CN113306545A - Vehicle trajectory tracking control method and system - Google Patents

Abstract

The invention relates to a vehicle trajectory tracking control method and system, which introduces a UniTire tire model with high precision in all working conditions, and embeds a linear time-varying MPC control algorithm to realize trajectory tracking, further expanding the adaptive scenarios of intelligent vehicle trajectory tracking control such as high speed , low adhesion road surface, large slip rate, etc., and improve tracking performance. At the same time, by using the numerical calculation method to calculate the local linearization process of the model, the computational complexity is reduced without affecting the calculation accuracy, and the desired trajectory can still be tracked and the vehicle can be stabilized under the condition of large tire slip rate.

Description

A vehicle trajectory tracking control method and system

technical field

The invention relates to the technical field of intelligent vehicle control, in particular to a vehicle trajectory tracking control method and system.

Background technique

Tires are the only contact parts between the vehicle and the ground. The tire dynamics model is closely related to the vehicle dynamics model and vehicle stability control. The accuracy of the tire model will directly affect the accuracy of vehicle control and the safety of driving.

The research of smart cars is usually divided into three aspects: perception, planning, and vehicle control. The control of smart cars usually refers to the tracking of the desired trajectory calculated by the perception layer and the planning layer. Trajectory tracking includes the tracking of the desired path and expected speed. At the same time, appropriate constraints are established to ensure the stability of the vehicle. Intelligent vehicle trajectory tracking is one of the key technologies in intelligent vehicle research, and there are many control methods that can be implemented.

However, in the existing vehicle trajectory tracking control, the tire model used often requires complex numerical operations, which leads to the simplification of tire characteristics in order to reduce the complexity of the solution, which affects the robustness of the control system.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a vehicle trajectory tracking control method and system, which improves the control accuracy by applying the UniTire tire model to the vehicle trajectory tracking control.

For achieving the above object, the present invention provides the following scheme:

A vehicle trajectory tracking control method, the method comprising:

Obtain the lateral control parameters and longitudinal control parameters of the vehicle;

Obtain the yaw moment according to the lateral control parameters combined with the UniTire tire model;

Obtaining the total longitudinal force through the sliding mode controller according to the longitudinal control parameter;

The control torque of the vehicle is obtained according to the yaw moment and the total longitudinal force, and the trajectory tracking control of the vehicle is realized according to the control torque.

Optionally, the obtaining of the yaw moment according to the lateral control parameters in combination with the UniTire tire model includes:

Build the output equation according to the slip angle formula in the UniTire tire model;

establishing a nonlinear state equation according to the lateral control parameters and the vehicle seven-degree-of-freedom model;

After discretizing and linearizing the output equation and the nonlinear state equation, a Jacobian matrix is obtained;

Build a first objective function according to the Jacobian matrix, and organize the first objective function to obtain a quadratic objective function;

The variation of the yaw moment when the quadratic objective function is the smallest is solved, and the yaw moment at the next moment is obtained according to the variation of the yaw moment.

Optionally, after obtaining the Jacobian matrix, also include:

The Jacobian matrix is calculated by a numerical derivation method to obtain a deformed Jacobian matrix, and the Jacobian matrix in the process of obtaining the quadratic objective function is replaced with the deformed Jacobian matrix.

Optionally, the first objective function includes:

为系统输出和参考输出的差值，

为横摆角，

为横摆角速度，Y为质心相对于大地坐标系的纵坐标，ξ为状态变量，t表示时刻，μ表示控制变量，ΔU(t)表示控制量的变化量，ρ为权重系数，ε表示松弛因子，

表示期望横摆角，

表示期望横摆角速度，Y_ref表示期望纵坐标，H_p为预测时域，H_c为控制时域，Q、R、S均为权重矩阵。in,

is the difference between the system output and the reference output,

is the yaw angle,

is the yaw rate, Y is the ordinate of the center of mass relative to the geodetic coordinate system, ξ is the state variable, t is the time, μ is the control variable, ΔU(t) is the change of the control variable, ρ is the weight coefficient, ε is the relaxation factor,

represents the desired yaw angle,

represents the desired yaw rate, _Yref represents the desired ordinate, _Hp is the prediction time domain, _Hc is the control time domain, and Q, R, and S are weight matrices.

Optionally, the obtaining the total longitudinal force through the sliding mode controller according to the longitudinal control parameter includes:

Build the switching function of the sliding mode controller;

Combining the switching function with the vehicle longitudinal dynamics equation yields the total longitudinal force.

Optionally, the obtaining the control moment of the vehicle according to the yaw moment and the total longitudinal force includes:

Construct the second objective function;

establishing constraints based on the yaw moment and the total longitudinal force;

Under the constraint conditions, solve the tire longitudinal force change amount when the second objective function is the smallest, and obtain the tire longitudinal force at the next moment according to the tire longitudinal force change amount;

The control torque is obtained according to the tire longitudinal force at the next moment in combination with the wheel dynamics model.

Optionally, when the yaw moment is obtained according to the lateral control parameter in combination with the UniTire tire model, the method further includes:

Obtain the front wheel turning angle according to the lateral control parameters combined with the UniTire tire model;

The trajectory tracking control of the vehicle is realized according to the front wheel rotation angle and the control torque.

Optionally, implementing the trajectory tracking control of the vehicle according to the front wheel angle and the control torque includes:

controlling the steering gear of the vehicle according to the front wheel angle;

Each wheel of the vehicle is independently controlled according to the control torque.

A vehicle trajectory tracking control system for implementing the above method, the system comprising: a lateral controller and a sliding mode controller;

The lateral controller is used to calculate the yaw moment according to the lateral control parameters; the lateral controller includes a linear time-varying model predictive controller and a UniTire tire model, and the UniTire tire model is used to provide a Jacobian matrix to the linear time-varying model participating in the calculation of the yaw moment in the model predictive controller;

The sliding mode controller is used to calculate the total longitudinal force according to the longitudinal control parameter;

The lateral controller and the sliding mode controller are respectively connected with the distribution controller;

The distribution controller is configured to receive the yaw moment and the total longitudinal force, and obtain the tire longitudinal force according to the yaw moment and the total longitudinal force;

The distribution controller is connected with a drive braking system; the drive braking system is used to receive the tire longitudinal force, obtain a control torque according to the tire longitudinal force, and realize the trajectory tracking of the vehicle according to the control torque control.

According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

The invention provides a vehicle trajectory tracking control method and system. By introducing the UniTire tire model, the complex lateral and longitudinal coupling characteristics are considered, and because the UniTire tire model has dimensionless force characteristic expression, theoretical model boundary conditions and other modeling characteristics , so that the trajectory tracking control of the vehicle has a higher global identification accuracy.

In the specific embodiment of the present invention, by using the numerical calculation method to calculate the process of local linearization of the model, the calculation complexity is reduced without affecting the calculation accuracy, and the desired trajectory can still be tracked under the condition of large tire slip rate And control the vehicle stability.

Description of drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings required in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some of the present invention. In the embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without any creative effort.

1 is a system block diagram of a vehicle trajectory tracking control system provided in Embodiment 1 of the present invention;

2 is a method flowchart of a vehicle trajectory tracking control method provided in Embodiment 2 of the present invention;

3 is a schematic diagram of a vehicle seven-degree-of-freedom model provided in Embodiment 2 of the present invention;

Figure 4(a) is a comparison diagram of the lateral position results of using two tire models in scenario 1 provided by Embodiment 2 of the present invention; Figure 4(b) is a scenario 1 provided by Embodiment 2 of the present invention using two types of tires The comparison chart of the yaw angle results of the model;

FIG. 5 is a comparison diagram of the results of using two tire models in Scenario 2 provided by Embodiment 2 of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

The purpose of the present invention is to provide a vehicle trajectory tracking control method and system, which has higher tracking control accuracy, and can track a desired trajectory and ensure vehicle stability even under a large tire slip rate.

In order to make the above objects, features and advantages of the present invention more clearly understood, the present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.

Example 1

The trajectory tracking control of intelligent vehicle is from the earliest proposed preview theory and driver model, which tracks the preview point by simulating the driving behavior of the driver, to the more advanced control methods applied to vehicle control, such as sliding mode control, linear matrix inequality, Neural network control, model predictive control, etc. Among them, model predictive control is good at dealing with multiple coupling problems with explicit constraints, and can better adapt to multiple free vehicle models and complex UniTire tire models. In addition, the existence of prediction, optimization, and feedback links and the solution of optimal control problems ensure that the Robustness of the algorithm and stationarity of the control.

The tire model is a highly nonlinear model. The tire models often used in vehicle trajectory tracking control usually include the following: The first is the simplest linear tire model, which can simplify the operation to the greatest extent, but the linear tire model It can only approximate the tire characteristics in the linear region of pure working conditions. It cannot describe the coupling characteristics under the compound conditions of cornering, longitudinal slip and roll at the same time and the nonlinear characteristics when the slip rate is large, which greatly limits the control. The application scenario of the algorithm; the second is a simple empirical model, the more commonly used is the Burckhardt tire model, this model can express the nonlinear tire characteristics to a certain extent, and the road adhesion coefficient is introduced into the model, but due to its parameters Too few, the expression accuracy under nonlinear and complex working conditions is very limited; the third is the complex tire model, the most widely used is the magic formula proposed by Pacejka, the magic formula includes multiple versions, but in vehicle control due to the complexity of the need In order to reduce the complexity of the solution, a simplified version of the mechanical characteristics of the composite working conditions is often used. The error caused by the simplification of the tire characteristics is compensated by the robustness of the control system.

Based on this, this embodiment provides a vehicle trajectory tracking control system. As shown in FIG. 1 , the system includes: a lateral controller and a sliding mode controller;

The lateral controller is used to calculate the yaw moment according to the lateral control parameters; the lateral controller includes a linear time-varying model predictive controller and a UniTire tire model, and the UniTire tire model is used to provide a Jacobian matrix to the linear time-varying model participating in the calculation of the yaw moment in the model predictive controller;

The sliding mode controller is used to calculate the total longitudinal force according to the longitudinal control parameter;

The lateral controller and the sliding mode controller are respectively connected with the distribution controller;

The distribution controller is configured to receive the yaw moment and the total longitudinal force, and obtain the tire longitudinal force according to the yaw moment and the total longitudinal force;

The distribution controller is connected with a drive braking system; the drive braking system is used to receive the tire longitudinal force, obtain a control torque according to the tire longitudinal force, and realize the trajectory tracking of the vehicle according to the control torque control.

As an optional implementation manner, the lateral controller is connected to the steering gear of the vehicle, and the lateral controller is used to output the rotation angle of the front wheels to control the rotation angle of the front wheels of the vehicle, and then combine the control torque and the rotation angle of the front wheels Jointly realize the trajectory tracking control of the vehicle.

Example 2

This embodiment provides a vehicle trajectory tracking control method, using the system described in Embodiment 1, as shown in FIG. 2 , the method includes:

Step 101: Acquire lateral control parameters and longitudinal control parameters of the vehicle;

Step 102: obtain the yaw moment according to the lateral control parameters combined with the UniTire tire model;

Step 103: obtain the total longitudinal force through the sliding mode controller according to the longitudinal control parameters;

Step 104: Obtain a control torque of the vehicle according to the yaw moment and the total longitudinal force, and implement trajectory tracking control of the vehicle according to the control torque.

In the following, the present embodiment will be specifically described by taking the trajectory tracking control applied to the four-wheel independent drive electric vehicle as an example.

期望横摆角速度

和期望纵坐标Y_ref，以及车辆当前的质心侧向速度v_y、横摆角

横摆角速度

质心相对于大地坐标系的横坐标X和纵坐标Y、质心纵向加速度a_x、质心纵向加速度a_y和车轮转动速度ω_i。When the trajectory tracking control of the four-wheel independent drive electric vehicle is performed, the path tracking and speed tracking are realized by controlling the four-wheel control torque T _i and the front wheel angle δ. The target vehicle to be controlled is an electric vehicle with four-wheel independent driving/braking and front wheel steering, so the four wheels of the target vehicle can be independently controlled, assuming that the front wheels of the vehicle have the same turning angle. The trajectory tracking controller proposed in this embodiment decouples the longitudinal and lateral vehicle dynamics. The lateral tracking control adopts linear time-varying model predictive control, and the lateral control parameters are input into a model predictive controller (Model Predictive Control, MPC), and the yaw moment and front wheel rotation angle are obtained according to the lateral control parameters combined with the UniTire tire model. The lateral control parameters include: desired yaw angle

Desired yaw rate

and the desired ordinate Y _ref , and the vehicle's current center of mass lateral velocity _vy , yaw angle

yaw rate

The abscissa X and ordinate Y of the center of mass relative to the geodetic coordinate system, the longitudinal acceleration of the center of mass a _x , the longitudinal acceleration of the center of mass a _y and the rotational speed of the wheel ω _i .

The longitudinal control adopts the sliding mode controller SMC, the longitudinal control parameters are input into the sliding mode controller SMC, and the total longitudinal force F _X is obtained according to the longitudinal control parameters. The longitudinal control parameters include: the desired vehicle speed V _xref , and the current longitudinal acceleration a _x and longitudinal velocity V _x of the center of mass of the vehicle.

Taking the yaw moment and the total longitudinal force as the input of the distribution controller of the lower layer, the quadratic programming is used to distribute the input in the optimal form to obtain the tire longitudinal force F _i , and the independent control of the four wheels is obtained according to the wheel dynamics. The torque _Ti , the control torque _Ti is the driving/braking torque _Ti , and the vehicle is directly controlled by the obtained front wheel angle and the driving/braking torque of the four wheels. Among them, the steering angle of the front wheel controls the steering gear of the vehicle, that is, the steering angle of the front wheel of the electric vehicle driven by the four-wheel motor that controls the independent driving/braking of the four wheels and the steering of the front wheels; the driving/braking torque control of each wheel is applied to the The driving force of the wheel, so as to realize the control of the target vehicle.

According to the balance principle of force and moment, the vehicle body coordinate system is used as the reference coordinate system, and the vehicle seven-degree-of-freedom model applied in the MPC controller is given, as shown in Fig. to two degrees of freedom.

In the vehicle 7-DOF model, the lateral motion equation is:

Yaw motion equation:

in,

和

为横摆角速度和横摆角加速度，l_a和l_b为质心至前/后轴距离，c为轮距。In the formula, Mz combines all the items including the longitudinal force in the yaw motion into one, and its physical meaning is the yaw moment generated by the longitudinal force of the wheel; and the Y-axis component generated by the longitudinal force in the lateral motion is small, so the prediction time Longitudinal forces in the domain are treated approximately as constants. In addition, δ is the front wheel rotation angle, F _xi and F _yi represent the tire longitudinal force and lateral force, respectively, i=1, 2, 3, 4 represent the left front, right front, left rear, and right rear four wheels of the vehicle, respectively, v _x and v _y are the longitudinal and lateral speeds of the center of mass, J _Z is the moment of inertia of the vehicle around the z-axis,

and

are the yaw angular velocity and yaw angular acceleration, l _a and l _b are the distance from the center of mass to the front/rear axle, and c is the wheel base.

The vehicle seven-degree-of-freedom model is established with the body coordinate system as the reference system. When the reference system of the vehicle is transformed into the geodetic coordinate system, according to the conversion principle, the longitudinal and lateral speeds of the vehicle's center of mass relative to the geodetic coordinate system can be expressed as:

At the same time, the yaw moment is calculated in combination with the full-condition UniTire tire model in the lateral control, and the lateral force in the full-condition UniTire tire model can be expressed as the following functional form:

F _yi =f _UniTire (α _i ,κ _i ,V _i ,F _zi )

Among them, the vertical tire load is:

where H is the height of the center of mass of the vehicle, L is the wheelbase of the vehicle, and a _x and a _y are the longitudinal and lateral accelerations of the center of mass of the vehicle.

The tire slip angle is:

In the geodetic coordinate system, the speed of the wheel center relative to the ground can be expressed as:

In the tire coordinate system, the longitudinal speed of the wheel center is:

The longitudinal slip rate is:

In the formula, ω _i is the wheel rolling angular velocity, and R _ie is the tire rolling radius.

控制变量μ＝[Mz δ]^T，并以轮胎侧偏角为约束条件，

为输出变量，根据七自由度车辆模型中的侧向运动、横摆运动两个自由度以及车辆质心相对于大地坐标系的纵向和侧向速度公式建立非线性状态方程，根据轮胎侧偏角公式建立输出方程，所述非线性状态方程和所述输出方程为：select state variable

The control variable μ=[Mz δ] ^T , and the tire slip angle is the constraint condition,

As the output variable, a nonlinear state equation is established according to the two degrees of freedom of lateral motion and yaw motion in the seven-degree-of-freedom vehicle model, and the longitudinal and lateral velocity formulas of the center of mass of the vehicle relative to the geodetic coordinate system. According to the tire slip angle formula The output equation is established, the nonlinear state equation and the output equation are:

in,

A linear time-varying system is obtained after discretizing and linearizing the nonlinear state equation and the output equation. In order to ensure the computational efficiency of the algorithm and reduce the complexity of solving the optimization function, in this embodiment, the seven-degree-of-freedom vehicle model used in the model predictive control is linearized, and the nonlinear model predictive control is transformed into a linear time-varying model prediction. Control problems, avoiding complex solutions for nonlinear model predictions. The local linearization of the system is achieved by ignoring the high-order items by taking the state and control variables of the system at the current moment as reference points to perform first-order Taylor expansion, and after discretization, the linear time-varying state equation and output equation are obtained:

ξ _k+1,t =A _k,t ξ _k,t +B _k,t μ _k,t +d _k,t , k=t,...,t+H _p -1

η _k,t =C _k,t ξ _k,t +D _k,t μ _k,t +e _k,t , k=t,...,t+H _p

in,

In order to reduce the complexity, the linear time-varying system is simplified according to the following rules:

A _k,t =A _t,t ,B _k,t =B _t,t ,d _k,t =d _t,t

C _k,t =C _t,t ,D _k,t =D _t,t ,e _k,t =e _t,t

Among them, A _t and B _t are the Jacobian matrices of the nonlinear state space equation relative to the state quantity and control quantity. For general nonlinear systems, the partial derivative solving function jacobian in Matlab can be used to calculate. However, considering that UniTire, as a semi-empirical tire model, contains empirical expressions to adapt to the test data, the above analytical method is still used to solve it. The resulting expression data is huge and the optimization function cannot be solved correctly. Therefore, this embodiment adopts the numerical derivation method to calculate A _t and B _t .

When performing numerical derivation, set the current moment as k, and the vehicle state at moment k-1 can be expressed as:

v _y (k-1)= _vy (k)-T·f ₁ (ξ(k),μ(k))

According to the definition of partial derivative, let the tire slip angle at time k-1 be:

In the same way, the wheel center velocity at time k-1 is:

The tire longitudinal slip rate at time k-1 is:

The tire lateral force at time k-1 is:

F _yi,δ (k-1)=f _UniTire (α _i,δ (k-1),κ _i,δ (k-1),V _i ,F _zi ), i=1, 2k-1 time vehicle mass center Lateral velocity and yaw velocity are:

According to the state of the vehicle at time k-1 and the current state of the vehicle (at time k), the differential method of the Jacobian matrix can be obtained, thereby replacing the traditional derivation method, and the method of directly calculating the Jacobian matrix is used to ensure the tire model. On the basis of the accuracy, the solution complexity caused by the nonlinear model is reduced to the greatest extent.

Finally, A _t and B _t calculated by numerical approximation are obtained:

The first objective function is constructed as follows:

为系统输出和参考输出的差值；第二项为控制量的变化量，对该项的惩罚代表了控制变化量平稳的要求；第三项为控制量，对该项的惩罚代表了控制量最大最小值的要求；第四项为松弛因子，该项扩大了软约束输出量的约束范围，使得可以用第一目标函数次优解代替最优解。H_p为预测时域，H_c为控制时域。Among them, the first

is the difference between the system output and the reference output; the second item is the variation of the control quantity, and the penalty for this item represents the requirement for a stable control variation; the third item is the control quantity, and the penalty for this item represents the control quantity The requirements of the maximum and minimum values; the fourth item is the relaxation factor, which expands the constraint range of the soft constraint output, so that the suboptimal solution of the first objective function can be used to replace the optimal solution. H _p is the prediction time domain, and H _c is the control time domain.

In the fourth item, ρ is the weight coefficient, which represents the penalty for each item. The larger the value, the greater the impact of the item on the objective function; ε represents the relaxation factor.

After sorting out the first objective function, we can get:

Wherein, Θ _trt =Υ _tr Θ _t ; Υ _tr is the block matrix to decompose the output variable;

和

为前文求解的雅可比矩阵A_t和B_t的变形。

and

Variations of the Jacobian matrices A _t and B _t solved for above.

Make the following settings:

G _t =[2σ(t) ^T Q _e Θ _trt +2U(t-1) ^T S _e M 0]

P _t =σ(t) ^T Q _e σ(t)+U(t-1) ^T S _e U(t-1)+ρε ²

Arranged into a standard quadratic objective function:

其中Δμ_t＝[Δδ_t,ΔM_Zt]，进一步得到下一时刻控制量μ(t|t)＝μ(t-1|t)+Δμ_t，前轮转角δ_exp直接作用于车辆，横摆力矩M_Zexp给到分配控制器进行轮胎力矩的分配。Solve for the control variable change that minimizes the function

where Δμ _t =[Δδ _t ,ΔM _Zt ], and further obtain the next moment control amount μ(t|t)=μ(t-1|t)+Δμ _t , the front wheel angle δ _exp acts directly on the vehicle, the yaw The torque M _{Zexp is} given to the distribution controller to distribute the tire torque.

Longitudinal control adopts sliding mode controller, first define the switching function:

s=V _x -V _xd +b∫(V _x -V _xd )dt

Choose an approach rate:

in,

联立可得期望总纵向力：According to the vehicle longitudinal dynamics equation

Simultaneously obtains the expected total longitudinal force:

The distribution controller distributes the tire longitudinal force according to the expected total yaw moment M _Zexp and the total longitudinal force F _Xexp , and firstly constructs the second objective function:

Among them: R ₁ , Q ₁ are weighting matrices; ρ ₁ , ρ ₂ are weighting coefficients; ΔF=[ΔF _x1 ,ΔF _x2 ,ΔF _x3 ,ΔF _x4 ] ^T , F=[F _x1 ,F _x2 ,F _x3 ,F _x4 ] ^T .

Write the quadratic programming problem in standard form:

In the formula: X=[ΔF ^T , ε ₁ , ε ₂ ] ^T , H=[R ₁ +Q ₁ , 0 _4×2 ; 0 _2×4 , ρ ₁ , ρ ₂ ],

Restrictions:

where:

lb=[lb ₁ ,lb ₂ ,lb ₃ ,lb ₄ ,0,0] ^T

ub=[ub ₁ ,ub ₂ ,ub ₃ ,ub ₄ ,ε _1max ,ε _2max ] ^T

得到驱动/制动力矩T_i作用于车辆，其中J_wi为每个车轮转动惯量；R_i为轮胎有效滚动半径；w_i为每个车轮转动速度；B_i为粘滞阻力系数。Solve the tire longitudinal force change ΔF when the function is minimized, and further obtain the tire longitudinal force at the next moment. According to the wheel dynamics

The driving/braking torque Ti acts on the vehicle, where _Jwi is the moment of inertia of each wheel; Ri is the effective rolling radius of the tire; _{wi is} the rotational speed of each wheel; B _i is the viscous resistance coefficient _.

Therefore, this embodiment introduces the UniTire tire model with high precision in all working conditions, and embeds the linear time-varying MPC control algorithm to realize trajectory tracking, further expanding the adaptive scenarios of intelligent vehicle trajectory tracking control, such as high speed, low adhesion road, large slippery shift rate, etc., and improve tracking performance. The UniTire tire model takes into account the complex lateral and longitudinal coupling characteristics. The UniTire model used in this embodiment has similar expression capabilities to the complex magic formula, but because of its dimensionless expression of force characteristics, dynamic friction coefficients, theoretical model boundary conditions, etc. Modeling characteristics, UniTire model not only has higher global identification accuracy, but also better model expansion ability. Therefore, the vehicle trajectory tracking control method based on the UniTire tire model provided in this embodiment can improve the trajectory tracking accuracy in a high-speed scenario and the vehicle stability in a large slip rate. In addition, the speed prediction, load prediction, and dynamic friction coefficient expression of the tire model itself can also improve the adaptability and expansibility of vehicle control to vehicle speed, vehicle type and tire type, and road conditions.

This embodiment simplifies the process of local linearization of the model in the control algorithm, and approximates it by using a numerical calculation method, avoiding complex formula derivation and complex numerical solution. Compared with the simplified tire model method, the designed control algorithm has higher tracking accuracy in high-speed scenarios, and can still track the desired trajectory and ensure the stability of the vehicle even under large tire slip rates.

In order to further prove the effect of this embodiment, this embodiment also builds a Matlab/Simulink and Carsim co-simulation platform to conduct trajectory tracking simulation verification of road conditions with different speeds and different adhesion coefficients, and compares the tire models using Pacejka5.2 and UniTire tracking accuracy.

Scenario 1: The road adhesion coefficient is 0.8, and the vehicle speed is 90km/h. As shown in Figure 4, (a) is the comparison chart of the lateral position results of the two models, and (b) is the comparison chart of the yaw angle results of the two models. It can be seen from the figure that using the UniTire tire model, the absolute value of the lateral position deviation is the largest The value is 3.06m, and the root mean square value of the lateral position deviation is 1.06m; using the PAC5.2 tire model, the maximum value is 3.21m, and the root mean square value is 1.13m. Using the UniTire tire model, the maximum absolute value of the yaw angle deviation is 9.16°, and the root mean square value of the yaw angle deviation is 3.55°; using the PAC5.2 tire model, the maximum value is 10.28°, and the root mean square value is 3.89 °.

Scenario 2: The road adhesion coefficient is 0.4, and the vehicle speed is 70km/h. As shown in Figure 5, (a) is the comparison chart of the lateral position results of the two models, and (b) is the comparison chart of the yaw angle results of the two models. It can be seen from the figure that the UniTire tire model is used, and the absolute value of the lateral position deviation is the largest The value is 2.45m, and the root mean square value of the lateral position deviation is 0.85m; using the PAC5.2 tire model, the maximum value is 2.50m, and the root mean square value is 0.90m. Using the UniTire tire model, the maximum absolute value of the yaw angle deviation is 9.78°, and the root mean square value of the yaw angle deviation is 3.67°; using the PAC5.2 tire model, the maximum absolute value of the yaw angle deviation is 9.34°, The root mean square value of the yaw angle deviation is 4.02°.

It can be seen that the performance of the vehicle trajectory tracking control using the UniTire tire model in this embodiment is significantly better than that of the solution using the PAC5.2 tire model, and the accuracy of the vehicle trajectory tracking control is improved.

Each embodiment in this specification focuses on the points that are different from other embodiments, and the same and similar parts between the various embodiments can be referred to each other.

In this paper, specific examples are used to illustrate the principles and implementations of the present invention. The descriptions of the above embodiments are only used to help understand the methods and core ideas of the present invention; meanwhile, for those skilled in the art, according to the present invention There will be changes in the specific implementation and application scope. In conclusion, the contents of this specification should not be construed as limiting the present invention.

Claims (10)

1. A vehicle trajectory tracking control method, characterized by comprising:

acquiring transverse control parameters and longitudinal control parameters of a vehicle;

obtaining a yaw moment by combining the transverse control parameters with a UniTire tire model;

obtaining a total longitudinal force through a sliding mode controller according to the longitudinal control parameters;

and obtaining the control moment of the vehicle according to the yaw moment and the total longitudinal force, and realizing the track tracking control of the vehicle according to the control moment.

2. The vehicle trajectory tracking control method of claim 1, wherein the deriving a yaw moment from the lateral control parameters in combination with a UniTire tire model comprises:

establishing an output equation according to a slip angle formula in a UniTire tire model;

establishing a nonlinear state equation according to the transverse control parameters and a seven-degree-of-freedom model of the vehicle;

discretizing and linearizing the output equation and the nonlinear state equation to obtain a Jacobian matrix;

constructing a first objective function according to the Jacobian matrix, and sorting the first objective function to obtain a quadratic objective function;

and solving the yaw moment variation when the quadratic form objective function is minimum, and obtaining the yaw moment at the next moment according to the yaw moment variation.

3. The vehicle trajectory tracking control method according to claim 2, further comprising, after obtaining the jacobian matrix:

and calculating the Jacobian matrix by using a numerical derivation method to obtain a deformed Jacobian matrix, and replacing the Jacobian matrix in the quadratic objective function acquisition process with the deformed Jacobian matrix.

4. The vehicle trajectory tracking control method according to claim 2, wherein the first objective function includes:

wherein,

is the difference between the system output and the reference output,

is the yaw angle of the vehicle,

is yaw rate, Y is ordinate of centroid relative to geodetic coordinate system, xi is state variable, t is time, mu is control variable, delta U (t) is variation of control quantity, rho is weight coefficient, epsilon is relaxation factor,

indicating that the desired yaw angle is desired,

indicating the desired yaw rate, Y_refDenotes the desired ordinate, H_pTo predict the time domain, H_cQ, R, S are weight matrices for the control time domain.

5. A vehicle trajectory tracking control method according to claim 1, wherein said deriving a total longitudinal force by a sliding mode controller in accordance with said longitudinal control parameter comprises:

constructing a switching function of the sliding mode controller;

and obtaining the total longitudinal force by combining the switching function and the vehicle longitudinal dynamic equation.

6. A vehicle trajectory tracking control method according to claim 1, wherein said deriving a control moment of the vehicle from said yaw moment and said total longitudinal force comprises:

constructing a second objective function;

establishing a constraint condition according to the yaw moment and the total longitudinal force;

solving the tire longitudinal force variation when the second objective function is minimum under the constraint condition, and obtaining the tire longitudinal force at the next moment according to the tire longitudinal force variation;

and obtaining a control moment according to the tire longitudinal force at the next moment and a wheel dynamic model.

7. The vehicle trajectory tracking control method of claim 1, wherein when a yaw moment is obtained in combination with a UniTire tire model according to the lateral control parameters, the method further comprises:

obtaining a front wheel corner by combining the transverse control parameter with a UniTire tire model;

and realizing the track tracking control of the vehicle according to the front wheel rotation angle and the control torque.

8. The vehicle trajectory tracking control method according to claim 7, wherein the performing trajectory tracking control of the vehicle based on the front wheel rotation angle and the control torque includes:

controlling a steering of the vehicle according to the front wheel steering angle;

and independently controlling each wheel of the vehicle according to the control torque.

9. A vehicle trajectory tracking control system for implementing the method of claim 1, the system comprising: a lateral controller and a sliding mode controller;

the transverse controller is used for calculating a yaw moment according to the transverse control parameters; the lateral controller comprises a linear time-varying model predictive controller and a UniTire tire model for providing a Jacobian matrix to the linear time-varying model predictive controller to participate in the calculation of the yaw moment;

the sliding mode controller is used for calculating total longitudinal force according to longitudinal control parameters;

the transverse controller and the sliding mode controller are respectively connected with the distribution controller;

the distribution controller is used for receiving the yaw moment and the total longitudinal force and obtaining a tire longitudinal force according to the yaw moment and the total longitudinal force;

the distribution controller is connected with a driving brake system; the driving and braking system is used for receiving the longitudinal force of the tire, obtaining a control moment according to the longitudinal force of the tire, and realizing the track tracking control of the vehicle according to the control moment.

10. A vehicle tracking control system according to claim 9, wherein the lateral controller is connected to a steering gear of the vehicle, and the lateral controller is configured to output a front wheel steering angle to control a steering angle of the front wheels of the vehicle.

Images