CN112051851A - Autonomous drift control method and system for electric four-wheel drive vehicle under limit working condition - Google Patents

Abstract

The invention relates to an autonomous drift control method of an electric four-wheel drive vehicle under a limit working condition, which comprises the following steps: establishing a control reference model which comprises a double-track three-degree-of-freedom vehicle dynamics model of the four-wheel-drive electric vehicle and a tire model considering longitudinal and transverse coupling characteristics; decoupling through a maximum irrelevant primitive control channel, and converting a coefficient matrix of an overdrive system of the electric four-wheel drive vehicle into a square matrix related to virtual control input; solving a square matrix related to virtual control input by adopting an integral fuzzy sliding mode controller to obtain virtual control input; and converting the virtual control input into actual physical input by adopting a control allocation method based on constraint optimization, and transmitting the actual physical input to the actuator and the whole vehicle model. The invention ensures that the automatic driving vehicle has high-level driving capability of a professional driver, can expand the application scene and the dynamic control boundary of the electric four-wheel drive vehicle, and furthest exerts the active safety performance and the dynamic potential of the automatic driving vehicle.

Description

Autonomous drift control method and system for electric four-wheel drive vehicle under extreme conditions

technical field

The invention relates to the technical field of active safety control of intelligent driving vehicles, in particular to an autonomous drift control method and system of an electric four-wheel drive vehicle under extreme working conditions.

Background technique

With the rapid development of the automobile industry, the active safety of automobiles is facing more and more severe challenges. At the same time, major domestic and foreign manufacturers have also developed and applied a variety of vehicle stability control technologies, including anti-lock braking systems, driving anti-skid systems, etc., but these technologies are mainly aimed at conventional driving conditions, and cannot cope with emergency scenarios and extreme driving conditions, such as icy and snowy roads, high-speed emergency collision avoidance scenarios, etc.

Extending the driving limits of a vehicle requires utilizing the tire's grip as much as possible. In racing, professional drivers usually consciously control the wheels to lock or slip to reduce lap time or avoid obstacles. This operation is called "drifting". The essence of drift is to precisely control the vehicle to be in a critical stable equilibrium condition under the oversteer state. At this time, the front wheel adhesion is close to saturation and the rear wheel reaches the adhesion limit. In professional competitions, rear-wheel-drive vehicles are generally selected to reduce the difficulty of the driver's operation. As a new direction for the development of electric vehicle technology, the electric four-wheel drive vehicle has the advantages of response speed and control accuracy, and the wheel torque can be flexibly distributed, the grip is stronger, and the slip angle limit range is larger. The control boundary and effect of vehicle dynamics under extreme conditions provide more possibilities, and also put forward higher requirements for control methods.

Therefore, it is necessary to study the essence of the drift operation of professional drivers, and to explore the dynamic characteristics and control strategies of electric four-wheel drive vehicles under the adhesion limit, so that the self-driving vehicles have the high-level driving ability of professional drivers, and can better explore the situation when the tires lock up. Or control potential in skid conditions, maximizing the application scenarios and dynamic control boundaries of autonomous vehicles.

SUMMARY OF THE INVENTION

In view of the above problems, the purpose of the present invention is to provide an autonomous drift control method and system for an electric four-wheel drive vehicle under extreme working conditions, which can expand the application scenarios and dynamic control boundaries of the electric four-wheel drive vehicle, and maximize its active Safety performance and power potential.

In order to achieve the above object, the present invention adopts the following technical solutions: an autonomous drift control method for an electric four-wheel drive vehicle under extreme working conditions, which comprises the following steps: 1) establishing a control reference model, including the two-track three-degree-of-freedom of the four-wheel drive electric vehicle The vehicle dynamics model and the tire model considering the longitudinal and lateral coupling characteristics; 2) Through the maximum irrelevant primitive control channel decoupling, the coefficient matrix of the overdrive system of the electric four-wheel drive vehicle is converted into a square matrix about the virtual control input, and a square matrix is obtained. 3) Using the integral fuzzy sliding mode controller to solve the square affine system to obtain the virtual control input; 4) Using the control allocation method based on constraint optimization to convert the virtual control input into the actual physical input, and transmit it to the execution equipment and vehicle models.

Further, the control reference model calculates the longitudinal speed, lateral speed and yaw angular speed corresponding to the steady-state drift of the four-wheel drive electric vehicle in the current driving environment, and sends the calculation results to the integral fuzzy sliding mode controller as the target. state quantity.

Further, the two-track three-DOF vehicle dynamics model is expressed as:

为车辆的横摆角速度，

为车辆的横摆角加速度，

为纵向加速度，

为横向加速度，F_xj和F_yj分别表示车轮切向及横向轮胎地面力，其中j＝1,2,3,4分别表示左前轮、右前轮、左后轮和右后轮，

为状态量相关的非线性项。Among them, V _x is the longitudinal vehicle speed, V _y is the lateral vehicle speed, ψ is the yaw angle of the vehicle,

is the yaw rate of the vehicle,

is the yaw angular acceleration of the vehicle,

is the longitudinal acceleration,

is the lateral acceleration, F _xj and F _yj represent the wheel tangential and lateral tire ground forces, respectively, where j=1, 2, 3, 4 represent the left front wheel, the right front wheel, the left rear wheel and the right rear wheel, respectively,

is the nonlinear term related to the state quantity.

Further, the tire model is expressed as:

为轮胎可用的最大侧向力。where μ is the road adhesion coefficient of the tire, F _z is the vertical load of the tire, D and E are the Pacejka tire model parameters, α is the tire side slip angle, α _cr is the tire critical side slip angle, and β is the vehicle center of mass side slip horn,

The maximum lateral force available for the tire.

Further, in the step 2), specifically:

The state equation of the overdriven system:

为系统的状态变量，

为系统的控制输入，F(x)为系统的非线性项，n、m分别为系统状态变量和控制输入的维数，

为有理数；系数矩阵

且rank(B)＝n<m，通过对过驱动系统的系数矩阵B转化为方形矩阵，得到各输入量之间的耦合关系，将具有相似控制效果的输入放进了同一个控制通道中，最终将系统转化为：in

is the state variable of the system,

is the control input of the system, F(x) is the nonlinear term of the system, n and m are the dimensions of the system state variable and control input, respectively,

is a rational number; coefficient matrix

And rank(B)=n<m, by converting the coefficient matrix B of the overdrive system into a square matrix, the coupling relationship between the various inputs is obtained, and the inputs with similar control effects are put into the same control channel, Ultimately transforming the system into:

The system represented in the formula is a square affine system with control decoupling; K′=[K _{1_1} ,K _{2_1} ,…,K _{n_1} ], K _{i_1} is the first column vector of the coefficient matrix K neutron matrix K _i , i=1,2,...,n.

Further, in the step 3), the input solution form of the integral fuzzy sliding mode controller is composed of two parts: the nominal control input v ₀ and the robust control input v ₁ :

v=v ₀ +v ₁ ;

The virtual control input v is:

In the formula, K′=[K _{1_1} , K _{2_1} ,…,K _{n_1} ], K _{i_1} is the first column vector of the submatrix K _i in the coefficient matrix K, i=1,2,…,n; F(x ) is the nonlinear term of the system; L is the state feedback diagonal matrix; e is the system deviation; x _d =[x _1d ,x _2d ,...,x _nd ] ^T , which is the ideal control system state quantity given by the reference model value; M(x) is the diagonal matrix; u _f is the output of the fuzzy system.

Further, the _v0 is set to the state feedback control rate that enables the system to track the ideal system trajectory:

where x _d =[x _1d , x _2d ,...,x _nd ] ^T , is the ideal value of the system state quantity given by the reference model; e=x _d -x is the system state deviation.

Further, for the robust control input v ₁ , by designing the integral sliding mode surface in the controller, the system state falls on the sliding mode surface from the initial moment, eliminating the approach stage in the classical sliding mode; The system replaces the saturation function in v ₁ , and the fuzzy system exhibits the characteristics of a saturation function with a nonlinear slope in the boundary layer. Taking gs and g ₁ Δs as the input variables of the fuzzy system, the output is u _f , and the nonlinear control law is expressed as:

v ₁ =-M(x)u _f

The diagonal elements of the diagonal matrix M(x) are set to be positive and larger than the upper limit of the corresponding system disturbance to ensure the convergence of the sliding mode.

Further, in the step 4), the virtual control input obtained by the integral fuzzy sliding mode controller is converted into the actual physical input by using the control assignment based on constraint optimization, and its objective function is:

In the formula, J ₁ is the penalty term in the objective function considering the utilization rate of the front wheel attachment, J ₂ is the penalty term in the objective function considering the dynamic characteristics of the actuator, u=[F _y1 ,F _y2 ,F _y3 ,F _y4 ,F _x1 , F _x2 , F _x3 , F _x4 ] ^T is the system input, k ₁ and k ₂ are the weight coefficients; W is the weighting matrix used to determine the change ratio of each tire force, and the diagonal term of W is set to each execution The reciprocal of the bandwidth ω _i over which the tire force is generated by the generator is:

The constraints ensure that the combination of physical input of each channel meets the requirements of virtual control input, and the tire force of the front wheel does not exceed the adhesion limit while the adhesion of the rear wheel is saturated.

An autonomous drift control system for an electric four-wheel drive vehicle under extreme working conditions, which includes a control reference model establishment module, a decoupling module, a solution module and an input module; the control reference model establishment module is used to establish a control reference model, including four The two-track three-degree-of-freedom vehicle dynamics model of the electric vehicle and the tire model considering the longitudinal and lateral coupling characteristics; the decoupling module is decoupled through the maximum irrelevant primitive control channel to convert the coefficient matrix of the overdrive system of the electric four-wheel drive vehicle. In order to obtain a square affine system about the square matrix of the virtual control input; the solving module adopts the integral fuzzy sliding mode controller to solve the square affine system to obtain the virtual control input; the input module adopts the control based on constraint optimization The distribution method converts the virtual control input into the actual physical input, which is transmitted to the actuator and the vehicle model.

Due to the adoption of the above technical solutions, the present invention has the following advantages: by studying the essence of the drift operation of professional drivers, the present invention performs autonomous drift control for extreme operating conditions during the driving of the electric four-wheel drive vehicle. Firstly, the overdrive system is transformed into a square affine system by the maximum irrelevant primitive control channel decoupling method, which is easy to solve the control input; and the integral fuzzy sliding mode controller is used to calculate the virtual control input, which eliminates the tendency of the classical sliding mode. At the same time, it can suppress the chattering of the control system induced by external disturbance and parameter uncertainty. Finally, the allocation method based on multi-objective optimal control under complex constraints converts virtual control input into actual physical input. The allocation strategy considers the influence of actuator dynamic response differences and high coupling characteristics on control effect. The invention enables the self-driving vehicle to possess the high-level driving ability of a professional driver, expands the application scenarios and dynamic control boundaries of the electric four-wheel drive vehicle, and maximizes its active safety performance and power potential.

Description of drawings

FIG. 1 is a schematic flowchart of the control method of the present invention.

FIG. 2 is a comparison diagram of the actual motion state of the vehicle and the expected drift state.

Figure 3 shows the lateral force, longitudinal force and the ratio of the resultant tire force to the vertical load for each wheel.

Figure 4 shows the trajectory of the center of mass of the vehicle.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some, but not all, embodiments of the present invention. Based on the described embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art fall within the protection scope of the present invention.

As shown in FIG. 1 , the present invention provides an autonomous drift control method for an electric four-wheel drive vehicle under extreme working conditions, which includes the following steps:

1) Establish a control reference model, including the two-track three-DOF vehicle dynamics model of the four-wheel drive electric vehicle and the tire model considering the longitudinal and lateral coupling characteristics.

In this embodiment, the two-track three-degree-of-freedom vehicle dynamics model is:

Among them, the coefficient matrix B is:

为车辆的横摆角速度，

为车辆的横摆角加速度，

为纵向加速度，

为横向加速度，m为车辆质量，I_z为车辆横摆转动惯量，δ为前轮转角，L_a和L_b分别为质心与前轴/后轴之间的直线距离，L_w为二分之一轮距，F_xj和F_yj分别表示车轮切向及横向轮胎地面力，其中j＝1,2,3,4分别表示左前轮、右前轮、左后轮和右后轮，F_roll和F_air分别为车辆的滚动阻力和空气阻力：where V _x is the longitudinal vehicle speed, V _y is the lateral vehicle speed, ψ is the yaw angle of the vehicle,

is the yaw rate of the vehicle,

is the yaw angular acceleration of the vehicle,

is the longitudinal acceleration,

is the lateral acceleration, m is the vehicle mass, I _z is the vehicle yaw moment of inertia, δ is the front wheel angle, L _a and L _b are the straight-line distances between the center of mass and the front/rear axle, respectively, and L _w is half Wheelbase, F _xj and F _yj represent the tangential and lateral tire ground forces, respectively, where j=1, 2, 3, 4 represent the left front wheel, the right front wheel, the left rear wheel and the right rear wheel, respectively, F _roll and F _air are the rolling resistance and air resistance of the vehicle, respectively:

F _roll = fmg

where f is the rolling resistance coefficient, g is the gravitational acceleration coefficient, ρ is the air density, C _d is the air resistance coefficient, and A is the cross-sectional area of the vehicle.

The tire model is:

为轮胎可用的最大侧向力，其中：In the formula, μ is the road adhesion coefficient of the tire, F _z is the vertical load of the tire, D and E are the Pacejka tire model parameters, obtained by fitting the actual tire test data, α is the tire side slip angle, and α _cr is the tire critical side declination angle, β is the side-slip angle of the vehicle center of mass,

is the maximum lateral force available to the tire, where:

The slip angles of the front and rear wheels are as follows:

The vertical load of each wheel is:

In the formula, L=L _a +L _b , h _g is the height of the center of mass of the vehicle.

The establishment of the control reference model lays the foundation for the subsequent control channel decoupling. At the same time, according to the environmental information such as the road adhesion coefficient and the curvature radius of the desired trajectory, the steady-state drift corresponding to the four-wheel drive electric vehicle in the current driving environment can be calculated. The longitudinal velocity, lateral velocity and yaw angular velocity are sent to the integral fuzzy sliding mode controller as the target state quantity.

2) From the vehicle state equation established in step 1), it can be seen that under the extreme condition of adhesion, the electric four-wheel drive vehicle is an overdrive system with complex nonlinear characteristics. Through the decoupling of the maximum irrelevant primitive control channel, step 1 ) The coefficient matrix B of the overdrive system of the electric four-wheel drive vehicle is transformed into a square matrix about the virtual control input, and a square affine system is obtained. Both classical and modern control methods can stabilize the transformed system.

The state equation of the overdriven system is:

为系统的状态变量，

为系统的控制输入，F(x)为系统的非线性项。系数矩阵

且rank(B)＝n<m。其中n、m分别为系统状态变量和控制输入的维数；

为有理数。in

is the state variable of the system,

is the control input of the system, and F(x) is the nonlinear term of the system. coefficient matrix

And rank(B)=n<m. where n and m are the dimensions of the system state variables and control input, respectively;

is a rational number.

Since the coefficient matrix B of the overdrive system is irreversible, the traditional controller design method cannot be directly used to solve the control input, so the present invention converts B into a square matrix through the maximum irrelevant primitive control channel decoupling method.

使得：For the coefficient matrix B, there must be an invertible transition matrix

makes:

Among them, K consists of n sub-matrices K _i of rank 1, and K _i satisfies:

且rank(K_i)＝1

and rank(K _i )=1

Therefore, through the transition matrix H, the linearly related column vectors in the coefficient matrix K can be grouped into the same sub-matrix K _i . It can be considered that the first column vector K _{i_1} of K _i is the primitive vector of the entire matrix, then K _i can be further expressed as:

In the formula, β _{i_k} is the proportional coefficient of the k-th column vector K _{i_k} to the primitive K _{i_1} .

Let each proportional coefficient β _{i_k} form a transverse quantity β _i :

Then _Ki can be further expressed as:

K _i =K _{i_1} β _i

From this, K can be further expressed as:

Perform the same transition matrix H transform on the input vector u:

与K_i的维数相同。in

_Same dimension as Ki.

After the above change process, the state equation of the overdrive system can be expressed as:

make:

K′=[K _{1_1} ,K _{2_1} ,…,K _{n_1} ]

v=[v ₁ v ₂ ... v _n ] ^T

By extracting the coupling information in the coefficient matrix B, the input variables u′ _{i_j} transformed by the transition matrix are linearly combined again, and the inputs with similar control effects are put into the same control channel. The different control channels v _i are independent of each other, realizing the decoupling of the control channels of the control system. Because _vi cannot guarantee a one-to-one correspondence with actual physical inputs, it is named virtual control input.

After the above derivation and transformation process, the state equation of the system is transformed into:

The system represented by the above equation is a square affine system with control decoupling, and the controller can be set by modern control theory methods to solve the virtual control input v.

3) The integral fuzzy sliding mode controller is used to solve the square affine system, and the virtual control input is obtained:

同时接受经由执行器及整车模型实时反馈的实际状态量

形成闭环控制。最终通过积分式模糊滑模控制器求得方形仿射系统的虚拟控制输入v。Specifically: based on the square affine system transformed in step 2), an integral fuzzy sliding mode controller is designed to avoid the approaching stage problem in the classical sliding mode control method and improve the robustness of the system. Based on the control reference model in step 1), the steady-state drift target state quantity of the integral fuzzy sliding mode controller under the preset road adhesion coefficient and trajectory radius can be obtained

At the same time, it accepts the actual state quantity fed back by the actuator and the vehicle model in real time.

form a closed loop control. Finally, the virtual control input v of the square affine system is obtained by the integral fuzzy sliding mode controller.

The input v solution form of the integral fuzzy sliding mode controller consists of two parts: the nominal control input v ₀ and the robust control input v ₁ :

v=v ₀ +v ₁

The sliding mode surface s of the integral fuzzy sliding mode controller is:

s=s ₀ +z

s ₀ =C ^T x

Among them, s ₀ is the linear combination of the system state quantity x, and C ^T is the linear combination coefficient. And z is the integral sliding mode term, which is obtained indirectly by the following equation:

z(0)=-s ₀ (x(0))

Therefore:

s(0)=s ₀ (x(0))+z(0)=0

It shows that if the design of the sliding mode surface is established, the system state will fall on the sliding mode surface from the initial moment, eliminating the approaching stage in the classical sliding mode. When the system is on the sliding surface, the equation of motion of the system can be obtained by taking the time derivative of the sliding surface:

Choosing the Lyapunov function and derivation of it, we get:

In order to suppress the chattering of the system, the robust control term v ₁ is often designed by the boundary layer theory, and the expression is as follows:

where ε>0 is the thickness of the saturation region; M(x)>Δd, where Δd is the observer disturbance estimation deviation.

However, when the vehicle is in a highly unstable boundary, external disturbances and parameter uncertainty may still induce buffeting of the control system. The present invention uses a fuzzy system to replace the saturation function, and the fuzzy system presents the characteristic of a saturation function with a nonlinear slope in the boundary layer. Taking gs and g ₁ Δs as the input variables of the fuzzy system, and the output of the fuzzy system as u _f , the nonlinear control law can be expressed as

v ₁ =-M(x)u _f

The input variables gs and g ₁ Δs of the fuzzy system are divided into three types of fuzzy sets: negative (N), zero (Z) and positive (P), and the output variables are divided into negative large (NB), negative medium (NM) and negative small (NS). ), zero (ZE), positive small (PS), positive middle (PM) and positive large (PB) 7 fuzzy sets; the trigonometric function is used as the fuzzy membership function; the Mamdani fuzzy model is applied, and the fuzzy rules of the system are shown in Table 1. ; Use the centroid method to defuzzify the output variables.

Table 1 System Fuzzy Rules

By choosing the linear combination coefficient matrix C ^T to make C ^T K' positive definite, and making the diagonal system of the diagonal matrix M(x) positive and greater than the upper limit of the corresponding system disturbance, it can be guaranteed that:

The only condition for an equal sign to exist:

That is, the sliding mode equation of motion of the system. Through the above derivation, the existence and convergence of the sliding mode are proved.

v ₀ is set to the state feedback control rate that makes the system follow the ideal system trajectory:

where x _d =[x _1d , x _2d ,...,x _nd ] ^T , is the ideal value of the system state quantity given by the control reference model in step 1); e=x _d -x is the system state deviation.

Put the expression of v ₀ into the system state equation, we can get:

Through the design of the state feedback diagonal matrix L, the above formula is a Hurwitz polynomial, then the system deviation e is asymptotically stable and must converge to 0 in a limited time.

After the design of the above nominal controller and robust controller, the virtual control input of the system is solved as:

4) In order to solve the actual physical input, while considering the dynamic response differences and highly coupled characteristics of the actuators such as driving, braking and steering and the tire forces generated by them, an online control assignment method based on multi-objective optimization under complex constraints is adopted.

After the virtual control input of the system is obtained by the integral fuzzy sliding mode controller, the control distribution method based on constraint optimization is used to convert the virtual control input into the actual physical input and transmit it to the actuator and the vehicle model.

In the drift state, the vehicle is at the high-speed instability boundary. The dynamic response difference and high coupling characteristics of the driving motor system, braking system, steering system and other executive systems will have a great impact on the control performance and efficiency. Therefore, the control allocation method cannot ignore the execution. difference in transient characteristics of the device. At the same time, since the rear wheel reaches the adhesion limit during the vehicle drifting process, it is necessary to reduce the tire adhesion utilization rate of the front wheel as much as possible to increase the control margin of the front wheel and prevent the vehicle from instability caused by additional operations under emergency conditions.

The objective function for optimal distribution of tire force is defined as:

In the formula, J ₁ is the penalty term in the objective function considering the utilization rate of the front wheel attachment, J ₂ is the penalty term in the objective function considering the dynamic characteristics of the actuator, u=[F _y1 ,F _y2 ,F _y3 ,F _y4 ,F _x1 , F _x2 , F _x3 , F _x4 ] ^T is the system input, and k ₁ and k ₂ are the weight coefficients. W is a weighting matrix, which is used to determine the change ratio of each tire force. The present invention believes that the output increment of tire force should be proportional to the bandwidth of each actuator to generate tire force, so the diagonal term of W is set as the reciprocal of the bandwidth _ωi of each actuator to generate tire force, namely:

ω _i can be determined by the dynamic characteristics of the actuator and the dynamic characteristics of the tire:

Among them, τ _a is the time constant after approximating the actuators such as driving/braking/steering as a first-order system, and τ _w is the longitudinal and lateral time constant of the tire. The specific value is set according to the nature and change of tire force.

In a second embodiment of the present invention, an autonomous drift control system for an electric four-wheel drive vehicle under extreme working conditions is provided, which includes a control reference model establishment module, a decoupling module, a solution module and an input module;

The control reference model building module is used to establish the control reference model, including the two-track three-degree-of-freedom vehicle dynamics model of the four-wheel drive electric vehicle and the tire model considering the longitudinal and lateral coupling characteristics;

The decoupling module decouples the maximum irrelevant primitive control channel, and converts the coefficient matrix of the overdrive system of the electric four-wheel drive vehicle into a square matrix about the virtual control input to obtain a square affine system;

The solving module adopts an integral fuzzy sliding mode controller to solve the square affine system, and obtains a virtual control input;

The input module adopts the control distribution method based on constraint optimization to convert virtual control input into actual physical input, and transmit it to the actuator and vehicle model.

To sum up, as shown in FIGS. 2 to 4 , it is a schematic diagram of the effect of the simulation test using the autonomous drift control method for an electric four-wheel drive vehicle provided by the present invention under extreme working conditions. After being controlled by the autonomous drift control method of the electric four-wheel drive vehicle under extreme working conditions provided by the present invention, as shown in Figure 2, the vehicle motion state can be stabilized in the steady state drift state under the current road conditions within a certain period of time; as shown in Figure 2 3, the longitudinal and lateral tire forces of each wheel can be dynamically adjusted according to the virtual control input command, while ensuring that the two wheels of the rear axle are always in the adhesion limit state (the road adhesion coefficient is set to 0.6), while the two wheels of the front axle still have certain control. space; as shown in Figure 4, the vehicle can start a circular motion of steady-state drift within a certain period of time, and the circle radius is the set curvature radius of the desired trajectory (the road curvature radius is set to 13 meters).

Although the present invention has been described with reference to the preferred embodiments and examples thereof, it will be understood by those skilled in the art that various features of the above-described embodiments of the present invention may be combined as appropriate to form variations, and those skilled in the art may Various other variations and modifications are made to the above-described embodiments, equivalent technical solutions are made, and applied in various fields without departing from the scope of the present invention.

Claims (10)

1. an autonomous drift control method of an electric four-wheel drive vehicle under extreme working conditions, is characterized in that, comprises the following steps: 1) Establish a control reference model, including the two-track three-DOF vehicle dynamics model of the four-wheel drive electric vehicle and the tire model considering the longitudinal and lateral coupling characteristics; 2) Through the decoupling of the maximum irrelevant primitive control channel, the coefficient matrix of the overdrive system of the electric four-wheel drive vehicle is converted into a square matrix about the virtual control input, and a square affine system is obtained; 3) The integral fuzzy sliding mode controller is used to solve the square affine system, and the virtual control input is obtained; 4) The control distribution method based on constraint optimization is used to convert the virtual control input into the actual physical input and transmit it to the actuator and the vehicle model. 2. The control method according to claim 1, wherein the control reference model calculates the longitudinal speed, lateral speed and yaw angular speed corresponding to the steady-state drift of the four-wheel drive electric vehicle in the current driving environment, and calculates the The calculation result is sent to the integral fuzzy sliding mode controller as the target state quantity. 3. The control method according to claim 1, wherein the vehicle dynamics model of the two-track three-degree-of-freedom vehicle is expressed as:

where V _x is the longitudinal vehicle speed, V _y is the lateral vehicle speed, ψ is the yaw angle of the vehicle,

is the yaw rate of the vehicle,

is the yaw angular acceleration of the vehicle,

is the longitudinal acceleration,

is the lateral acceleration, F _xj and F _yj represent the wheel tangential and lateral tire ground forces, respectively, where j=1, 2, 3, 4 represent the left front wheel, the right front wheel, the left rear wheel and the right rear wheel, respectively,

is the nonlinear term related to the state quantity. 4. The control method according to claim 1, wherein the tire model is expressed as:

where μ is the road adhesion coefficient of the tire, F _z is the vertical load of the tire, D and E are the Pacejka tire model parameters, α is the tire side slip angle, α _cr is the tire critical side slip angle, and β is the vehicle center of mass side slip horn,

The maximum lateral force available for the tire. 5. control method as claimed in claim 1, is characterized in that, in described step 2), is specifically: The state equation of the overdriven system:

in

is the state variable of the system,

is the control input of the system, F(x) is the nonlinear term of the system, n and m are the dimensions of the system state variable and control input, respectively,

is a rational number; coefficient matrix

And rank(B)=n<m, by converting the coefficient matrix B of the overdrive system into a square matrix, the coupling relationship between the various inputs is obtained, and the inputs with similar control effects are put into the same control channel, Ultimately transforming the system into:

The system represented in the formula is a square affine system with control decoupling; K′=[K _{1_1} , K _{2_1} , ..., K _{n_1} ], K _{i_1} is the first column vector of the coefficient matrix K neutron matrix K _i , i=1,2,...,n. 6. control method as claimed in claim 1 is characterized in that, in described step 3), the input solution form of described integral fuzzy sliding mode controller consists of nominal control input v ₀ and robust control input v ₁ two. Partial composition: v=v ₀ +v ₁ ; The virtual control input v is:

In the formula, K′=[K _{1_1} , K _{2_1} ,…, K _{n_1} ], K _{i_1} is the first column vector of the submatrix K _i in the coefficient matrix K, i=1, 2,…,n; F(x ) is the nonlinear term of the system; L is the state feedback diagonal matrix; e is the system deviation; x _d = [x _1d , x _2d , ..., x _nd ] ^T , which is the ideal control system state quantity given by the reference model value; M(x) is the diagonal matrix; u _f is the output of the fuzzy system. 7. The control method of claim 6, wherein the v ₀ is set to a state feedback control rate that enables the system to track an ideal system trajectory:

where x _d =[x _1d , x _2d , . . . , x _nd ] ^T , is the ideal value of the system state quantity given by the reference model; e=x _d -x is the system state deviation. 8. The control method according to claim 6, characterized in that, for the robust control input v ₁ , by designing an integral sliding mode surface in the controller, the system state falls on the sliding mode surface from the initial moment, eliminating the At the same time, the fuzzy system is used to replace the saturation function in v ₁ , and the fuzzy system presents the characteristic of the saturation function with nonlinear slope in the boundary layer. Taking gs and g ₁ Δs as the input variables of the fuzzy system, The output is u _f , and the nonlinear control law is expressed as: v ₁ =-M(x)u _f The diagonal elements of the diagonal matrix M(x) are set to be positive and larger than the upper limit of the corresponding system disturbance to ensure the convergence of the sliding mode. 9. control method as claimed in claim 1 is characterized in that, in described step 4), adopts the control distribution based on constraint optimization to convert the virtual control input obtained by integral fuzzy sliding mode controller into actual physical input, its The objective function is:

In the formula, J ₁ is the penalty item in the objective function considering the utilization rate of front wheel attachment, J ₂ is the penalty item in the objective function considering the dynamic characteristics of the actuator, u=[F _y1 , F _y2 , F _y3 , F _y4 , F _x1 , F _x2 , F _x3 , F _x4 ] ^T is the system input, k ₁ , k ₂ are the weight coefficients; W is the weighting matrix used to determine the change ratio of each tire force, and the diagonal term of W is set to each execution The reciprocal of the bandwidth ω _i over which the tire force is generated by the generator is:

The constraints ensure that the combination of physical input of each channel meets the requirements of virtual control input, and the tire force of the front wheel does not exceed the adhesion limit while the adhesion of the rear wheel is saturated. 10. An autonomous drift control system for an electric four-wheel drive vehicle under extreme working conditions, characterized in that it comprises: a control reference model establishment module, a decoupling module, a solution module and an input module; The control reference model establishment module is used to establish a control reference model, including a two-track three-degree-of-freedom vehicle dynamics model of a four-wheel drive electric vehicle and a tire model considering longitudinal and lateral coupling characteristics; The decoupling module is decoupled through the maximum irrelevant primitive control channel, and converts the coefficient matrix of the overdrive system of the electric four-wheel drive vehicle into a square matrix about the virtual control input to obtain a square affine system; The solving module adopts an integral fuzzy sliding mode controller to solve the square affine system, and obtains a virtual control input; The input module adopts a control distribution method based on constraint optimization to convert virtual control input into actual physical input, and transmit it to the actuator and the vehicle model.

Images