Steering control method and system for autonomous vehicle, storage medium and vehicle

Abstract

The invention provides an automatic driving vehicle steering control method and system, a storage medium and a vehicle. The method comprises the steps that a three-degree-of-freedom vehicle dynamics model with an active transverse stabilizer bar is established; converting the three-degree-of-freedom vehicle dynamics model into a state space equation form; performing discretization processing on the state-space equation by adopting a model predictive control algorithm, constructing an objective function and constraint conditions by taking tracking of an expected trajectory and minimum change of a controlled quantity as a control objective, and obtaining an optimal controlled quantity through rolling optimization solution; the optimal control quantity comprises a steering wheel turning angle, a vehicle rear wheel camber angle and a transverse stabilizer bar torque; and performing steering control on the vehicle according to the optimal control quantity. By implementing the method, the steering capability of the automatic driving vehicle can be improved, and meanwhile, the roll stability and the riding comfort of the vehicle during steering are improved.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving, and particularly to a steering control method, system, storage medium and vehicle for an autonomous driving vehicle. Background Art

[0002] At present, when an autonomous driving vehicle travels at a high speed on a large-curvature bend or quickly changes lanes in a scenario with a large traffic flow, due to the insufficient lateral force provided by the wheel sideslip angle, dangerous situations such as vehicle sideslip are likely to occur. To avoid such dangers, autonomous driving vehicles currently mostly adopt the method of reducing the vehicle speed, but this will greatly reduce the traffic efficiency of the vehicle.

[0003] The active wheel camber control technology can provide an additional lateral force for the wheels by actively controlling the camber angle, improving the adhesion characteristics of the tires, thereby improving the handling stability of the vehicle during steering. However, when the vehicle travels at a high speed with a large steering angle, considering the roll dynamics characteristics of the vehicle, the roll angle will increase, which is not conducive to the roll stability of the vehicle.

[0004] In the scenario where an autonomous driving vehicle passes through a large-curvature bend at a high speed or needs to quickly change lanes, the problem of insufficient steering ability may still lead to dangerous situations such as sideslip. When the vehicle turns at a high speed, the lateral force provided by the sideslip angle reaches the limit, and the vehicle is extremely prone to sideslip. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a steering control method, system, storage medium and vehicle for an autonomous driving vehicle, which can improve the steering ability of the autonomous driving vehicle, and at the same time improve the roll stability and ride comfort of the vehicle during steering.

[0006] As one aspect of the present invention, a vehicle steering control method is provided, which includes the following steps:

[0007] Establish a three-degree-of-freedom vehicle dynamics model with an active anti-roll bar, where the three degrees of freedom include yaw, lateral, and roll directions;

[0008] Convert the three-degree-of-freedom vehicle dynamics model into the form of a state-space equation;

[0009] Adopt a model predictive control algorithm to discretize the state-space equation, and take tracking the desired trajectory and minimizing the change of the control quantity as the control objective, construct an objective function and constraint conditions, and obtain the optimal control quantity through rolling optimization, where the optimal control quantity includes the steering wheel angle, the rear wheel camber angle of the vehicle, and the anti-roll bar torque;

[0010] According to the optimal control quantity, perform steering control on the vehicle.

[0011] Among them, a three-degree-of-freedom vehicle dynamics model oriented to control is established, including the vehicle's dynamic equations in three degrees of freedom, including:

[0012] The following lateral motion equation is established:

[0013]

[0014] where a y is the lateral acceleration, m is the vehicle mass, m s is the sprung mass, h s is the distance from the vehicle's center of mass to the roll center (h s =H - h RC ), H is the height of the sprung mass from the ground when no roll occurs, h RC is the height of the roll center from the ground, is the roll angular acceleration, F y1 F y2 、F y31 and F y4 are the lateral forces of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively;

[0015] The following yaw motion equation is established:

[0016]

[0017] where I z is the yaw moment of inertia, is the yaw angular acceleration, l f and l r are the distances from the center of mass to the front axle and to the rear axle respectively;

[0018] The following roll motion equation is established:

[0019]

[0020] where, I x is the roll moment of inertia, I xs is the roll moment of inertia considering the influence of the sprung mass, is the roll angular acceleration, is the roll angle, is the roll angular velocity, M d is the active roll moment, and are the suspension roll stiffness and roll damping respectively;

[0021] where:

[0022] F y1 =C α1 α1,

[0023] F y2 = C α2 α2,

[0024] F y3 = C α3 α3 + C γ3 γ3,

[0025] F y4 = C α4 α4 + C γ4 γ4

[0026] where α1, α2, α3, and α4 are the sideslip angles of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively, C α1 , C α2 , C α3 and C α4 are the sideslip stiffnesses of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively; C γ3 and C γ3 are the camber stiffnesses of the left rear wheel and right rear wheel respectively; γ3 and γ4 are the camber angles of the left rear wheel and right rear wheel respectively.

[0027] Among them, the conversion of the three-degree-of-freedom vehicle dynamics model into the state-space equation form and the discretization process include:

[0028] Converting the three-degree-of-freedom vehicle dynamics model into the following state-space equation:

[0029]

[0030] Where:

[0031] x is the state vector,

[0032] u is the state vector, u = [δ f γ3 γ4 M d T

[0033] y is the output vector,

[0034] A is the system matrix, B is the input matrix, C is the output matrix, and E is the feedforward matrix, which are as follows:

[0035]

[0036] Among them, a 13 = -(C α1 + C α2 )E f , a 14 = 0, ​ a 23 =-l f E f (C α1 +C α2 ),a 24 =0,a 31 =0,a 32 =0,a 33 =0,a 34 =1,a 41 =0,a 42 =-m s h s v x , b 11 =-(C α1 +C α2 ), b 14 =0,b 21 =-l f (C α1 +C α2 ), b 24 =0,b 31 =0,b 32 =0,b 33 =0,b 34 =0,b 41 =0,b 42 =0,b 43 =0,b 44 =-1。

[0037] Among them, the model predictive control algorithm is adopted to discretize the state space equation, and with the goal of tracking the desired trajectory and minimizing the change of the control quantity, an objective function and constraint conditions are constructed, and the optimal control quantity is obtained by solving through rolling optimization, including:

[0038] The state space equation is discretized to obtain the discretized state space equation:

[0039] A T =I+E -1 AT,

[0040] B T =E -1 BT;

[0041] Among them, A T is the transition matrix of the discrete state, I is the identity matrix, A is the system matrix, T is the sampling period, and E -1 is the inverse matrix of the feedforward matrix; B T is the input matrix of the discrete state, and B is the input matrix;

[0042] Using the model predictive control algorithm, the discretized state - space equation is used to predict the state trajectories {x(k + 1), x(k + 2), …, x(k + Np)} and output trajectories {y(k + 1), y(k + 2), …, y(k + Np)} in the next N p steps at each sampling time. If the current state is x(k), the next - step prediction is x(k + 1) = A T x(k)+B T u(k);

[0043] Among them, taking the front - wheel steering angle, the camber angles of the left and right rear wheels, and the anti - roll bar torque as control variables, the model predictive control algorithm is used for rolling optimization to solve the following online optimization problem with constraints online to obtain the optimal control variables:

[0044] Among them, the objective function is:

[0045]

[0046] The constraint conditions are:

[0047] Δu min ≤Δu(k + i)≤Δu max

[0048] u min ≤u(k + i)≤u max

[0049] y mon ≤y(k + i)≤y max

[0050] The reference output quantity is:

[0051]

[0052] Among them, where R is the reciprocal of the lane curvature and y ref is the yaw rate.

[0053] Among them, according to the optimal control variables, the vehicle steering control is carried out, including:

[0054] Applying the first element in the optimal control variable sequence to the autonomous vehicle to realize vehicle steering control;

[0055] After realizing the vehicle steering control, the vehicle state is feedback - corrected to adjust the subsequent control variables in real time.

[0056] Correspondingly, as another aspect of the present invention, a vehicle steering control system is further provided, which includes:

[0057] A dynamic model establishment module, which is used to establish a three-degree-of-freedom vehicle dynamic model with an active anti-roll bar, and the three degrees of freedom include yaw, lateral, and roll directions;

[0058] A state space conversion module, which is used to convert the three-degree-of-freedom vehicle dynamic model into the form of a state space equation;

[0059] A prediction processing module, which is used to adopt a model predictive control algorithm to discretize the state space equation, and construct an objective function and constraint conditions with the goals of tracking the desired trajectory and minimizing the change of the control quantity, and obtain the optimal control quantity through rolling optimization solution. The optimal control quantity includes the steering wheel angle, the camber angle of the vehicle's rear wheels, and the anti-roll bar torque;

[0060] A control execution module, which is used to perform steering control on the vehicle according to the optimal control quantity.

[0061] Among them, the dynamic model establishment module includes:

[0062] A lateral equation establishment unit, which is used to establish the following lateral motion equation:

[0063]

[0064] Among them, a y is the lateral acceleration, m is the vehicle mass, and m s is the sprung mass, h s is the distance from the vehicle's center of mass to the roll center (h s =H - h RC ), H is the height of the sprung mass from the ground when there is no roll, and h RC is the height of the roll center from the ground, is the roll angular acceleration, and F y1 F y2 、F y31 and F y4 are the lateral forces of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively;

[0065] Among them:

[0066] F y1 =C α1 α1,

[0067] F y2 =C α2 α2,

[0068] F y3 =C α3 α3 + C γ3 γ3,

[0069] F y4 =Cα4 α4 + C γ4 γ4

[0070] where α1, α2, α3, and α4 are the sideslip angles of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively, and C α1 , C α2 , C α3 and C α4 are the sideslip stiffnesses of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively; C γ3 and C γ3 are the camber stiffnesses of the left rear wheel and right rear wheel respectively; γ3 and γ4 are the camber angles of the left rear wheel and right rear wheel respectively;

[0071] Yaw equation establishment unit, for establishing the following yaw motion equation:

[0072]

[0073] where I z is the yaw moment of inertia, is the yaw angular acceleration, l f and l r are the distances from the center of mass to the front axle and to the rear axle respectively;

[0074] Roll equation establishment unit, for establishing the following roll motion equation:

[0075]

[0076] where, I x is the roll moment of inertia, I xs is the roll moment of inertia considering the influence of the sprung mass, is the roll angular acceleration, is the roll angle, is the roll angular velocity, M d is the active roll moment, and are the suspension roll stiffness and roll damping respectively.

[0077] where, in the state space conversion module, specifically, the three - degree - of - freedom vehicle dynamics model is converted into the following state space equation:

[0078]

[0079] where:

[0080] x is the state vector,

[0081] u is the state vector, u = [δ f γ3 γ4 M dT

[0082] y is the output vector,

[0083] A is the system matrix, B is the input matrix, C is the output matrix, and E is the feedforward matrix, which are as follows:

[0084]

[0085] where, a 13 = -(C α1 + C α2 )E f , a 14 = 0, a 23 = -l f E f (C α1 + C α2 ), a 24 = 0, a 31 = 0, a 32 = 0, a 33 = 0, a 34 = 1, a 41 = 0, a 42 = -m s h s v x , b 11 = -(C α1 + C α2 ), b 14 = 0, b 21 = -l f (C α1 + C α2 ), b 24 = 0, b 31 = 0, b 32 = 0, b 33 = 0, b 34 = 0, b 41 = 0, b 42 = 0, b 43 = 0, b 44 = -1;

[0086] where, the prediction processing module includes:

[0087] A discretization processing unit for discretizing the state - space equation to obtain the discretized state - space equation: ​

[0088] A t = I + E -1 AT,

[0089] B t = E -1 BT;

[0090] Wherein, A T is the transition matrix of discrete states, I is the identity matrix, A is the system matrix, T is the sampling period, and E -1 is the inverse matrix of the feedforward matrix; B T is the input matrix of discrete states, and B is the input matrix;

[0091] The algorithm construction unit is used to adopt the model predictive control algorithm to predict the state trajectories {x(k + 1), x(k + 2), …, x(k + Np)} and output trajectories {y(k + 1), y(k + 2), …, y(k + Np)} of the next N o steps at each sampling moment by using the discretized state - space equation. If the current state is x(k), then the next - step prediction is x(k + 1) = A T x(k)+B T u(k);

[0092] The rolling optimization unit is used to take the front - wheel steering angle, the camber angles of the left and right rear wheels, and the anti - roll bar torque as control variables, and perform rolling optimization by using the model predictive control algorithm to online solve the following constrained online optimization problem to obtain the optimal control variables:

[0093] Wherein, the objective function is:

[0094]

[0095] The constraint conditions are:

[0096] Δu min ≤Δu(k + i)≤Δu max

[0097] u min ≤u(k + i)≤u max

[0098] y min ≤y(k + i)≤y max

[0099] The reference output quantity is:

[0100] y ref (k + i)=[0 ω d 0] T (i = 1, 2, …, N p )

[0101] Among them, where R is the reciprocal of the lane curvature, and y ref is the yaw rate.

[0102] Among them, the control execution module includes:

[0103] A steering control unit, which is used to apply the first element in the optimal control quantity sequence to the autonomous vehicle to achieve vehicle steering control;

[0104] A feedback correction unit, which is used to perform feedback correction on the vehicle state after achieving vehicle steering control so as to adjust subsequent control quantities in real time.

[0105] As another aspect of the present invention, there is also provided a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor module, the steps of the method as described above are implemented.

[0106] As another aspect of the present invention, there is also provided a vehicle, which includes:

[0107] One or more processor modules;

[0108] A memory, which is used to store one or more computer programs;

[0109] When the one or more computer programs are executed by the one or more processor modules, the one or more processor modules implement the method as described above.

[0110] Implementing the embodiments of the present invention has the following beneficial effects:

[0111] The present invention relates to a vehicle steering control method, system, storage and vehicle. Specifically, based on the three-degree-of-freedom vehicle dynamics model of the vehicle, by introducing key factors such as the camber stiffness of the left and right rear wheels, the camber angles of the left and right rear wheels, and the active roll moment, a new dynamics model is constructed. Subsequently, the established dynamics model is converted into a discretized state space equation, and the front wheel steering angle, the camber angles of the left and right rear wheels, and the anti-roll bar moment are set as control quantities. By solving the constrained online optimization problem, the specific values of the front wheel steering angle, the camber angles of the left and right rear wheels, and the anti-roll bar moment are accurately determined.

[0112] The method of the present invention applies the coordinated control strategy of three control variables, namely, steering wheel angle, wheel camber angle, and anti-roll bar torque, to autonomous vehicles. With the help of the active anti-roll bar, the roll angle of the vehicle can be effectively reduced, thereby expanding the boundary of the vehicle's steering ability. This method can not only significantly improve the steering performance of the vehicle, enabling it to smoothly pass through curves with large curvatures and achieve rapid lane changes at high speeds, but also reduce the roll angle of the vehicle during steering, enhance the roll stability of the vehicle, and thus improve the riding comfort. BRIEF DESCRIPTION OF THE DRAWINGS

[0113] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following-described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, obtaining other drawings based on these drawings still belongs to the scope of the present invention.

[0114] Figure 1 It is a schematic main process diagram of an embodiment of a vehicle steering control method provided by the present invention;

[0115] Figure 2 It is a schematic diagram of a three-degree-of-freedom vehicle dynamics model in a front view angle involved in the method of the present invention;

[0116] Figure 3 It is a schematic diagram of a three-degree-of-freedom vehicle dynamics model in a top view angle involved in the method of the present invention;

[0117] Figure 4 It is a schematic structural diagram of an embodiment of a vehicle steering control system provided by the present invention;

[0118] Figure 5 For Figure 4 It is a schematic structural diagram of the dynamic model establishment module in

[0119] Figure 6 For Figure 4 It is a schematic structural diagram of the prediction processing module in

[0120] Figure 7 For Figure 4 It is a schematic structural diagram of the control execution module in DETAILED DESCRIPTION OF THE EMBODIMENTS

[0121] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings.

[0122] As Figure 1As shown, a main flowchart of an embodiment of a vehicle steering control method provided by the present invention is shown. In combination with Figure 2 , Figure 3 shown, in this embodiment, the vehicle steering control method at least includes the following steps:

[0123] Step S10, establish a three-degree-of-freedom vehicle dynamics model with an active anti-roll bar, where the three degrees of freedom include yaw, lateral, and roll directions;

[0124] In the present invention, a vehicle dynamics model with an active anti-roll bar is adopted. Among them, the anti-roll bar is equivalent to a torsion bar spring. When the vehicle is driving straight, the left and right wheels hardly undergo load transfer, and the deformations of the left and right suspensions remain the same. At this time, the anti-roll bar does not act on the vehicle. When the vehicle is turning, the vehicle undergoes a roll motion, the loads of the left and right wheels are transferred, and the deformation amounts of the left and right suspensions are different, causing the anti-roll bar to twist and generate a torsional moment. The torsional moment generates a pair of equal and opposite acting forces at the connection between the anti-roll bar and the vehicle body. This pair of acting forces is manifested as an anti-roll moment M d , as Figure 2 shown. The anti-roll moment can effectively suppress the roll motion of the vehicle body and improve the roll stability of the vehicle.

[0125] The active anti-roll bar adds an actuator on the basis of the traditional passive anti-roll bar. It can control the deformation amount of the anti-roll bar according to the current state of the vehicle and adjust the anti-roll moment in real time, so as to achieve the purpose of controlling the roll attitude of the vehicle body.

[0126] In combination with Figure 3 shown, the three-degree-of-freedom vehicle dynamics model in this step includes three degrees of freedom: yaw, lateral, and roll. This vehicle model can well express the coupling relationship between the three degrees of freedom. According to Newton's second law, the dynamic equations in each direction can be obtained:

[0127]

[0128] The above three-degree-of-freedom vehicle dynamics model will be described in detail below.

[0129] In a specific example, step S10 further includes:

[0130] Step S100, establish the following lateral motion equation:

[0131]

[0132] where a y is the lateral acceleration, m is the vehicle mass, m sis the sprung mass, h s is the distance from the vehicle's center of mass to the roll center (h s = H - h RC ), H is the height of the sprung mass from the ground when there is no roll, h RC is the height of the roll center from the ground, is the roll angular acceleration, F y1 F y2 、F y31 and F y4 are the lateral forces of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively;

[0133] Furthermore, ma y represents the inertial force of the vehicle in the lateral direction, represents the additional lateral force caused by the roll motion; F y1 + F y2 + F y3 + F y4 is the sum of the lateral forces of the four wheels and can represent the external force received by the vehicle in the lateral direction;

[0134] This lateral motion equation describes the force balance relationship of the vehicle in the lateral direction, taking into account the vehicle's inertial force, the additional lateral force caused by the roll motion, and the action of the wheel lateral forces.

[0135] Step S101, establish the following yaw motion equation:

[0136]

[0137] where, I z is the yaw moment of inertia, is the yaw angular acceleration, l f and l r are the distances from the center of mass to the front axle and to the rear axle respectively;

[0138] Furthermore, represents the inertial moment of the vehicle in the yaw direction; l f (F y1 + F y2 ) represents the moment contribution of the front wheel lateral force to the vehicle's yaw motion; l r (F y3 + F y4 ) represents the moment contribution of the rear wheel lateral force to the vehicle's yaw motion, and its sign is negative because the moment generated by the rear wheel lateral force is opposite to the front wheel direction.

[0139] This yaw motion equation describes the moment balance relationship of the vehicle in the yaw direction, taking into account the vehicle's inertial moment and the influence of the front and rear wheel lateral forces on the yaw motion.

[0140] Step S102, establish the following roll motion equation:

[0141]

[0142] Wherein, I x is the roll moment of inertia, and I xs is the roll moment of inertia considering the influence of the sprung mass distribution changing with the roll angle, is the roll angular acceleration, is the roll angle, is the roll angular velocity, M d is the active roll moment, and are the suspension roll stiffness and roll damping respectively;

[0143] Furthermore, represents the inertial moment of the vehicle in the roll direction; m s a y h s represents the roll moment caused by the lateral acceleration; represents the roll restoring moment caused by gravity; represents the elastic restoring moment of the suspension system to the roll motion; represents the damping moment of the suspension system to the roll motion; M d is the externally applied control moment for actively controlling the roll motion of the vehicle.

[0144] In this roll motion equation, the moment balance relationship of the vehicle in the roll direction is described, considering the effects of the vehicle inertial moment, the roll moment caused by the lateral acceleration, the roll restoring moment caused by gravity, the elastic restoring moment and damping moment of the suspension system, and the active roll moment.

[0145] In the above steps S100 to S102, assuming that the front wheel steering angle is relatively small and the tire model is within the linear range, then there is:

[0146]

[0147] Wherein, α i (i = 1, 2, 3, 4) are the left front wheel, right front wheel, left rear wheel and right rear wheel sideslip angles respectively, and δ f is the front wheel steering angle.

[0148] Thus, the lateral forces of each wheel are obtained:

[0149] F y1 = C α1 α1,

[0150] F y2 = Cα2 α2,

[0151] F y3 = C α3 α3 + C γ3 γ3,

[0152] F y4 = C α4 α4 + C γ4 γ4

[0153] where α1, α2, α3, and α4 are the sideslip angles of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively, and C α1 , C α2 , C α3 , and C α4 are the sideslip stiffnesses of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively; C γ3 and C γ3 are the camber stiffnesses of the left rear wheel and right rear wheel respectively; γ3 and γ4 are the camber angles of the left rear wheel and right rear wheel respectively.

[0154] Substitute the calculation formulas of F y1 , F y2 , F y3 , F y4 into the above lateral motion equation and yaw motion equation, and a complete three - degree - of - freedom vehicle dynamics model can be obtained.

[0155] Step S11: Convert the three - degree - of - freedom vehicle dynamics model into the form of a state - space equation;

[0156] Specifically, in an actual example, step S11 specifically includes:

[0157] Convert the three - degree - of - freedom vehicle dynamics model into the following state - space equation:

[0158]

[0159] where:

[0160] x is the state vector, These variables can comprehensively describe the lateral and roll dynamic states of the vehicle during steering.

[0161] u is the input vector, u = [δ f γ3 γ4 M d T , and these are the input quantities that can be adjusted through the vehicle's control system to affect the vehicle's motion state and achieve the control of vehicle steering and roll.

[0162] y is the output vector, ​These output quantities are usually the key indicators of vehicle motion that we are concerned about, and are used to evaluate the steering performance and roll stability of the vehicle, etc.

[0163] A is the system matrix, which describes the internal dynamic relationship between system state variables, reflecting the physical and dynamic characteristics of the vehicle itself, such as the influence of mass, inertia, tire characteristics, etc. on the vehicle motion state. For example, the element a 11 、a 12 in the matrix A, etc. respectively represent the influence coefficients of lateral velocity and yaw angular velocity on themselves and other state variables, and these coefficients are determined by vehicle parameters (such as tire cornering stiffness, vehicle center of mass position, wheelbase, etc.). Specifically:

[0164]

[0165] B is the input matrix, which represents the influence relationship of control inputs on state variables, that is, how control quantities such as front wheel steering angle, rear wheel camber angle, and anti-roll bar torque act on the vehicle to change state variables such as lateral velocity, yaw angular velocity, roll angle, and roll angular velocity. For example, the element b 11 in the matrix B represents the influence coefficient of the front wheel steering angle on the lateral velocity, and b 23 represents the influence coefficient of the right rear wheel camber angle on the yaw angular velocity, etc. Specifically:

[0166]

[0167] C is the output matrix, which defines the linear combination relationship between state variables and output variables, that is, how to obtain the output quantities we are concerned about from state variables, such as sideslip angle, yaw angular velocity, and roll angle. For example, the element in the first row and first column of the matrix C is 1 / v_x, representing the proportional relationship between lateral velocity and sideslip angle, while other elements indicate that the yaw angular velocity and roll angle directly correspond to the corresponding output quantities; specifically:

[0168]

[0169] E is the feedforward matrix, which contains parameters such as the mass and moment of inertia of the vehicle, and is used to transform the dynamic equation into the standard form of the state space equation. For example, m is the vehicle mass, I z is the yaw moment of inertia, m s is the sprung mass of the vehicle, h s is the distance from the vehicle center of mass to the roll center, and I xs is the roll moment of inertia.

[0170]

[0171] Among them, a 13 = -(Cα1 +C α2 )E f ,a 14 = 0, a 23 = -l f E f (C α1 +C α2 ),a 24 = 0,a 31 = 0,a 32 = 0,a 33 = 0,a 34 = 1,a 41 = 0,a 42 = -m s h s v x , b 11 = -(C α1 +C α2 ), b 14 = 0,b 21 = -l f (C α1 +C α2 ), b 24 = 0,b 31 = 0,b 32 = 0,b 33 = 0,b 34 = 0,b 41 = 0,b 42 = 0,b 43 = 0,b 44 = -1.

[0172] It can be understood that the state - space equation is the basis of the model predictive control algorithm (MPC). After converting the dynamic model into this form, it is convenient to use the MPC algorithm for rolling optimization to solve the optimal control quantities, such as the front - wheel steering angle, the camber angles of the left and right rear wheels, and the anti - roll bar torque, etc., to achieve the steering control and roll - stability control of the vehicle.

[0173] Through the state - space equation, it can be clearly shown how the state variables of the vehicle system (lateral velocity, yaw rate, roll angle, and roll - angle velocity) change with the control inputs (front - wheel steering angle, rear - wheel camber angle, and anti - roll bar torque) and the dynamic characteristics of the system itself (determined by matrices A and B), which is convenient for analyzing and predicting the dynamic behavior of the vehicle.

[0174] Step S12: Using the model predictive control algorithm, discretize the state space equation, and construct the objective function and constraint conditions with the objectives of tracking the desired trajectory and minimizing the change in the control quantity. Then, obtain the optimal control quantity through rolling optimization. The optimal control quantity includes the steering wheel angle, the camber angle of the vehicle's rear wheels, and the torque of the anti-roll bar.

[0175] In a specific example, step S12 further includes:

[0176] Step S120: Discretize the state space equation to obtain the discretized state space equation:

[0177] A T = I + E -1 AT,

[0178] B T = E -1 BT;

[0179] where A T is the transition matrix of the discrete state, I is the identity matrix, A is the system matrix, T is the sampling period, and E -1 is the inverse matrix of the feedforward matrix; B T is the input matrix of the discrete state, and B is the input matrix; where A T = I + E -1 AT can describe the linear evolution of the state within the sampling period and can be achieved through the first-order Taylor approximation. And B T = E -1 BT can describe the cumulative effect of the input within the sampling period.

[0180] It can be understood that discretizing the continuous-time state space equation can better adapt to the implementation of digital control systems. In the actual control of autonomous vehicles, digital computers are usually used for control calculations, and the discretized state space equation can be directly used to write control programs to achieve the digital implementation of control algorithms.

[0181] Step S121: Using the model predictive control algorithm, predict the state trajectories {x(k + 1), x(k + 2), …, x(k + Np)} and output trajectories {y(k + 1), y(k + 2), …, y(k + Np)} of the next N p steps at each sampling moment using the discretized state space equation. If the current state is x(k), then the next prediction is x(k + 1) = A T x(k) + B T u(k);

[0182] Among them, \(x(k)\) is the system state vector at discrete time \(k\), which describes the internal state of the system at a certain moment. \(u(k)\) is the system input vector at discrete time \(k\), representing the control action of the outside on the system. \(y(k)\) is the system output vector at discrete time \(k\), reflecting the observable behavior of the system under the action of the input and state.

[0183] Step S13: Take the front wheel steering angle, the camber angles of the left and right rear wheels, and the anti-roll bar torque as control variables, and use the model predictive control algorithm to perform rolling optimization to solve the following constrained online optimization problem online to obtain the optimal control variables:

[0184] Among them, the objective function is:

[0185]

[0186] It can be understood that this objective function aims to minimize the error between the predicted output \(y(k + i)\) and the reference output \(y_{ref}(k + i)\) (weighted by the weight matrix \(Q\)), the magnitude of the control input increment \(\Delta u(k + i)\) (weighted by the weight matrix \(R\)), and the relevant term of a relaxation factor \(\varepsilon\) (with a coefficient of \(\rho\)). \(N\) p is the prediction horizon of the control system, and \(N\) c is the control horizon.

[0187] The constraint conditions are:

[0188] Condition 1, \(\Delta u\) min \(\leq\Delta u(k + i)\leq\Delta u\) max

[0189] Condition 2, \(u\) min \(\leq u(k + i)\leq u\) max

[0190] Condition 3, \(y\) min \(\leq y(k + i)\leq y\) max

[0191] Among them, Condition 1 is used to constrain the upper and lower bounds of the control input increment \(\Delta u(k + i)\) to ensure that the change amount of the control input is within a reasonable range and avoid overly drastic control actions;

[0192] Condition 2 is used to constrain the upper and lower bounds of the control input \(u(k + i)\) to limit the value range of the control input itself to protect the actuator or meet the physical limitations of the actual system;

[0193] Condition 3 is used to implement the upper and lower bounds of the system output \(y(k + i)\) to ensure that the system output is within a safe or desired range and prevent the system from operating beyond the allowed boundaries.

[0194] These constraints work together to ensure that during the optimization process, the obtained control input sequence can not only minimize the objective function but also meet the actual operating requirements and safety restrictions of the system.

[0195] The reference output is:

[0196] y ref (k + i) = [0 ω d 0] T (i = 1, 2, …, N p )

[0197] Among them, where R is the reciprocal of the lane curvature, and y ref is the yaw rate.

[0198] Step S13: Perform steering control on the vehicle according to the optimal control quantity.

[0199] In a specific example, the step S13 further includes:

[0200] Apply the first element in the optimal control quantity sequence to the autonomous vehicle to achieve vehicle steering control; that is, apply the optimal wheel camber angle, the optimal steering wheel angle, and the optimal anti-roll bar torque in the optimal control quantity sequence to the autonomous vehicle, and let the autonomous vehicle perform steering control according to this parameter;

[0201] After achieving vehicle steering control, perform feedback correction on the vehicle state (such as vehicle yaw rate, roll angle and other state variables) to adjust the subsequent control quantity in real time.

[0202] It can be understood that the method provided by the present invention introduces the left and right rear wheel camber stiffness, the left and right rear wheel camber angles, and the active roll torque into the three-degree-of-freedom vehicle dynamics model of the vehicle, thereby establishing a new dynamics model; then convert the dynamics model into a discretized state-space equation, taking the front wheel steering angle, the left and right rear wheel camber angles, and the anti-roll bar torque as control quantities, and determine the front wheel steering angle, the left and right rear wheel camber angles, and the anti-roll bar torque by solving the constrained online optimization problem.

[0203] The method provided by the present invention can actively coordinate and control the steering wheel angle, the wheel camber angle, and the anti-roll bar torque of the autonomous vehicle. When the autonomous vehicle is driving at high speed through a large-curvature bend or needs to quickly change lanes, it can provide additional lateral force for the vehicle through the active camber angle, and can reduce the roll angle of the vehicle through the active anti-roll bar, improving the steering ability and roll stability of the autonomous vehicle.

[0204] Such as Figure 4As shown, a schematic structural diagram of an embodiment of a vehicle steering control system provided by the present invention is shown. In combination with Figure 5 and Figure 6 shown, in this embodiment, the vehicle steering control system includes:

[0205] A kinetic model establishment module 10, configured to establish a three-degree-of-freedom vehicle kinetic model with an active anti-roll bar, where the three degrees of freedom include yaw, lateral, and roll directions;

[0206] A state space conversion module 11, configured to convert the three-degree-of-freedom vehicle kinetic model into a state space equation form;

[0207] A prediction processing module 12, configured to use a model predictive control algorithm to discretize the state space equation, and construct an objective function and constraint conditions with the goals of tracking the desired trajectory and minimizing the change in the control quantity, and obtain the optimal control quantity through rolling optimization, where the optimal control quantity includes the steering wheel angle, the vehicle rear wheel camber angle, and the anti-roll bar torque;

[0208] A control execution module 13, configured to perform steering control on the vehicle according to the optimal control quantity.

[0209] As Figure 5 shown, in an embodiment of the present invention, the kinetic model establishment module 10 includes:

[0210] A lateral equation establishment unit 110, configured to establish the following lateral motion equation:

[0211]

[0212] where a y is the lateral acceleration, m is the vehicle mass, m s is the sprung mass, h s is the distance from the vehicle center of mass to the roll center (h s =H - h RC ), H is the height of the sprung mass from the ground when no roll occurs, h RC is the height of the roll center from the ground, is the roll angular acceleration, F y1 F y2 、F y31 and F y4 are the lateral forces of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively;

[0213] where:

[0214] F y1 =C α1 α1,

[0215] Fy2 = C α2 α2,

[0216] F y3 = C α3 α3 + C γ3 γ3,

[0217] F y4 = C α4 α4 + C γ4 γ4

[0218] where α1, α2, α3, and α4 are the sideslip angles of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively, and C α1 , C α2 , C α3 and C α4 are the sideslip stiffnesses of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively; C γ3 and C γ3 are the camber stiffnesses of the left rear wheel and right rear wheel respectively; γ3 and γ4 are the camber angles of the left rear wheel and right rear wheel respectively;

[0219] The yaw equation establishment unit 111 is used to establish the following yaw motion equation:

[0220]

[0221] where I z is the yaw moment of inertia, is the yaw angular acceleration, l f and l r are the distances from the center of mass to the front axle and to the rear axle respectively;

[0222] The roll equation establishment unit 112 is used to establish the following roll motion equation:

[0223]

[0224] where, I x is the roll moment of inertia, I xs is the roll moment of inertia considering the influence of the sprung mass, is the roll angular acceleration, is the roll angle, is the roll angular velocity, M d is the active roll moment, and are the suspension roll stiffness and roll damping respectively.

[0225] More specifically, in the state space conversion module 11, the three-degree-of-freedom vehicle dynamics model is specifically converted into the following state space equation:

[0226]

[0227] Wherein:

[0228] x is the state vector,

[0229] u is the state vector, u = [δ f γ3 γ4 M d T

[0230] y is the output vector,

[0231] A is the system matrix, B is the input matrix, C is the output matrix, and E is the feedforward matrix, which are as follows:

[0232]

[0233] Wherein, a 13 = -(C α1 + C α2 )E f , a 14 = 0, a 23 = -l f E f (C α1 + C α2 ), a 24 = 0, a 31 = 0, a 32 = 0, a 33 = 0, a 34 = 1, a 41 = 0, a 42 = -m s h s v x , b 11 = -(C α1 + C α2 ), b 14 = 0, b 21 = -l f (C α1 + C α2 ), b 24 = 0, b 31 = 0, b 32 = 0, b 33 = 0, b 34 = 0, b 41 = 0, b 42 = 0, b​43 = 0, b 44 = -1;

[0234] As Figure 6 shown, in an embodiment of the present invention, the prediction processing module 12 includes:

[0235] A discrete processing unit 120, configured to discretize the state space equation to obtain a discretized state space equation:

[0236] A T = I + E -1 AT,

[0237] B T = E -1 BT;

[0238] wherein, A T is the transition matrix of the discrete state, I is the identity matrix, A is the system matrix, T is the sampling period, and E -1 is the inverse matrix of the feedforward matrix; B T is the input matrix of the discrete state, and B is the input matrix;

[0239] An algorithm construction unit 121, configured to use the model predictive control algorithm to predict the state trajectories {x(k + 1), x(k + 2), …, x(k + Np)} and output trajectories {y(k + 1), y(k + 2), …, y(k + Np)} of the next N p steps at each sampling moment by using the discretized state space equation. If the current state is x(k), the next prediction is x(k + 1) = A T x(k) + B T u(k);

[0240] A rolling optimization unit 122, configured to use the front wheel steering angle, the camber angles of the left and right rear wheels, and the lateral stabilizer bar torque as control variables, and perform rolling optimization by using the model predictive control algorithm to online solve the following online optimization problem with constraints to obtain the optimal control variables:

[0241] wherein, the objective function is:

[0242]

[0243] The constraint conditions are:

[0244] Δu min ≤ Δu(k + i) ≤ Δi max

[0245] u min ≤ u(k + i) ≤ u max

[0246] ymin y(k + i) ≤ y max

[0247] The reference output is:

[0248] y ref (k + i) = [0 ω d 0] T (i = 1, 2, …, N p )

[0249] Among them, where R is the reciprocal of the lane curvature, and y ref is the yaw rate.

[0250] As Figure 7 shown, in an embodiment of the present invention, the control execution module 13 includes:

[0251] A steering control unit 130, configured to apply the first element in the optimal control quantity sequence to the autonomous driving vehicle to achieve vehicle steering control;

[0252] A feedback correction unit 131, configured to perform feedback correction on the vehicle state after achieving vehicle steering control to adjust subsequent control quantities in real time.

[0253] For more details, reference can be made to and combined with the foregoing description of Figures 1 to 3 , and details will not be elaborated herein.

[0254] As yet another aspect of the present invention, there is also provided a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor module, the steps of the method described as Figures 1 to 3 are implemented. For more details, reference can be made to and combined with the foregoing description of Figures 1 to 3 , and details will not be elaborated herein.

[0255] As yet another aspect of the present invention, there is also provided a vehicle, which includes:

[0256] One or more processor modules;

[0257] A memory, configured to store one or more computer programs;

[0258] When the one or more computer programs are executed by the one or more processor modules, the one or more processor modules implement the method described as Figures 1 to 3 . For more details, reference can be made to and combined with the foregoing description of Figures 1 to 3 , and details will not be elaborated herein.

[0259] Implementing the embodiments of the present invention has the following beneficial effects:

[0260] The present invention relates to a vehicle steering control method, system, storage, and vehicle. Specifically, based on the three-degree-of-freedom vehicle dynamics model of the vehicle, by introducing key factors such as the camber stiffness of the left and right rear wheels, the camber angles of the left and right rear wheels, and the active roll moment, a new dynamics model is constructed. Subsequently, the established dynamics model is converted into a discretized state-space equation, and the front-wheel steering angle, the camber angles of the left and right rear wheels, and the anti-roll bar moment are set as control variables. By solving the constrained online optimization problem, the specific values of the front-wheel steering angle, the camber angles of the left and right rear wheels, and the anti-roll bar moment are accurately determined.

[0261] The method of the present invention applies the coordinated control strategy of three control variables, namely the steering wheel angle, the wheel camber angle, and the anti-roll bar moment, to autonomous vehicles. With the help of the active anti-roll bar, the roll angle of the vehicle can be effectively reduced, thereby expanding the boundary of the vehicle's steering ability. This method can not only significantly improve the steering performance of the vehicle, enabling it to smoothly pass through large-curvature curves and achieve rapid lane changes at high vehicle speeds, but also reduce the roll angle of the vehicle during steering, enhance the roll stability of the vehicle, and thus improve the riding comfort.

[0262] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor module of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor module of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0263] The above-disclosed is only a preferred embodiment of the present invention, and of course, it cannot be used to limit the scope of the rights of the present invention. Therefore, equivalent changes made according to the claims of the present invention still fall within the scope covered by the present invention.

Claims

1. A vehicle steering control method, characterized in that, Including the following steps: Establish a three-degree-of-freedom vehicle dynamics model with an active anti-roll bar, where the three degrees of freedom include yaw, lateral, and roll directions; Convert the three-degree-of-freedom vehicle dynamics model into the form of state-space equations; Adopt a model predictive control algorithm to discretize the state-space equations, and with the goals of tracking the desired trajectory and minimizing the change in control variables, construct an objective function and constraint conditions, and obtain the optimal control variables through rolling optimization, where the optimal control variables include steering wheel angle, vehicle rear wheel camber angle, and anti-roll bar torque; According to the optimal control variables, perform steering control on the vehicle.

2. The method according to claim 1, wherein Establish a control-oriented three-degree-of-freedom vehicle dynamics model, including the dynamic equations of the vehicle in the three degrees of freedom directions, including: Establish the following lateral motion equation: Among them, a y is the lateral acceleration, m is the vehicle mass, m s is the sprung mass, h s is the distance from the vehicle's center of mass to the roll center (h s = H - h RC ), H is the height of the sprung mass from the ground when no roll occurs, h RC is the height of the roll center from the ground, is the roll angular acceleration, F y1 F y2 、F y31 and F y4 are the lateral forces of the left front wheel, right front wheel, left rear wheel and right rear wheel respectively; Establish the following yaw motion equation: Among them, I z is the yaw moment of inertia, is the yaw angular acceleration, l f and l r are the distances from the center of mass to the front axle and to the rear axle respectively; Establish the following roll motion equation: Among them, I x is the roll moment of inertia, and I xs is the roll moment of inertia considering the influence of the sprung mass, is the roll angular acceleration, is the roll angle, is the roll angular velocity, M d is the active roll moment, and are the suspension roll stiffness and roll damping respectively; Where: F y1 = C α1 α1, F y2 = C α2 α2, F y3 = C α3 α3 + C γ3 γ3, F y4 = C α4 α4 + C γ4 γ4 Among them, α1, α2, α3, and α4 are the sideslip angles of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively, C α1 , C α2 , C α3 , and C α4 are the sideslip stiffnesses of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively; C γ3 , and C γ3 are the camber stiffnesses of the left rear wheel and right rear wheel respectively; γ3 and γ4 are the camber angles of the left rear wheel and right rear wheel respectively.

3. The method according to claim 2, wherein The conversion of the three-degree-of-freedom vehicle dynamics model into the form of state-space equations and the discretization process include: Convert the three-degree-of-freedom vehicle dynamics model into the following state-space equations: Where: x is the state vector, u is the state vector, u = [δ f γ3γ4M d T ​ y is the output vector, A is the system matrix, B is the input matrix, C is the output matrix, and E is the feedforward matrix, which are respectively as follows: Among them, a 13 = -(C α1 +C α2 )E f ,a 14 =0, a 23 =-l f E f (C α1 +C α2 ),a 24 =0,a 31 =0, a 32 = 0, a 33 = 0, a 34 = 1, a 41 = 0, a 42 = -m s h s v x , b 11 = -(C α1 + C α2 ), b 14 = 0, b 21 = -l f (C α1 + C α2 ) b 24 = 0, b 31 = 0, b 32 = 0, b 33 = 0, b 34 = 0, b 41 = 0, b 42 = 0, b 43 = 0, b 44 = -1 4. The method according to claim 3, wherein Adopt a model predictive control algorithm to discretize the state-space equations, and with the goals of tracking the desired trajectory and minimizing the change in control variables, construct an objective function and constraint conditions, and obtain the optimal control variables through rolling optimization, including: Discretize the state-space equations to obtain the discretized state-space equations: A T = I + E -1 AT, B T = E -1 BT; Among them, A T is the transition matrix of the discrete state, I is the identity matrix, A is the system matrix, T is the sampling period, E -1 is the inverse matrix of the feedforward matrix; B T is the input matrix of the discrete state, B is the input matrix; Using the model predictive control algorithm, the predicted state trajectories {x(k + 1), x(k + 2), …, x(k + Np)} and output trajectories {y(k + 1), y(k + 2), …, y(k + Np)} for the next N p steps are predicted at each sampling instant using the discretized state space equation. If the current state is x(k), the next prediction is x(k + 1) = A T x(k) + B T u(k); Where, taking the front wheel steering angle, left and right rear wheel camber angles, and anti-roll bar torque as control variables, adopt a model predictive control algorithm for rolling optimization, and online solve the following constrained online optimization problem to obtain the optimal control variables: Where, the objective function is: The constraint conditions are: Δu min ≤Δu(k + i)≤Δu max u min u(k + i) ≤ u max y mon ≤ y(k + i) ≤ y max The reference output quantity is: y ref (k + i) = [0ω d 0] T (i = 1, 2, …, N p ) Among them, where R is the reciprocal of the lane curvature, and y ref is the yaw rate.

5. The method according to claim 4, wherein According to the optimal control variables, perform steering control on the vehicle, including: Apply the first element in the optimal control variable sequence to the autonomous vehicle to achieve vehicle steering control; After achieving vehicle steering control, perform feedback correction on the vehicle state to adjust the subsequent control variables in real time.

6. A vehicle steering control system, characterized in that, Including: A dynamic model establishment module for establishing a three-degree-of-freedom vehicle dynamics model with an active anti-roll bar, where the three degrees of freedom include yaw, lateral, and roll directions; A state-space conversion module for converting the three-degree-of-freedom vehicle dynamics model into the form of state-space equations; A prediction processing module for adopting a model predictive control algorithm to discretize the state-space equations, and with the goals of tracking the desired trajectory and minimizing the change in control variables, construct an objective function and constraint conditions, and obtain the optimal control variables through rolling optimization, where the optimal control variables include steering wheel angle, vehicle rear wheel camber angle, and anti-roll bar torque; A control execution module for performing steering control on the vehicle according to the optimal control variables.

7. The system according to claim 6, wherein The dynamic model establishment module includes: A lateral equation establishment unit for establishing the following lateral motion equation: Wherein, a y is the lateral acceleration, m is the vehicle mass, m s is the sprung mass, h s is the distance from the vehicle's center of mass to the roll center (h s = H - h RC ), H is the height of the sprung mass from the ground when there is no roll, h RC is the height of the roll center from the ground, is the roll angular acceleration, F y1 F y2 、F y31 and F y4 are the lateral forces of the left front wheel, right front wheel, left rear wheel and right rear wheel respectively; Where: F y1 = C α1 α1, F y2 = C α2 α2, F y3 = C α3 α3 + C γ3 γ3, F y4 = C α4 α4 + C γ4 γ4 Among them, α1, α2, α3, and α4 are the sideslip angles of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively, C α1 , C α2 , C α3 , and C α4 are the sideslip stiffnesses of the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively; C γ3 , and C γ3 are the camber stiffnesses of the left rear wheel and right rear wheel respectively; γ3 and γ4 are the camber angles of the left rear wheel and right rear wheel respectively; A yaw equation establishment unit for establishing the following yaw motion equation: Among them, I z is the yaw moment of inertia, is the yaw angular acceleration, l f and l r are the distances from the center of mass to the front axle and to the rear axle, respectively; A roll equation establishment unit for establishing the following roll motion equation: Among them, I x is the roll moment of inertia, and I xs is the roll moment of inertia considering the influence of the sprung mass, is the roll angular acceleration, is the roll angle, is the roll angular velocity, and M d is the active roll moment, and are the suspension roll stiffness and roll damping respectively.

8. The system according to claim 7, wherein In the state space conversion module, specifically, a three-degree-of-freedom vehicle dynamics model is converted into the following state space equation: Where: x is the state vector, u is the state vector, u = [δ f γ3γ4M d T ​ y is the output vector, A is the system matrix, B is the input matrix, C is the output matrix, and E is the feedforward matrix, which are respectively as follows: Among them, a 13 = -(C α1 +C α2 )E f ,a 14 =0, a 23 =-l f E f (C α1 +C α2 ),a 24 =0,a 31 =0, a 32 = 0, a 33 = 0, a 34 = 1, a 41 = 0, a 42 = -m s h s v x , b 11 = -(C α1 + C α2 ), b 14 = 0, b 21 = -l f (C α1 +C α2 ) b 24 = 0, b 31 = 0, b 32 = 0, b 33 = 0, b 34 = 0, b 41 = 0, b 42 = 0, b 43 = 0, b 44 = -1 9. The system according to claim 8, wherein The prediction processing module includes: A discretization processing unit for discretizing the state space equation to obtain a discretized state space equation: A T = I + E -1 AT, B T = E -1 BT; Among them, A T is the transition matrix of discrete states, I is the identity matrix, A is the system matrix, T is the sampling period, E -1 is the inverse matrix of the feedforward matrix; B T is the input matrix of discrete states, B is the input matrix; An algorithm construction unit, which uses a model predictive control algorithm to predict the state trajectories {x(k + 1), x(k + 2), …, x(k + Np)} and output trajectories {y(k + 1), y(k + 2), …, y(k + Np)} of the next N p steps using the discretized state space equation at each sampling moment. If the current state is x(k), the next prediction is x(k + 1) = A T x(k) + B T u(k); A rolling optimization unit for using the front wheel steering angle, the camber angles of the left and right rear wheels, and the lateral stabilizer bar torque as control variables, and performing rolling optimization using a model predictive control algorithm to online solve the following constrained online optimization problem to obtain optimal control variables: Where the objective function is: The constraint conditions are: Δu min ≤Δu(k + i)≤Δu max u min u(k + i) ≤ u max y min ≤ y(k + i) ≤ y max The reference output quantity is: y ref (k + i) = [0ω d 0] T (i = 1, 2, …, N p ) Among them, where R is the reciprocal of the lane curvature, and y ref is the yaw rate.

10. The system according to claim 9, characterized in that, The control execution module includes: A steering control unit for applying the first element in the optimal control variable sequence to the autonomous vehicle to achieve vehicle steering control; A feedback correction unit for performing feedback correction on the vehicle state after achieving vehicle steering control to adjust subsequent control variables in real time.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor module, the steps of the method according to any one of claims 1 to 5 are implemented.

12. A vehicle, characterized in that, Including: One or more processor modules; A memory for storing one or more computer programs; When the one or more computer programs are executed by the one or more processor modules, the one or more processor modules implement the method according to any one of claims 1 to 5.

Images