CN104881030B - Side longitudinal coupling tracking control method for unmanned vehicle based on fast terminal sliding mode principle - Google Patents

Abstract

The invention provides a kind of unmanned vehicle side Longitudinal data tracking and controlling method based on fast terminal sliding formwork principle, including：1) input and output parameter of selection unmanned vehicle side Longitudinal data tracking control unit；2) tracing control error and its state equation are defined；3) fast terminal sliding formwork is set up as state variable with tracing control error；4) fast terminal tendency rate is used as the sliding formwork tendency rate of tracing control；5) coupled relation between desired driving force or brake force and desired side force is derived from according to vehicle dynamic model and the above；6) expectation front wheel angle is calculated；7) according to the positive and negative of the inverse longitudinal force model output tried to achieve, judgement should currently calculate desired throttle opening or desired braking moment；8) desired throttle opening or braking moment are calculated.Be can see from simulation result, the present invention not only improves the error precision of Position Tracking Control, it may have the error precision of good speed tracking control.

Description

Side longitudinal coupling tracking control method for unmanned vehicle based on fast terminal sliding mode principle

【Technical field】

The invention relates to the technical field of autonomous driving and sliding mode control of an unmanned vehicle, in particular to a tracking control method for realizing side longitudinal coupling of an unmanned vehicle.

【Background technique】

The motion control of unmanned vehicles is the key technology of the manipulation control layer in the navigation control architecture, and the motion control algorithm is the core research content of this layer. The motion control here is the tracking control based on the planned trajectory. Tracking control of autonomous vehicles includes longitudinal tracking control and lateral tracking control. Longitudinal tracking control is to control the driving speed of the vehicle to be consistent with the planned speed, and to maintain a certain driving distance with the vehicle in front. Lateral tracking control is to control the driving direction of the vehicle so that it can travel along the planned path and meet the basic performance requirements. Including tracking accuracy, adaptability to vehicle parameters and environmental changes (robustness), and ride comfort.

When the unmanned vehicle performs small curvature planning actions such as lane keeping, the coupling between the vehicle's lateral motion and longitudinal motion can often be ignored, and the lateral and longitudinal decoupling control is performed, and the longitudinal tracking controller and lateral tracking controller are designed separately. The study of longitudinal tracking control algorithm is dedicated to solving the problem of improving the longitudinal control accuracy under the influence of strong nonlinearity of vehicle longitudinal dynamics system and the influence of large longitudinal disturbance conditions. The methods adopted include intelligent control such as PID control, optimal control, adaptive control, sliding mode control, fuzzy control and artificial neural network control. The automatic cruise system developed based on the above algorithm has been applied in some advanced commercial vehicles. Since the lateral tracking control is based on two-dimensional space to solve the tracking problem, coupled with the change of vehicle and road parameters, and the inherent nonlinear characteristics of lateral motion, the design of the controller is often a complex problem, with self-correcting real-time The control algorithm of learning ability is the focus of research. On the other hand, when the vehicle performs some complex planned maneuvers such as changing lanes and turning at intersections in an environment with moving obstacles, it is required to track the planned position sequence and planned velocity sequence at the same time. At this time, the coupling is strong, and if the side-to-longitudinal decoupling control algorithm is still used, a large tracking error will be generated.

Some researchers have conducted research on side-longitudinal coupling control strategies based on the coupling effect of vehicle side-longitudinal motion, but the current coupling control strategy of side-longitudinal motion is mainly aimed at vehicle platoon control, for the control of high mobility behavior of a single autonomous vehicle, such as Vehicles perform some complex planning maneuvers such as changing lanes and turning at intersections in an environment with moving obstacles, and the research is less or the control accuracy is low.

The relevant literature retrieved is given below:

[1] Wu Qing, He Zhiwei, Chu Xiumin, Zong Chengqiang. Key technologies and research progress of vehicle platooning control in intelligent vehicle road system [J]. Transportation and Computer, 2008, 26(4): 154-157.

[2] Rajamani R. Vehicle Dynamics and Control. New York: Springer, 2006.

[3] Zhou QZ, Wang FY and Li L. Robust Sliding Mode Control of 4WS Vehicles for Automatic Path Tracking [C]. Proceedings of IEEE Intelligent Vehicles Symposium, Piscataway, USA, 2005: 819-826.

[4]Hou Yongping, Hu Yujin, Li Chenggang, Guo Konghui. Study on characteristics of sideslip relaxation length at large slip angles[J].Automotive Engineering, 2001(2):78-81.

[5]Edward H, Lim M and Hedrick JK.Lateral and Longitudinal Vehicle Control Coupling for Automated Vehicle Operation[C].Proceedings of the American Control Conference San Diego,California,Jun.1999:3676-3680.

[6] Rajamani R, Tan HS, Law BK, et al. Demonstration of Integrated Longitudinal and Lateral Control for the Operation of Automated Vehicles in Platoons [J]. IEEE Transactions on Control Systems Technology, 2000, 8(4): 695-708.

[7] Li Yinong, Yang Liu, Zheng Ling, etc. Vehicle longitudinal and lateral coupling control based on sliding mode control [J]. China Mechanical Engineering, 2007,18(7):866-870.

[8] Yu X, Man Z, Wu Y. Terminal Sliding Modes with Fast Transient Performance [C]. Proceedings of the 36th IEEE International Conference on Control & Decision, San Diego, 1997: 962-963.

[9] Yu Shuanghe, Research on Fast Robust Sliding Mode Control Method with Low Chattering [D]. Doctoral Dissertation of Harbin Institute of Technology, 2000.

【Content of invention】

The purpose of the present invention is to provide an unmanned vehicle side longitudinal coupling tracking control method based on the principle of fast terminal sliding mode, to solve the defects or deficiencies in the above-mentioned existing theory and design; the control method is to improve the tracking of the controller Performance, by directly compensating the coupling in the controller design stage, the unmanned vehicle can be robust when performing some complex and highly maneuverable planning behaviors such as changing lanes and turning at intersections in a multi-movement obstacle environment. High-precision tracking planning behavior, including tracking planning position status and speed status.

In order to achieve the above-mentioned purpose, the present invention adopts following technical scheme:

A side longitudinal coupling tracking control method for unmanned vehicles based on the principle of fast terminal sliding mode, including the following steps:

Step 1: Select the input parameters and output parameters of the longitudinal coupling tracking controller on the side of the unmanned vehicle;

Step 2: Define the tracking control error and its state equation:

ε ₁ =XX _d +YY _d

where (X,Y, ) are the position coordinates and velocity vector yaw angle of the controlled vehicle on the two-dimensional plane, respectively, (X _d , Y _d , ) are the position coordinates and velocity vector yaw angle of the vehicle at the planning point on the trajectory planned by the trajectory planning algorithm, respectively;

Step 3: Using tracking control errors ε ₁ and ε ₂ as state variables, establish a first-order nonlinear switching function, that is, design fast terminal sliding mode surfaces S ₁ and S ₂ :

Among them, p _α1 ＞0, p _β1 ＞0, p _α2 ＞0 and p _β2 ＞0 are the coefficients of the first-order item and the exponential item of the sliding surface, and the values used in simulation are p _α1 ＝p _α2 ＝p _β1 ＝p _β2 ＝2 ;p _q1 , p _p1 , p _q2 , p _p2 are all odd numbers, they are power exponent parameters, and satisfy p _q1 <p _p1 <2p _q1 , p _q2 <p _p2 <2p _q2 , and the value in simulation is p _q1 =p _q2 = 5, p _p1 = p _p2 = 7;

Step 4: Use the fast terminal approach rate formula as the sliding mode approach rate of the tracking controller:

Among them, f _α1 >0, f _β1 >0, f _α2 >0 and f _β2 >0 are the coefficients of the first-order term and exponential term of the fast terminal approach rate, and the values in the simulation are f _α1 =f _α2 =80, f _β1 = f _β2 = 2; f _q1 , f _p1 , f _q2 , and f _p2 are all odd numbers, which are power exponent parameters, and satisfy f _q1 <f _p1 <2f _q1 , f _q2 <f _p2 <2f _q2 , and take the value of f _q1 =f _q2 =5, f _p1 =f _p2 =7;

Step 5: According to the vehicle dynamics model and the above steps 1 to 4, derive the coupling relationship between the expected driving force or braking force F _trbrd and the expected lateral force F _ld :

Step 6: Calculate and obtain the expected front wheel angle:

where C _f is the cornering stiffness of the front tire, l _f is the distance from the front axle of the vehicle to the center of mass; υ is the linear velocity of the vehicle; β is the side slip angle of the vehicle's center of mass; r is the angular velocity of the vehicle, and ψ is the heading angle of the vehicle ,

Step 7: Calculate the inverse longitudinal force model with the expected driving force or expected braking force F _trbrd as input for the vehicle longitudinal force model; if the calculation result of F _trbrd is positive, it is the expected driving force F _trd , and then calculate the expected Throttle opening α _thb ; on the contrary, if the calculation result of F _trbrd is negative, it is the expected braking force F _brd , and then calculate the expected braking torque P _brd ;

Step 8: The calculation method of the expected throttle opening α _thb is: without considering the elastic deformation of the tire and the drive train, according to the vehicle longitudinal force model, the expected engine torque T _ed and the current engine speed ω _e , using the inverse engine model The desired throttle opening α _thb can be obtained:

α _thb ＝f _ieng (T _ed ,ω _e )

where f _ieng (T _ed ,ω _e ) represents the inverse engine torque characteristic function;

Alternatively, the calculation formula of the expected braking torque P _brd is: P _brd =F _brd /K _br , where K _br is a proportional coefficient of the linear relationship between the expected braking force and the braking torque.

In step 1, the current coordinates of the center of mass of the unmanned vehicle are selected from the coordinate error XX _d and YY _d of the planned point on the planned trajectory, and the error of the yaw angle of the speed vector of the unmanned vehicle is the same as the yaw angle of the planned point is the input parameter of the tracking controller; the throttle opening α _th or the braking torque F _br , and the front wheel rotation angle δ _f are the output parameters of the tracking controller.

The value of the cornering stiffness _Cf of the front tire is twice the cornering stiffness of the single tire.

Compared with the prior art, the advantages of the present invention are as follows: when the unmanned vehicle autonomously drives to complete the highly maneuverable driving behavior, if the current vehicle state and the expected state frequently have a large tracking error, based on the method of the present invention, the The controller can quickly reduce the tracking error, so that the unmanned vehicle can quickly track the desired state, and not only has strong robustness to the uncertainty factors of the system in the whole control process, but also can obtain satisfactory dynamic quality, and at the same time control Simple and easy to implement. The invention not only improves the error precision of position tracking control, but also has very good error precision of speed tracking control.

The derivation in step 5 to step 7 of the present invention all adopts the vehicle side longitudinal coupling dynamics model, which has better model accuracy in the behavior of strong vehicle mobility.

In step three of the present invention, tracking control errors ε ₁ and ε ₂ are used as state variables, and the first-order nonlinear switching function is used to establish fast terminal sliding modes S ₁ and S ₂ , so that the tracking error of the vehicle tracking control system reaches the state of the sliding mode surface Converge quickly to zero, that is, the current centroid position of the unmanned vehicle coincides with the expected position, and the current velocity vector heading angle coincides with the expected heading angle.

In step 4 of the present invention, based on the fast terminal sliding modes S ₁ and S ₂ , the fast terminal approach rate is established as the sliding mode approach rate of the tracking controller, which can make the tracking error of the vehicle tracking control system not located on the sliding mode surface, in the It can quickly reach the sliding surface from any state with weak or no jitter in a limited time, and it has stronger robustness than conventional methods.

【Description of drawings】

Figure 1 is a schematic diagram of an unmanned vehicle tracking control system;

Fig. 2 is a flow chart of the longitudinal coupling tracking control method on the side of the unmanned vehicle based on the fast terminal sliding mode principle;

Fig. 3 is a schematic diagram of an inverse engine torque characteristic function;

Fig. 4 is a schematic diagram of a target trajectory, a tracking trajectory under coupling control, and a tracking trajectory under decoupling control;

Figure 5 is a comparison diagram of tracking error under coupling control and tracking error under decoupling control; among them, Figure 5(a) is a comparison diagram of position tracking error in X direction, and Figure 5(b) is a comparison diagram of position tracking error in Y direction Comparison diagram, Fig. 5(c) is a comparison diagram of velocity tracking error in X direction, Fig. 5(d) is a comparison diagram of velocity tracking error in Y direction, Fig. 5(e) is a comparison diagram of yaw angle (direction) tracking error ;

Figure 6 is a schematic diagram of the control amount of the side-to-longitudinal coupling control algorithm; Figure 6(a) is a schematic diagram of the accelerator opening of the longitudinal tracking control, the unit is %; Figure 6(b) is a schematic diagram of the brake pedal pressure of the longitudinal tracking control, the unit is N, Figure 6(c) is a schematic diagram of the front wheel angle of the lateral tracking control, the unit is rad, and the control quantity shown in the figure conforms to the physical characteristic constraints of the actuator.

【detailed description】

According to the principle of side-to-longitudinal coupling control, this design is further described in detail.

Referring to Fig. 1, the present invention is used for the design of the tracking controller when the track of the unmanned vehicle is tracked and planned.

Referring to Fig. 2, the unmanned vehicle side longitudinal coupling tracking control method based on the fast terminal sliding mode principle of the present invention is divided into the following nine steps, each step is as follows:

1) Select the input parameters and output parameters of the unmanned vehicle side longitudinal coupling tracking controller according to the planned trajectory:

(1a) Input parameters: the current coordinates of the center of mass of the unmanned vehicle and the coordinate error of the planned point on the planned trajectory XX _d , YY _d ; the error of the yaw angle of the unmanned vehicle speed vector from the planned point yaw angle where (X,Y, ) are the position coordinates and velocity vector yaw angle of the controlled vehicle on the two-dimensional plane, respectively, (X _d , Y _d , ) are the position coordinates and velocity vector yaw angle of the vehicle at the planning point on the trajectory planned by the trajectory planning algorithm, respectively.

(1b) Output parameters: throttle opening α _th (or braking torque F _br ); front wheel rotation angle δ _f .

2) According to the principle of fast terminal sliding mode, the state variable is defined as the tracking control error, and its state equation is given:

(2a) State variables:

ε ₁ =XX _d +YY _d

(2b) The state equation of the state variable:

The kinematic equation of the vehicle is:

where υ is the linear velocity of the vehicle, r is the angular velocity of the vehicle, ψ is the heading angle of the vehicle, β is the sideslip angle of the vehicle's center of mass and

According to the kinematic equation of the vehicle, the state equation of the state variable can be deduced as:

The state equation for the state variable is further derived:

3) Design the fast terminal sliding surface based on the vehicle kinematics equation:

(3a) Using tracking control errors ε ₁ and ε ₂ as state variables, establish a first-order nonlinear switching function, that is, design fast terminal sliding mode surfaces S ₁ and S ₂ :

Among them, p _α1 ＞0, p _β1 ＞0, p _α2 ＞0 and p _β2 ＞0 are the coefficients of the first-order item and the exponential item of the sliding surface, and the values used in simulation are p _α1 ＝p _α2 ＝p _β1 ＝p _β2 ＝2 ;p _q1 , p _p1 , p _q2 , p _p2 are all odd numbers, they are power exponent parameters, and satisfy p _q1 <p _p1 <2p _q1 , p _q2 <p _p2 <2p _q2 , and the value in simulation is p _q1 =p _q2 =5, p _p1 =p _p2 =7.

(3b) Deriving the fast terminal sliding surface S ₁ and S ₂ and bringing it into the vehicle kinematics equation:

4) Using the fast terminal approach rate formula as the sliding mode approach rate of the tracking controller:

Among them, f _α1 >0, f _β1 >0, f _α2 >0 and f _β2 >0 are the coefficients of the first-order term and exponential term of the fast terminal approach rate, and the values in the simulation are f _α1 =f _α2 =80, f _β1 = f _β2 = 2; f _q1 , f _p1 , f _q2 , and f _p2 are all odd numbers, which are power exponent parameters, and satisfy f _q1 <f _p1 <2f _q1 , f _q2 <f _p2 <2f _q2 , and take the value of f _q1 =f _q2 =5, f _p1 =f _p2 =7.

5) According to the fast terminal sliding mode surface, the sliding mode reaching law and the coupling equation of vehicle dynamics, the coupling relationship of the expected control quantity of the tracking controller is obtained.

(5a) Vehicle dynamics coupling equation:

Among them, F _l and R _l are the lateral forces acting on the chassis by the front and rear tires respectively. F _f and F _r are the longitudinal forces acting on the front and rear axles, generated by the powertrain and braking system, respectively. F _ad is the forward air resistance of the vehicle. δ _f is the front wheel rotation angle, l _f and l _r are the distances from the front and rear axles of the vehicle to the center of mass. l is the distance l _f +l _r between the front and rear axles of the vehicle. μ is the viscosity coefficient of conventional road surface, usually μ=1 in the case of dry road surface, and μ=0.5 in the case of watery road surface, m is the mass of the vehicle, and I _ψ is the moment of inertia of the vehicle.

(5b) According to the derivation of the fast terminal sliding mode surface in (3b), the approach rate of the sliding mode in 4) and the vehicle dynamics coupling equation in (5a), the expected control quantities F _trbrd and F _ld of the tracking controller can be deduced The coupling relationship:

where F _trbrd is the expected driving force or braking force, F _ld is the expected lateral force, F _rr is the ground friction force of the rear wheel, q ₃ =cos(ψ+β)+sin(ψ+β), q ₄ =cos(ψ+β)-sin(ψ+β).

6) According to the linear tire model and the relationship between the comprehensive front wheel slip angle and the front wheel angle, calculate the expected front wheel angle:

(6a) Linear tire model:

F _l = C _f δ ₁

Where _Cf is the cornering stiffness of the front tire, where its value is twice the cornering stiffness of a single tire, and δ1 is the comprehensive side slip angle _of the front wheel.

(6b) The relationship between the front wheel comprehensive side slip angle and the front wheel rotation angle:

(6c) According to the expected lateral force F _ld in (5b), the expected front wheel angle of the steering control amount is obtained:

7) In order to obtain the longitudinal control amount throttle opening α _th or the braking torque P _br , it is necessary to obtain the inverse longitudinal force model with the expected driving force or expected braking force F _tbrd as input for the vehicle longitudinal force model. If the calculation result of F _trbrd is positive, then it is the expected driving force F _trd , then the expected throttle opening α _thb needs to be calculated. Conversely, if the calculation result of F _trbrd is negative, it means the expected braking force F _brd , and then the expected braking torque P _brd needs to be calculated.

8) Assuming that F _trbrd is the desired driving force F _trd , calculate the desired throttle opening α _thb .

(8a) According to the assumptions of the vehicle longitudinal force model established above, without considering the elastic deformation of tires and drive train, according to the model, the expected driving force can be expressed as:

F _trd ＝η _D T _ed τ _T (ω _t /ω _e )f _gear (α _th ,υ)i ₀ /h _t

where η _D is the power transmission coefficient of the vehicle driveline, T _ed is the desired engine output torque, τ _T is the torque ratio coefficient, and its independent variable is ω _t /ω _e , is the turbine speed ω _t of the fluid coupling and the current engine The ratio of rotational speed ω _e , f _gear (α _th , υ) is the transmission ratio function of automatic transmission, which is a nonlinear function of throttle opening α _th (fully closed is 0, fully open is 100%) and vehicle speed υ; i ₀ The transmission ratio of the main reducer; h _t is the effective radius of the wheel.

(8b) From the expected engine torque T _ed and the current engine speed ω _e , the expected throttle opening α _thb can be obtained by using the inverse engine model. The inverse engine model is expressed as follows:

α _thb ＝f _ieng (T _ed ,ω _e )

where f _ieng (T _ed ,ω _e ) represents the inverse engine torque characteristic function. By using the engine torque characteristic function f _eng (ω _e ,α _th ) and the provided data, the inverse engine torque characteristic function f _ieng (T _ed ,ω _e ) for calculating the throttle opening from the engine output torque and engine speed can be obtained. Refer to image 3.

Alternatively, let F _trbrd be the desired braking torque P _brd . According to the relationship between braking force and braking torque, the inverse braking system model can be obtained as:

P _brd =F _brd /K _br

where K _br is the scaling factor of the linear relationship between the desired braking force and braking torque.

In the present invention, a longitudinal coupling tracking control method based on the principle of fast terminal sliding mode is provided, and the control method is simulated and verified by using MATLAB. It can be seen from the simulation results of Fig. 4-Fig. 6 that when the unmanned vehicle autonomously drives to complete the driving behavior with strong maneuverability, if the current vehicle state and the expected state frequently have large tracking errors, based on the method of the present invention, the The controller can quickly reduce the tracking error, so that the unmanned vehicle can quickly track the desired state. The invention not only improves the error precision of the position tracking control, but also has good error precision of the speed tracking control, and the control amount is smooth and conforms to the physical characteristic constraint of the actuator.

Claims (3)

1. The unmanned vehicle side longitudinal coupling tracking control method based on the rapid terminal sliding mode principle is characterized by comprising the following steps of:

the method comprises the following steps: selecting input parameters and output parameters of the longitudinal coupling tracking controller at the side of the unmanned vehicle;

step two: defining a tracking control error and a state equation thereof:

₁＝X-X_d+Y-Y_d

${\stackrel{\&CenterDot;}{\mathrm{\&epsiv;}}}_1=\stackrel{\&CenterDot;}{X}-{\stackrel{\&CenterDot;}{X}}_d+\stackrel{\&CenterDot;}{Y}-{\stackrel{\&CenterDot;}{Y}}_d$

whereinRespectively a position coordinate and a speed vector yaw angle on a two-dimensional plane of the controlled vehicle,respectively obtaining the position coordinates and the speed vector yaw angle of the vehicle at the planned point on the planned track by the track planning algorithm;

step three: controlling errors by tracking₁、₂As a state variable, establishing a first-order nonlinear switching function, namely designing a rapid terminal sliding mode surface S₁And S₂：

$S_1={\stackrel{\&CenterDot;}{\mathrm{\&epsiv;}}}_1+p_{\mathrm{\&alpha;}1}{\mathrm{\&epsiv;}}_1+p_{\mathrm{\&beta;}1}{{\mathrm{\&epsiv;}}_1}^{p_{q1}/p_{p1}}$

$S_2={\stackrel{\&CenterDot;}{\mathrm{\&epsiv;}}}_2+p_{\mathrm{\&alpha;}2}{\mathrm{\&epsiv;}}_2+p_{\mathrm{\&beta;}2}{{\mathrm{\&epsiv;}}_2}^{p_{q2}/p_{p2}}$

Wherein p is_α1>0，p_β1>0，p_α2>0 and p_β2>0 is a sliding mode surface first-order term and an exponential term coefficient, and the value is p in simulation_α1＝p_α2＝p_β1＝p_β2＝2；p_q1，p_p1，p_q2，p_p2Are all odd numbers and are powersIndex parameter, and satisfies p_q1<p_p1<2p_q1，p_q2<p_p2<2p_q2The value is p in simulation_q1＝p_q2＝5，p_p1＝p_p2＝7；

Step four: adopting a quick terminal approach rate formula as a sliding mode approach rate of a tracking controller:

${\stackrel{\&CenterDot;}{S}}_1=-f_{\mathrm{\&alpha;}1}{\mathrm{\&beta;}}_1-f_{\mathrm{\&beta;}1}{S_1}^{f_{q1}/f_{p1}}$

${\stackrel{\&CenterDot;}{S}}_2=-f_{\mathrm{\&alpha;}2}{S}_2-f_{\mathrm{\&beta;}2}{S_2}^{f_{q2}/f_{p2}}$

wherein f is_α1>0，f_β1>0，f_α2>0 and f_β2>0 is the coefficient of the first order term and the exponential term of the approach rate of the rapid terminal, and the value is f in simulation_α1＝f_α2＝80，f_β1＝f_β2＝2；f_q1，f_p1，f_q2，f_p2Are all odd numbers, are power exponent parameters, and satisfy f_q1<f_p1<2f_q1，f_q2<f_p2<2f_q2The value is f in simulation_q1＝f_q2＝5，f_p1＝f_p2＝7；

Step five: deriving the expected driving force or braking force F according to the vehicle dynamic model and the content of the steps 1 to 4_trbrdAnd a desired lateral force F_ldThe coupling relationship between:

$\begin{array}{c}{F}_{trbrd}=-(q_3{f}_1+q_4{f}_2-\left({\stackrel{\&CenterDot;\&CenterDot;}{X}}_d+{\stackrel{\&CenterDot;\&CenterDot;}{Y}}_d\right)+p_{\mathrm{\&alpha;}1}{\stackrel{\&CenterDot;}{\mathrm{\&epsiv;}}}_1+\left(p_{\mathrm{\&beta;}1}{p}_{q1}/p_{p1}\right){{\mathrm{\&epsiv;}}_1}^{(p_{q1}-p_{p1})/p_{p1}}{\stackrel{\&CenterDot;}{\mathrm{\&epsiv;}}}_1\\ +f_{\mathrm{\&alpha;}1}{S}_1+f_{\mathrm{\&beta;}1}{S_1}^{f_{q1}/f_{p1}})/q_1{q}_3+F_{ld}(q_3{\mathrm{\&delta;}}_f-q_4)/q_3\end{array}$

wherein,F_rrin order to provide the ground friction force of the rear wheel, q₃＝cos(ψ+β)+sin(ψ+β)，q₄＝cos(ψ+β)-sin(ψ+β)，_fis a front wheel corner, R_lIs the lateral force of the rear tire acting on the chassis; f_fIs a longitudinal force acting on the front axle, generated by the powertrain and the braking system; l_fAnd l_rIs the distance of the front and rear axles of the vehicle to the center of mass; l is the distance between the front and rear axles of the vehicle_f+l_r； Mu is the viscosity coefficient of a conventional road surface, m is the mass of the vehicle, I_ψIs the moment of inertia of the vehicle;

step six: calculating to obtain an expected front wheel turning angle:

${\mathrm{\&delta;}}_{fd}=F_{ld}/C_f+(\mathrm{\&beta;}+l_f\frac{r}{\mathrm{\&upsi;}})$

wherein C is_fFor cornering stiffness of the front tyre, /)_fThe distance from the front axle of the vehicle to the center of mass, upsilon is the linear velocity of the vehicle, β is the center of mass slip angle of the vehicle, r is the angular velocity of the vehicle, psi is the heading angle of the vehicle,

step seven: determination of a desired driving force or a desired braking force F for a vehicle longitudinal force model_trbrdIs an input inverse longitudinal force model; if F_trbrdIf the calculated result of (b) is positive, the expected driving force F is obtained_trdThen, a desired throttle opening α is calculated_thb(ii) a On the contrary, if F_trbrdIf the result of the calculation of (c) is negative, the desired braking force F is obtained_brdThen calculating the desired braking torque P_brd；

Step eight desired throttle opening α_thbThe calculation method comprises the following steps: expected engine torque T according to a vehicle longitudinal force model without considering elastic deformation of tires and a drive train_edAnd the current engine speed omega_eUsing inverse engine modelsThe desired throttle opening α can be obtained_thb：

α_thb＝f_ieng(T_ed,ω_e)

Wherein f is_ieng(T_ed,ω_e) Representing an inverse engine torque characteristic function;

or, a desired braking torque P_brdThe calculation formula of (2) is as follows: p_brd＝F_brd/K_br，K_brIs a proportionality factor of a linear relationship between the desired braking force and the braking torque.

2. The unmanned vehicle side longitudinal coupling tracking control method based on the fast terminal sliding mode principle according to claim 1, characterized in that: step one, selecting the coordinate error X-X of the current coordinate of the center of mass position of the unmanned vehicle and the position coordinate error of a planned point on a planned track_d、Y-Y_dAnd unmanned vehicle velocity vector yaw angle is with planning some yaw angle errorFor tracking input parameters of the controller, throttle opening α_thOr braking torque F_brFront wheel steering angle_fParameters are output for the tracking controller.

3. The unmanned vehicle side longitudinal coupling tracking control method based on the fast terminal sliding mode principle according to claim 1, characterized in that: cornering stiffness C of front tire_fThe value of (c) is taken to be twice the cornering stiffness per tire.