[0164] A model building module 81 is used to build an online driving style identification model;

[0165] A characteristic index acquisition module 82 is used to obtain the driver's online lane change data and extract characteristic indicators;

[0166] an identification module 83 for inputting characteristic indicators into an online driving style identification model to obtain the driver's driving style;

[0167] The reference yaw rate acquisition module 84 is configured to determine a vehicle stability factor based on the driver's driving style; determine a driving style gain coefficient based on the vehicle stability factor; and obtain a reference yaw rate based on the driving style gain coefficient.

[0168] an instability degree index determination module 85 for determining an instability degree index based on a reference yaw rate and an ideal center of mass sideslip angle;

[0169] The control variable acquisition module 86 is configured to conduct cooperative and non-cooperative games based on the reference yaw rate, the ideal center of mass sideslip angle, and the instability index to obtain a preliminary front wheel angle increment, a preliminary rear wheel angle increment, and a final direct yaw moment; and utilize the final direct yaw moment to influence the preliminary front wheel angle increment and the preliminary rear wheel angle increment to obtain a final front wheel angle increment and a final rear wheel angle increment.

[0170] The control module 87 is configured to apply the final front wheel steering angle increment and the final rear wheel steering angle increment to the steering wheel, and convert the final direct yaw moment into tire longitudinal force and distribute it to each wheel.

[0171] The AFWS / DYC lateral stability coordination control device considering driving style of this embodiment has the same inventive concept as the AFWS / DYC lateral stability coordination control method considering driving style mentioned above. Therefore, the specific implementation method of the device can be seen in the embodiment part of the AFWS / DYC lateral stability coordination control method considering driving style mentioned above, and its technical effects correspond to the technical effects of the above method, which will not be repeated here.

[0172] An embodiment of the present application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the above-mentioned AFWS / DYC lateral stability coordinated control method considering driving style is implemented.

[0173] The above descriptions are merely examples of various embodiments of the present application, but the scope of protection of the present application is not limited thereto. Any modifications or substitutions that can be readily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included within the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. An AFWS / DYC lateral stability coordinated control method considering driving style, characterized in that: include: Build an online driving style identification model; Obtain driver's online lane-changing data and extract characteristic indicators; Inputting the characteristic index into the driving style online identification model to obtain the driver's driving style; determining a vehicle stability factor based on the driver's driving style; determining a driving style gain coefficient according to the vehicle stability factor; obtaining a reference yaw rate according to the driving style gain coefficient; determining an instability degree index according to the reference yaw rate and the ideal center of mass sideslip angle; performing cooperative and non-cooperative games based on the reference yaw rate, the ideal sideslip angle at the center of mass, and the instability index to obtain a preliminary front wheel steering angle increment, a preliminary rear wheel steering angle increment, and a final direct yaw moment; and utilizing the final direct yaw moment to influence the preliminary front wheel steering angle increment and the preliminary rear wheel steering angle increment to obtain a final front wheel steering angle increment and a final rear wheel steering angle increment; Applying the final front wheel steering angle increment and the final rear wheel steering angle increment to the steering wheel, and converting the final direct yaw moment into tire longitudinal force and distributing it to each wheel; The reference yaw rate is: in, is the reference yaw rate, Gain is the driving style gain coefficient, ω r_d is the ideal yaw rate; The final direct yaw moment is: in, is the final direct yaw moment, k is the sampling time, L1 is the coefficient matrix of the direct yaw moment, and θ1 is the variable; for The amount of decomposition, is the weight matrix, T stands for transpose; for The amount of decomposition, is the weight matrix, Ω is the coefficient matrix; is the predicted stability control target; Λ is the coefficient matrix, X(k) is the matrix variable; Υ is the coefficient matrix; Weight Matrix for: r1＝10 -(5+4LA) Among them, r1 is the control output weight of the direct yaw moment, LA is the parameter, S ta is an index of instability, and N represents the prediction time domain.

2. The method according to claim 1, wherein in, Build an online driving style identification model, including: Obtain lane-changing process data from multiple drivers, and extract feature indicators from each driver's lane-changing process data as training samples; Perform cluster analysis on all training samples to determine the driving style category of each training sample; The training samples and the corresponding driving style categories constitute a driving style dataset; The GRNN is trained based on the driving style dataset to obtain an online driving style identification model.

3. The method according to claim 1, wherein in, Obtain driver online lane change data and extract characteristic indicators, including: Step S21: Obtain the distance change Δdis between the vehicle and the lane line at the current time t0 lane , if Δdis lane >3, then go to step S22, otherwise, repeat step S21 at the next moment; Step S22: Calculate the distance dis between the vehicle and the obstacle at time t0-5 front , if 40>dis front >0, then go to step S23, otherwise go to step S21; Step S23, determining the lane change start time and end time, includes: If str sw (t)<10° and str sw (t+0.5)>15°, t represents the time, t∈[t0-5,t0+5], str sw (t) is the steering wheel angle at time t, then the first time t1 = t; If str sw (t)<-15° and str sw (t+0.5)>-10°, then the second moment t2=t; Start time t start =min(t1,t2), end time t end =max(t1,t2); Step S24, extracting the vehicle driving data between the lane change start time and the lane change end time, i.e., the driver's online lane change data; Step S25 , extracting characteristic indicators based on the driver's online lane change data, wherein the characteristic indicators include speed standard deviation, maximum absolute value of lateral acceleration, standard deviation of lateral acceleration, average absolute value of steering wheel angle, and average absolute value of steering wheel angular velocity.

4. The method according to claim 1, wherein in, Determining an instability index based on the reference yaw rate and the ideal center of mass sideslip angle includes: The β-ω phase plane is divided into three regions: stable region I, transition region II, and unstable region III. determining a region to which the vehicle belongs based on the reference yaw rate and the ideal center-of-mass sideslip angle; According to the area to which the vehicle belongs, the instability index is determined using the following formula: Among them, S ta is the instability index, Δγ r is the deviation between the actual yaw rate and the ideal yaw rate, Δγ max is the maximum deviation between the actual yaw rate and the ideal yaw rate, Δβ r is the deviation between the actual center of mass sideslip angle and the ideal center of mass sideslip angle, Δβ max It is the maximum deviation between the actual sideslip angle and the ideal sideslip angle.

5. The method according to claim 1, wherein The final front wheel turning angle increment and the final rear wheel turning angle increment are: in, is the final front wheel angle increment, is the final rear wheel steering angle increment, K1 and K2 are feedback gains, is the inverse of K2, E is the coefficient matrix, x(k) is the state variable, L2 is the coefficient matrix of the front wheel steering angle increment, L3 is the coefficient matrix of the rear wheel steering angle increment, is the coefficient matrix of the prediction equation, L1 is the coefficient matrix of the direct yaw moment, θ1 is the variable, and Z is the coefficient matrix.

6. An AFWS / DYC lateral stability coordinated control device considering driving style, characterized in that: include: Model building module, used to build an online driving style identification model; A characteristic index acquisition module is used to obtain the driver's online lane change data and extract characteristic indicators; an identification module, configured to input the characteristic index into the driving style online identification model to obtain the driver's driving style; a reference yaw rate acquisition module, configured to determine a vehicle stability factor based on the driver's driving style; determining a driving style gain coefficient according to the vehicle stability factor; obtaining a reference yaw rate according to the driving style gain coefficient; an instability degree index determination module, configured to determine an instability degree index based on the reference yaw angular velocity and the ideal center of mass sideslip angle; a control variable acquisition module, configured to perform a cooperative game and a non-cooperative game based on the reference yaw rate, the ideal center of mass sideslip angle, and the instability degree index to obtain a preliminary front wheel steering angle increment, a preliminary rear wheel steering angle increment, and a final direct yaw moment; and utilize the final direct yaw moment to influence the preliminary front wheel steering angle increment and the preliminary rear wheel steering angle increment to obtain a final front wheel steering angle increment and a final rear wheel steering angle increment; a control module, configured to apply the final front wheel steering angle increment and the final rear wheel steering angle increment to the steering wheel, and convert the final direct yaw moment into a tire longitudinal force and distribute it to each wheel; The reference yaw rate is: in, is the reference yaw rate, Gain is the driving style gain coefficient, ω r_d is the ideal yaw rate; The final direct yaw moment is: in, is the final direct yaw moment, k is the sampling time, L1 is the coefficient matrix of the direct yaw moment, and θ1 is the variable; for The amount of decomposition, is the weight matrix, T stands for transpose; for The amount of decomposition, is the weight matrix, Ω is the coefficient matrix; is the predicted stability control target; Λ is the coefficient matrix, X(k) is the matrix variable; Υ is the coefficient matrix; Weight Matrix for: r1＝10 -(5+4LA) Among them, r1 is the control output weight of the direct yaw moment, LA is the parameter, S ta is an index of instability, and N represents the prediction time domain.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the AFWS / DYC lateral stability coordinated control method considering driving style as described in any one of claims 1-5.

Images