Trajectory tracking method and system based on three-degree-of-freedom vehicle model under action of strong wind

Abstract

The application discloses a trajectory tracking method and system of a three-degree-of-freedom vehicle model based on strong wind action, and a vehicle dynamics model including three degrees of freedom of longitudinal, lateral and yaw is established according to vehicle parameters. A tire model is established according to tire structure and parameters. In a vehicle coordinate system, a resultant force of the action of strong wind on the vehicle is decomposed into x, y and z direction components to establish a strong wind model. The influence of vertical load, roll moment and pitch moment is represented by the change of tire cornering stiffness to obtain a tire model of front and rear wheel cornering stiffness change. The influence of the tire model of front and rear wheel cornering stiffness change and the strong wind model is added to the three-degree-of-freedom vehicle dynamics model to obtain a three-degree-of-freedom vehicle dynamics model under the action of strong wind; and a trajectory tracking controller of the vehicle is further established to realize trajectory tracking of the vehicle. The application has high trajectory tracking accuracy and fast tracking timeliness.

Description

Technical Field

[0001] The invention discloses a trajectory tracking method and system based on a three-degree-of-freedom vehicle model under strong winds, belonging to the technical field of intelligent vehicle trajectory tracking control. Background Art

[0002] Since the beginning of the 21st century, with the rapid development of highway construction in my country, the mileage of highways and bridges in mountainous and hilly areas has also increased rapidly. The number of passenger buses and passenger and freight transport volumes has also grown daily, leading to a surge in demand for high-performance commercial vehicles. Highway tunnels in mountainous areas are often very long, and due to the deep canyons between tunnels, strong winds are easily generated due to geographical factors. When a vehicle exits a tunnel, it enters an area of ​​high crosswind intensity from a completely wind-free environment. Or, when a vehicle is driving normally on a bridge and is suddenly affected by a sea breeze, the large lateral surface area of ​​a commercial vehicle often prevents the driver from reacting due to inertia, leading to traffic accidents. High winds have a significant impact on vehicle handling stability and safety. Therefore, studying the impact of high winds on the trajectory tracking accuracy and safety of unmanned commercial vehicles has important theoretical and practical significance.

[0003] Currently, existing technologies mainly target crosswinds and consider the impact of strong winds on the vehicle's yaw moment. For example, Zhao Youqun from Nanjing University of Aeronautics and Astronautics considered the obstacle avoidance problem of high-speed vehicles under the influence of crosswinds, designed a second-order active disturbance rejection controller, and used Carsim and Simulink for joint simulation. However, they did not consider the combined effects of strong winds on the vehicle's vertical force, longitudinal force, roll moment, and pitch moment, resulting in poor trajectory tracking control accuracy. Summary of the Invention

[0004] Purpose of the invention: In order to solve the problem in the prior art that strong winds do not take into account the combined effects of vertical force, longitudinal force, roll moment and pitch moment on the vehicle, resulting in poor trajectory tracking control accuracy, the present invention provides a trajectory tracking method based on a three-degree-of-freedom vehicle model under strong winds.

[0005] Technical solution: To achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A trajectory tracking method based on a three-degree-of-freedom vehicle model under strong winds comprises the following steps:

[0007] Step 1: Establish a three-degree-of-freedom dynamic model of the vehicle including longitudinal, lateral and yaw degrees of freedom according to vehicle parameters.

[0008] Step 2: Establish a tire model based on tire structure and parameters.

[0009] Step 3, in the vehicle coordinate system, decompose the resultant force of the strong wind acting on the vehicle into x, y and z direction components, and set the distance between the resultant force point of the strong wind and the mass center of the vehicle in x, y and z directions as Δx, Δy and Δz, and establish a strong wind model.

[0010] Step 4, the influence of vertical load, roll moment and pitch moment is represented by the change of tire cornering stiffness, and the relationship between tire cornering stiffness and vertical load is established. The average value of left and right wheel cornering stiffness is taken as the cornering stiffness of front and rear wheels in the three-degree-of-freedom vehicle dynamics model under the action of strong wind, and then the tire model of front and rear wheel cornering stiffness change is obtained through the change of front and rear wheel cornering stiffness.

[0011] Step 5, the tire model of front and rear wheel cornering stiffness change, the x and y direction components F wx , F wy of the resultant force of the strong wind in the strong wind model, and the influence of the yaw moment M wz caused by the strong wind are added to the three-degree-of-freedom vehicle dynamics model to obtain the three-degree-of-freedom vehicle dynamics model under the action of strong wind.

[0012] Step 6, according to the strong wind model and the three-degree-of-freedom vehicle dynamics model under the action of strong wind, vehicle trajectory tracking control is carried out.

[0013] Preferably, the dynamics and kinematics equations of the three-degree-of-freedom vehicle dynamics model under the action of strong wind in step 5 are as follows:

[0014]

[0015]

[0016]

[0017]

[0018]

[0019] In the formula, m is the mass of the vehicle, v x is the lateral speed of the front wheel, v y is the longitudinal speed, is the yaw angular velocity, F xf is the longitudinal force on the front wheel, F xr is the longitudinal force on the rear wheel, δ f is the front wheel steering angle, C f is the cornering stiffness of the front wheel, C r is the cornering stiffness of the rear wheel, α f , α r are the front and rear wheel cornering angles, C Dx is the air coefficient in x direction, CDy C is the air coefficient in the y direction, C Dz C is the air coefficient in the z direction, p is the air density, A is the frontal area of the vehicle, v rx v is the relative wind speed in the x direction, v ry v is the relative wind speed in the y direction, v rz v is the relative wind speed in the z direction, I z I is the moment of inertia of the vehicle body about the z axis, a is the distance from the center of mass to the front axle, and b is the distance from the center of mass to the rear axle.

[0020] Preferably, in step 4, when calculating the change in vertical load of the wheels, it is assumed that the vertical load, roll moment and pitch moment caused by the strong wind are positive, and the strong wind roll moment causes the left and right side wheel vertical load change amount to be the pitch moment causes the front and rear wheel vertical load change amount to be the vertical load causes the front and rear wheel vertical load change amount to be

[0021] The vehicle four wheel vertical load change amount is:

[0022]

[0023]

[0024]

[0025]

[0026] where, ΔF wx is the left and right side wheel vertical load change amount caused by the strong wind roll moment, M wx is the x direction wind generated moment, B is the track, ΔF wy is the front and rear wheel vertical load change amount caused by the pitch moment, M wy is the y direction wind generated moment, L is the wheelbase, ΔF zf is the front wheel vertical load change amount caused by the vertical load, b is the distance from the center of mass to the rear axle, ΔF z is the vehicle vertical load change amount, ΔF zr is the rear wheel vertical load change amount caused by the vertical load, a is the distance from the center of mass to the front axle, ΔF fl is the left front wheel vertical load change amount, ΔF fr is the right front wheel vertical load change amount, ΔF rl is the left rear wheel vertical load change amount, ΔF rr is the right rear wheel vertical load change amount.

[0027] Preferably, the strong wind model includes the force and moment of the strong wind, which are as follows:

[0028]

[0029]

[0030] wherein F wk is the force of the strong wind, M wk is the moment of the strong wind, p is the air density, v rk is the relative wind speed in the k direction, C Dk is the air coefficient in the k direction, A is the windward area of the vehicle, and Ak is the position of the resultant force of the strong wind from the center of mass in the k axis direction, wherein k = x, y, z.

[0031] Preferably, the tire model in step 2 includes the relationship between the longitudinal force and vertical force of the wheel, the longitudinal slip ratio, and the lateral force of the front and rear wheels.

[0032] The relationship between the longitudinal force and vertical force of the wheel is as follows:

[0033]

[0034] The longitudinal slip ratio is as follows:

[0035]

[0036] wherein F x is the longitudinal force of the wheel, p is the road adhesion coefficient, F z is the vertical force of the wheel, s is the longitudinal slip ratio, r is the rolling radius of the wheel, w is the angular velocity of the wheel, and v x is the longitudinal speed of the wheel.

[0037] The normal force of the front and rear wheels is as follows: F zf is the normal force of the front wheel, and F zr is the normal force of the rear wheel, g is the acceleration of gravity, and the wheelbase L = a + b.

[0038] The lateral force of the front and rear wheels is as follows:

[0039] F yf = C f a f ;

[0040] F yr = C r a r ;

[0041] wherein C f is the cornering stiffness of the front wheel, and af is the front wheel yaw angle, C r is the cornering stiffness of the rear wheel, a r is the rear wheel yaw angle.

[0042] Preferably, under the assumption of small angle of the wheel,

[0043] Preferably, when the action point of the large wind resultant force is not located at the center of mass of the vehicle in step 3, the vehicle will also bear the action of the yaw moment, the roll moment and the pitch moment.

[0044] Another object of the present application is to provide a trajectory tracking control system based on a three-degree-of-freedom vehicle model under the action of large wind, comprising an MPC controller, a state estimator, a turning angle matching unit, a large wind model unit, and a comparison unit, the MPC controller comprising a prediction model unit, a target function constraint condition unit, and an optimal solution unit, wherein:

[0045] The prediction model unit is provided with a three-degree-of-freedom vehicle dynamics model under the action of large wind.

[0046] The large wind model unit obtains state data of the vehicle under the action of large wind according to the action of large wind.

[0047] The state estimator receives real-time parameters of the controlled vehicle and state data under the action of large wind in real time to obtain the latest estimated state.

[0048] The comparison unit compares the reference trajectory with the latest control trajectory to obtain input parameters.

[0049] The MPC controller controls the front wheel turning angle based on the three-degree-of-freedom vehicle dynamics model under the action of large wind according to the input parameters, and optimizes the solution through the target function constraint condition unit and the optimal solution unit to output the optimal control value.

[0050] Another object of the present application is to provide an electronic device comprising at least one processor, at least one memory, and a communication interface. The processor, memory, and communication interface communicate with each other. The memory stores program instructions executable by the processor, and the processor invokes the program instructions to execute the trajectory tracking method based on the three-degree-of-freedom vehicle model under the action of large wind.

[0051] Another object of the present application is to provide a computer-readable storage medium, which stores program instructions that, when executed by a processor, cause the processor to execute the trajectory tracking method based on the three-degree-of-freedom vehicle model under the action of large wind.

[0052] Compared with the prior art, the present application has the following beneficial effects:

[0053] 1. The x, y direction component force F wx , F wy and yaw moment M wz caused by the large wind are added to the three-degree-of-freedom vehicle dynamics model, and the tracking trajectory accuracy is high by adding the x, y direction component force F wx , F wy and yaw moment M wz caused by the large wind.

[0054] 2. Only a three-degree-of-freedom model is used, and the pitch moment and roll moment are converted into the side stiffness variation of the front and rear wheels through the relationship between the tire vertical load and the side stiffness; finally, due to the reduction of the two degrees of freedom of pitch and roll, the calculation amount of the algorithm is reduced, the calculation time is shortened, and the timeliness of tracking is improved. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 is a three-degree-of-freedom vehicle dynamics model.

[0056] Figure 2 is a tire dynamics model.

[0057] Figure 3 is the relationship between the side stiffness of the tire and the vertical load.

[0058] Figure 4 is a model predictive control prediction model with a large wind force model.

[0059] Figure 5 is a vehicle body coordinate system.

[0060] Figure 6 is a process of converting the large wind moment into the side stiffness of the front and rear wheels.

[0061] Figure 7 is a double lane shift reference trajectory and fan position under the overtaking working condition.

[0062] Figure 8 is a trajectory tracking effect diagram under the condition of 10 m / s lateral wind speed and road adhesion coefficient 0.8.

[0063] Figure 9 is a yaw angle change over time diagram under the condition of 10 m / s lateral wind speed and road adhesion coefficient 0.8.

[0064] Figure 10 is a yaw angular velocity change over time diagram under the condition of 10 m / s lateral wind speed and road adhesion coefficient 0.8.

[0065] Figure 11 is a trajectory tracking effect diagram under the condition of 10 m / s lateral wind speed and road adhesion coefficient 0.5.

[0066] Figure 12 Figure 6 is a lateral angle-time graph under the condition of a lateral wind speed of 10 m / s and a road adhesion coefficient of 0.5.

[0067] Figure 13 Figure 7 is a lateral angle velocity-time graph under the condition of a lateral wind speed of 10 m / s and a road adhesion coefficient of 0.5.

[0068] Figure 14 Figure 8 is a trajectory tracking effect graph under the condition of a lateral wind speed of 10 m / s and a road adhesion coefficient of 0.3.

[0069] Figure 15 Figure 9 is a lateral angle-time graph under the condition of a lateral wind speed of 10 m / s and a road adhesion coefficient of 0.3.

[0070] Figure 16 Figure 10 is a lateral angle velocity-time graph under the condition of a lateral wind speed of 10 m / s and a road adhesion coefficient of 0.3. DETAILED DESCRIPTION

[0071] The present application will be further illustrated below in conjunction with the accompanying drawings and specific examples. It should be understood that these examples are only used to illustrate the present application and are not used to limit the scope of the present application. After reading the present application, those skilled in the art can make various modifications to the equivalent forms of the present application, which fall within the scope defined by the appended claims.

[0072] EMBODIMENT

[0073] During the driving of the automobile, various road conditions and environmental disturbances are faced, such as a very deep canyon between tunnels, which is easy to cause strong wind due to geographical factors. When the vehicle leaves the tunnel, it will enter the high-intensity side wind area from the environment without side wind, or when the vehicle is normally driving on the bridge and is suddenly affected by the sea wind. Due to the influence of the wind force in each direction, the front and lateral area of the unmanned commercial vehicle is large, and it is necessary to prevent the vehicle from being affected by the strong wind to cause the risk of fishtailing and sliding, etc. Therefore, it has important theoretical significance and practical value to study the influence of strong wind on the trajectory tracking accuracy and safety of the unmanned commercial vehicle. For this purpose, a trajectory tracking method based on a three-degree-of-freedom vehicle model under the action of strong wind is proposed, which includes the following steps:

[0074] Step 1, a three-degree-of-freedom vehicle dynamics model including longitudinal, lateral and yaw three degrees of freedom is established according to the vehicle parameters.

[0075] In order to simplify the dynamic model of the vehicle, the present embodiment assumes that the wheel side slip angle is small, and the mass of the vehicle is the sprung mass, and the mass of the suspension and the wheel is ignored. A model including longitudinal, lateral and yaw three degrees of freedom is established, as shown in Figure 1 、 Figure 5 x ​is the lateral velocity of the front wheels, v y is the longitudinal velocity, a is the distance from the mass center to the front axle, b is the distance from the mass center to the rear axle, F xf , F xr are the longitudinal forces on the front and rear wheels, respectively, F yf , F yr are the lateral forces on the front and rear wheels, respectively, δ f are the front wheel steering angles, α f is the front wheel yaw angle, α r is the rear wheel yaw angle, is the yaw rate, F wx , F wy , M wz are the x and y components of the resultant wind force and the yaw moment due to the wind, respectively.

[0076] The dynamic and kinematic equations of the three-degree-of-freedom vehicle dynamics model are:

[0077]

[0078]

[0079]

[0080]

[0081]

[0082] where m is the vehicle mass, v x is the lateral velocity of the front wheels, v y is the longitudinal velocity, is the yaw rate, F xf is the longitudinal force on the front wheels, F xr is the longitudinal force on the rear wheels, δ f are the front wheel steering angles, C f is the front wheel cornering stiffness, C r is the rear wheel cornering stiffness, α f , α r are the front and rear wheel cornering angles, C Dx is the x-direction air coefficient, C Dy is the y-direction air coefficient, C Dz is the z-direction air coefficient, p is the air density, A is the vehicle frontal area, v rx is the x-direction relative wind speed, v ry is the y-direction relative wind speed, v rz is the z-direction relative wind speed, I zIt is expressed as the moment of inertia of the vehicle body around the z-axis, a is the distance from the center of mass to the front axle, b is the distance from the center of mass to the rear axle, F wx 、F wy 、M wz They are the components of the strong wind force in the x and y directions and the yaw moment caused by the strong wind.

[0083] Step 2: Establish a tire model based on tire structure and parameters.

[0084] Among the existing nonlinear tire models, the Pacejka tire model is widely used in the industry. It determines the structure and parameters of the model by measuring tire data and can better describe tire behavior and performance. According to the assumptions of this paper, the Pacejka tire model can be described under linear working conditions. Its dynamic model is as follows: Figure 2 The tire model includes the relationship between the longitudinal force and vertical force of the wheel, the longitudinal slip rate, and the lateral force of the front and rear wheels.

[0085] The relationship between the longitudinal force and the vertical force of the wheel is:

[0086] F x =μF z ;

[0087] The longitudinal slip rate is:

[0088]

[0089] Among them, F x is the longitudinal force of the wheel, μ is the road adhesion coefficient, F z is the vertical force of the wheel, s is the longitudinal slip rate, r is the rolling radius of the wheel, ω is the angular velocity of the wheel, v x is the longitudinal speed of the wheel;

[0090] The normal forces on the front and rear wheels are F zf is the normal force on the front wheel, F zr is the normal force on the rear wheel, g is the acceleration due to gravity, and the wheelbase L = a + b.

[0091] The lateral forces on the front and rear wheels are:

[0092] F yf =C f α f ;

[0093] F yr =C r α r ;

[0094] Among them, C f is the cornering stiffness of the front wheel, α f is the front wheel yaw angle, Cr is the cornering stiffness of the rear wheel, a r is the yaw angle of the rear wheel.

[0095] Under the assumption of small angle of the wheel,

[0096] Step 3, in the vehicle coordinate system, the resultant force of the strong wind acting on the vehicle is decomposed into x, y and z direction components, when the action point of the resultant force of the strong wind is not located at the vehicle mass center position, the vehicle will also bear the action of the yaw moment, roll moment and pitch moment. The action point of the resultant force of the strong wind is set as Δx, Δy and Δz in the x, y and z directions from the vehicle mass center position, and the strong wind model is established.

[0097] This paper assumes that the vehicle will not slip under the action of the strong wind. In the vehicle coordinate system, the resultant force of the strong wind acting on the vehicle is decomposed into x, y and z direction components, and are respectively represented by F wx , F wy , and F wz . When the action point of the resultant force of the strong wind is not located at the vehicle mass center position, the vehicle will also bear the action of the yaw moment, roll moment and pitch moment. The action point of the resultant force of the strong wind is set as Δx, Δy and Δz in the x, y and z directions from the vehicle mass center position, and the strong wind model includes the action force and its moment generated by the strong wind. According to the theory of aerodynamics and theoretical mechanics, the action force and its moment generated by the strong wind are as follows:

[0098]

[0099]

[0100] In the formula, F wk is the action force generated by the strong wind, M wk is the moment generated by the strong wind, p is the air density, v rk is the relative wind speed in the k direction, C Dk is the air coefficient in the k direction, A is the windward area of the vehicle, and Δk is the position of the action point of the resultant force of the strong wind in the k axis direction from the mass center, where k=x, y, z.

[0101] Step 4, as shown in Figure 6 , the influence of vertical load, roll moment and pitch moment is represented by the change of tire cornering stiffness, and the relationship between tire cornering stiffness and vertical load is established. The average value of the left and right wheel cornering stiffness is taken as the cornering stiffness of the front and rear wheels in the three-degree-of-freedom vehicle dynamics model under the action of the strong wind, and then the tire model of the change of the front and rear wheel cornering stiffness is obtained through the change of the front and rear wheel cornering stiffness.

[0102] In this paper, the x and y direction components F wx , F wyand the yaw moment M caused by the strong wind wz are added into the three-DOF vehicle dynamics model. Since the vertical load, roll moment and pitch moment are not considered in the three-DOF vehicle model, the effects of the vertical load, roll moment and pitch moment are represented by the change of the tire cornering stiffness in this paper, and the average of the left and right wheel cornering stiffness is taken as the cornering stiffness of the front and rear wheels in the model. The relationship between the tire cornering stiffness and the vertical load is shown in Fig. 2. Figure 1 Figure 3

[0103] When calculating the change of the wheel vertical load, it is assumed that the vertical load, roll moment and pitch moment caused by the strong wind are positive (the signs are opposite when they are negative), the change of the left and right wheel vertical load caused by the roll moment of the strong wind is the change of the front and rear wheel vertical load caused by the pitch moment of the strong wind is the change of the front and rear wheel vertical load caused by the vertical load is

[0104] The change of the vertical load of the four wheels of the vehicle is:

[0105]

[0106]

[0107]

[0108]

[0109] where, ΔF wx is the change of the left and right wheel vertical load caused by the roll moment of the strong wind, M wx is the moment caused by the x-direction strong wind, B is the wheel track, ΔF wy is the change of the front and rear wheel vertical load caused by the pitch moment, M wy is the moment caused by the y-direction strong wind, L is the wheel base, ΔF zf is the change of the front wheel vertical load caused by the vertical load, b is the distance from the mass center to the rear axle, ΔF z is the change of the vehicle vertical load, ΔF zr is the change of the rear wheel vertical load caused by the vertical load, a is the distance from the mass center to the front axle, ΔF fl is the change of the left front wheel vertical load, ΔF fr is the change of the right front wheel vertical load, ΔF rl is the change of the left rear wheel vertical load, ΔF rr is the change of the right rear wheel vertical load.

[0110] ​​Step 5, the influence of the tire model of the front wheel side stiffness variation, the x, y direction component force F of the wind model of the wind force is added to the three freedom dynamics model of the vehicle, and the three freedom dynamics model of the vehicle under the action of the wind is obtained. wx wy And the yaw moment M caused by the wind wz The dynamics and kinematics equation of the three freedom dynamics model of the vehicle under the action of the wind is:

[0111] The dynamics and kinematics equation of the three freedom dynamics model of the vehicle under the action of the wind is:

[0112]

[0113]

[0114]

[0115]

[0116]

[0117] In the formula, m is the mass of the vehicle, v x is the lateral velocity of the front wheel, v y is the longitudinal velocity, is the yaw angular velocity, F xf is the longitudinal force received by the front wheel, F xr is the longitudinal force received by the rear wheel, δ f is the front wheel rotation angle, respectively, C f is the side stiffness of the front wheel, C r is the side stiffness of the rear wheel, α f , α r is the front, rear wheel side angle, respectively, C Dx is the air coefficient in the x direction, C Dy is the air coefficient in the y direction, C Dz is the air coefficient in the z direction, ρ is the air density, A is the windward area of the vehicle, v rx is the relative wind speed in the x direction, v ry is the relative wind speed in the y direction, v rz is the relative wind speed in the z direction, I z represents the moment of inertia of the vehicle body around the z axis, a is the distance from the mass center to the front axle, and b is the distance from the mass center to the rear axle.

[0118] Step 6, according to the wind model and the three freedom dynamics model of the vehicle under the action of the wind, trajectory tracking control is carried out.

[0119] ​Since the three-degree-of-freedom vehicle dynamics model is used in this paper, the vertical load, roll moment and pitch moment are not considered, and the influence of the vertical load, roll moment and pitch moment can only be represented by the change of the wheel adhesion and side stiffness, and the average value of the left and right wheel side stiffness is taken as the side stiffness of the front and rear wheels in the model. The side stiffness of the tire is related to the vertical load as shown in the following figure. In order to facilitate the simulation of the effect of the strong wind in the Carsim software, it is assumed that the strong wind acting on the vehicle is in the Y direction of the vehicle coordinate system, and the strong wind is given by the Carsim fan.

[0120] In order to compensate for the error of the state quantity and the control quantity under the influence of the strong wind, a strong wind model is introduced into the MPC controller in this paper, and the principle diagram of the vehicle model predictive control is as shown in Figure 4

[0121] Therefore, in another embodiment, a trajectory tracking control system based on a three-degree-of-freedom vehicle model under the action of strong wind is provided, as shown in Figure 4 The MPC controller includes a prediction model unit, a target function constraint condition unit, an optimal solution unit, wherein:

[0122] The prediction model unit is provided with a three-degree-of-freedom vehicle dynamics model under the action of strong wind.

[0123] The strong wind model unit obtains the state data of the vehicle under the action of strong wind under the action of strong wind.

[0124] The state estimator receives the real-time parameters of the controlled vehicle and the state data under the action of strong wind in real time to obtain the latest estimated state.

[0125] The comparison unit compares the reference trajectory with the latest control trajectory to obtain the input parameters.

[0126] The MPC controller controls the front wheel steering angle based on the three-degree-of-freedom vehicle dynamics model under the action of strong wind according to the input parameters, and optimizes the solution through the target function constraint condition unit and the optimal solution unit to output the optimal control value.

[0127] (1) Model prediction

[0128] The prediction model is based on a discretized linear time-varying model, and the system prediction output expression can be obtained as follows:

[0129] Y(t) = ψ t ξ(tt) + Θ t ΔU(t);

[0130] Wherein,

[0131] ​

[0132] Y(t) is the control output vector, ψ t , Θ t is the coefficient matrix, ξ(t|t) is the system state quantity based on time t, ΔU(t) is the control increment of the system, A, B, C are the coefficient matrices.

[0133] wherein,

[0134]

[0135]

[0136]

[0137] (2) Rolling optimization

[0138] The determination of the optimization objective function is based on minimizing the vehicle state error and control input increment, optimizing the system state error under the constraint condition, and avoiding the mutation of the actuator output, so as to maintain the continuity of the control input. The objective function used in this paper is in the form of:

[0139]

[0140] The constraint conditions are as follows:

[0141] -12°<β<12°, good road surface;

[0142] -2°<β<2°, icy road surface;

[0143] -15°<α f <15°;

[0144]

[0145] ε>0;

[0146] ΔU min ≤ΔU t ≤ΔU max ;

[0147] ΔU min ≤AΔU t +ΔU t ≤ΔU max ;

[0148] In the formula, η(k+i|t), ΔU(k+i|t) are the system output state quantity and the reference output state quantity at time t+i based on time t, respectively; Q, R, ρ are weight coefficients; N p is the prediction time domain; N cwhere T is the control time domain; ε is the relaxation factor. The first term reflects the system's ability to follow the reference trajectory, and the second term reflects the requirement for smooth changes in the control variable. The function of the entire expression is to make the system track the desired trajectory quickly and smoothly.

[0149] (3) Feedback correction

[0150] After solving the objective function in each control cycle, the control input increment sequence in the control time domain is as follows:

[0151]

[0152] The first element of the control sequence (front wheel steering angle) acts on the controlled vehicle, i.e., the actual control input increment:

[0153]

[0154] After entering the next control cycle, the loop continues the above steps to achieve trajectory tracking of the autonomous vehicle.

[0155] Let the state quantity of the vehicle be where is the longitudinal velocity, is the lateral velocity, is the yaw angle at the center of mass of the vehicle, is the yaw rate at the center of mass of the vehicle, X is the longitudinal displacement, Y is the lateral displacement, and ΔF zij is the change in vertical load received by the four wheels. The output control variable of the controller with the added wind model is u d = δ f In the trajectory tracking controller, the established wind model is added to the vehicle dynamic model as a prediction model.

[0156] In another embodiment, an electronic device is provided, comprising: at least one processor, at least one memory, and a communication interface. The processor, memory, and communication interface communicate with each other. The memory stores program instructions executable by the processor, and the processor invokes the program instructions to execute the trajectory tracking method based on the three-degree-of-freedom vehicle model under the action of the wind.

[0157] In another embodiment, a computer-readable storage medium is provided, which stores program instructions that, when executed by a processor, cause the processor to execute the trajectory tracking method based on the three-degree-of-freedom vehicle model under the action of the wind.

[0158] Simulation

[0159] To verify the effectiveness of vehicle trajectory tracking control under strong winds, a vehicle simulation model was built in Carsim software. A van model provided with the software was selected, with its main parameters shown in Table 1. The vehicle state, road adhesion coefficient, and wind speed were set as inputs, and the front wheel angle was set as output. Simulink software was also used to build a three-degree-of-freedom vehicle dynamics model, a Dugoff tire model, a road adhesion coefficient estimation model based on the unscented Kalman filter algorithm, and a strong wind model. Carsim and Simulink were then used for joint simulation.

[0160] Table 1 Vehicle dynamic parameters

[0161]

[0162] The double lane change test, also known as the dynamic roll test, is mainly used to test the sharp turning maneuvers of the vehicle's steering angle, which are rapidly reversed. It is an internationally accepted test item. Because it can effectively evaluate the trajectory tracking level of the control algorithm and reflect the vehicle's handling stability, it is often used in vehicle dynamics simulation and stability testing. The double lane change reference trajectory and fan position set in the Carsim software in this article are as follows: Figure 10 The simulation test used fans on one side of the road to simulate high side wind conditions. Simulations were performed using Carsim / Simulink, and compared with simulations without high winds.

[0163] The test vehicle adopts front wheel steering control, assuming that the wind speed is constant, the wind speed is V = 10m / s, simulating the situation of level 5 gale, the wind speed direction is along Figure 7 The Y-axis direction remains unchanged. Assuming the vehicle's speed remains constant, the wind direction and torque acting on the vehicle body will change as the vehicle travels along a double lane. This article calculates the changes in force and torque caused by high winds based on the wheel force output and high wind model of the Carsim software.

[0164] Under the condition of simulating high-speed driving on a good road, the road adhesion coefficient is set to 0.8, the simulation speed is 96km / h, and the wind speed is 10m / s. The simulation results are as follows: Figures 8-10 As shown in the figure, compared to the controller without the high wind model, the trajectory tracking control with the high wind model adheres closer to the reference trajectory and achieves better trajectory tracking. Furthermore, the vehicle's yaw angle is reduced by approximately 2°, or 15%, compared to the control without the high wind model. It reaches a stable state after 7 seconds, demonstrating faster convergence. The yaw rate also decreases to approximately 12° / s, a decrease of approximately 25%, and reaches stability after 8 seconds.

[0165] Under the conditions of simulating general road urban driving speed, the road adhesion coefficient is set to 0.5, the speed is 72km / h, and the wind speed is 10m / s. The simulation results are as follows Figures 11-13The tracking trajectory with the controller with the wind model is more consistent with the reference trajectory, and the trajectory tracking control effect is better. The vehicle's yaw angle is controlled within 10°, and the yaw angular velocity is controlled within 15° / s. Compared with the controller without the wind model, the yaw angle is reduced by about 2°, which is reduced by about 16%, and the yaw angular velocity is reduced by about 6° / s, which is reduced by about 23%. The yaw angle and the yaw angular velocity tend to stabilize faster than the control without the wind model.

[0166] In the simulation of trajectory tracking control on icy road surface, the road adhesion coefficient is set to 0.3, the vehicle speed is 54km / h, and the wind speed received by the vehicle is 10m / s. The simulation results are shown in Figures 14-16 The control trajectory with the wind model is basically consistent with the reference trajectory, and the yaw angle is controlled within 7°, and the yaw angular velocity is controlled within 4° / s. Compared with the controller without the wind model, the yaw angle is reduced by about 4°, which is reduced by about 30%, and the yaw angular velocity is reduced by about 4° / s, which is reduced by about 20%.

[0167] From the above simulation analysis, in the double-migration dynamic simulation, under different road adhesion coefficients and different vehicle speed conditions, the tracking trajectory with the wind model is more consistent with the reference trajectory than the controller without the wind model, and the trajectory tracking control effect is better. The vehicle's yaw angle and yaw angular velocity are smaller than the control without the wind model, and the stabilization speed is faster. The simulation results verify the effectiveness of the trajectory tracking control with the wind model.

[0168] Under the condition of high-speed driving on good road surface, the yaw angle is reduced by about 15% and the yaw angular velocity is reduced by about 31% with the wind model controller compared with the controller without the wind model. Under the condition of urban driving at a certain speed on general road surface, the yaw angle is reduced by about 16% and the yaw angular velocity is reduced by about 23%. In the simulation of trajectory tracking control on icy road surface, the yaw angle is reduced by about 30% and the yaw angular velocity is reduced by about 20%. The trajectory tracking control is effectively realized, and the vehicle driving safety is improved.

[0169] In the double-migration dynamic simulation, under different vehicle speed conditions, the tracking trajectory with the wind model is more consistent with the reference trajectory than the controller without the wind model. On the road surface with adhesion coefficients of 0.8, 0.5 and 0.3, the vehicle's yaw angle and yaw angular velocity are smaller than the control without the wind model, and the stabilization speed is faster. By comparing the estimation results of high, medium and low road adhesion coefficients, it can be found that the trajectory tracking control effect is better on the road surface with high adhesion coefficient. The simulation results verify the effectiveness of the trajectory tracking control with the wind model.

[0170] The above merely preferred embodiments of the present application, it should be noted that for those of ordinary skill in the art, without departing from the principles of the present application, can also make a number of improvements and refinements, these improvements and refinements should also be considered within the scope of the present application.

Claims

1. A trajectory tracking method based on a three-degree-of-freedom vehicle model of action of strong wind, characterized by, The method comprises the following steps: Step 1, a vehicle dynamics model including three degrees of freedom of longitudinal, lateral and yaw is established according to vehicle parameters; Step 2, a tire model is established according to tire structure and parameters; Step 3, in the vehicle coordinate system, the resultant force of the strong wind on the vehicle is decomposed into x, y and z direction components, and the points of action of the resultant force of the strong wind in the x, y and z directions are set as Δx, Δy and Δz from the position of the vehicle center of mass, and a strong wind model is established; Step 4, the influence of vertical load, roll moment and pitch moment is represented by the change of tire cornering stiffness, the relationship between tire cornering stiffness and vertical load is established, and the average values of the cornering stiffness of the front left, right and rear left, right wheels are taken as the cornering stiffness of the front and rear wheels in the three-degree-of-freedom vehicle dynamics model under the action of the strong wind, and then the tire model of the change of the cornering stiffness of the front and rear wheels is obtained through the change of the cornering stiffness of the front and rear wheels; In calculating the change of the vertical load of the wheels, it is assumed that the vertical load, the roll moment and the pitch moment generated by the strong wind are positive, the roll moment of the strong wind causes the change of the vertical load of the left and right wheels to be the pitch moment causes the change of the vertical load of the front and rear wheels to be the vertical load causes the change of the vertical load of the front and rear wheels to be The change amount of the vertical load of the four wheels of the vehicle is: wherein, ΔF wx is the vertical load variation of the left and right wheels caused by the large wind roll moment, M wx is the moment generated by the x-direction large wind, B is the wheel track, ΔF wy is the vertical load variation of the front and rear wheels caused by the pitch moment, M wy is the moment generated by the y-direction large wind, L is the wheelbase, ΔF zf is the vertical load variation of the front wheels caused by the vertical load, b is the distance from the center of mass to the rear axle, ΔF z is the vertical load variation of the vehicle, ΔF zr is the vertical load variation of the rear wheels caused by the vertical load, a is the distance from the center of mass to the front axle, ΔF fl is the vertical load variation of the left front wheel, ΔF fr is the vertical load variation of the right front wheel, ΔF rl is the vertical load variation of the left rear wheel, ΔF rr is the vertical load variation of the right rear wheel. Step 5, the x, y direction components of the large wind resultant force F wx , and the yaw moment M wz caused by the large wind are added to the three-DOF vehicle dynamics model to obtain a three-DOF vehicle dynamics model under the action of the large wind; Step 6, according to the strong wind model and the three-degree-of-freedom vehicle dynamics model under the action of the strong wind, a trajectory tracking controller of the vehicle is established to track the trajectory.

2. The trajectory tracking method based on the three-degree-of-freedom vehicle model of large wind action according to claim 1, characterized in that, The dynamics and kinematics equations of the three-degree-of-freedom vehicle dynamics model under the action of the strong wind in step 5 are: where m is the vehicle mass, v x is the lateral velocity of the front wheels, v y is the longitudinal velocity, is the yaw rate, F xf is the longitudinal force on the front wheels, F xr is the longitudinal force on the rear wheels, δ f is the front wheel steering angle, C f , C r are the front and rear cornering stiffnesses, α f , α r are the front and rear side slip angles, C Dx is the x-direction air coefficient, C Dy is the y-direction air coefficient, C Dz is the z-direction air coefficient, p is the air density, A is the vehicle frontal area, v rx is the x-direction relative wind speed, v ry is the y-direction relative wind speed, v rz is the z-direction relative wind speed, I z is the moment of inertia about the z-axis of the vehicle body, a is the distance from the center of mass to the front axle, and b is the distance from the center of mass to the rear axle.

3. The trajectory tracking method based on the three-degree-of-freedom vehicle model of large wind action according to claim 2, characterized in that, The strong wind model includes the force and moment generated by the strong wind, which are as follows: where F wk is the force generated by the wind, M wk is the moment generated by the wind, p is the air density, v rk is the relative wind speed in the k direction, C Dk is the air coefficient in the k direction, A is the wind- facing area of the vehicle, and Δk is the position of the wind resultant force point in the k-axis direction from the center of mass, where k = x, y, z.

4. The trajectory tracking method based on the three-degree-of-freedom vehicle model of large wind action according to claim 3, characterized in that: The tire model in step 2 includes the relationship between the longitudinal force and vertical force of the wheel, the longitudinal slip ratio, and the lateral force of the front and rear wheels; The relationship between the longitudinal force and vertical force of the wheel is: F x = μF z ; The longitudinal slip ratio is: where F x is the longitudinal force of the wheel, μ is the road adhesion coefficient, F z is the vertical force of the wheel, s is the longitudinal slip ratio, r is the rolling radius of the wheel, ω is the angular velocity of the wheel, v x is the longitudinal velocity of the wheel; The normal forces of the front and rear wheels are respectively F zf F is the normal force of the front wheel, F zr F is the normal force of the rear wheel, g is the acceleration of gravity, and the wheelbase L=a+b. The lateral force of the front and rear wheels is: F yf = C f α f ; F yr = C r α r ; where C f is the cornering stiffness of the front wheels, a f is the front wheel alignment angle, C r is the cornering stiffness of the rear wheels, a r is the rear wheel alignment angle.

5. The trajectory tracking method based on the three-degree-of-freedom vehicle model of large wind action according to claim 4, characterized in that: Under the assumption of small angle of the wheel, 6. The trajectory tracking method based on the three-degree-of-freedom vehicle model of large wind action according to claim 5, characterized in that: When the point of action of the resultant force of the strong wind is not located at the position of the vehicle center of mass in step 3, the vehicle will also bear the action of the yaw moment, roll moment and pitch moment.

7. A trajectory tracking control system based on a three-degree-of-freedom vehicle model of action of strong wind, characterized by, To realize the trajectory tracking method of the three-degree-of-freedom vehicle model based on the action of the strong wind according to any one of claims 1 to 6, comprising an MPC controller, a state estimator, a steering angle matching unit, a strong wind model unit and a comparison unit, wherein: The prediction model unit is provided with a three-degree-of-freedom vehicle dynamics model under the action of the strong wind; The strong wind model unit obtains the state data of the vehicle under the action of the strong wind according to the action of the strong wind; The state estimator receives the real-time parameters of the controlled vehicle and the state data under the action of the strong wind in real time to obtain the latest estimated state; The comparison unit compares the reference trajectory with the latest control trajectory to obtain the input parameters; The MPC controller predicts the vehicle speed and front wheel steering angle based on the three-degree-of-freedom vehicle dynamics model under the action of the strong wind according to the input parameters, and optimizes and solves through the target function constraint unit and the optimal solution unit to output the optimal control value.

8. An electronic device, comprising: It comprises: At least one processor, at least one memory and a communication interface; the processor, the memory and the communication interface communicate with each other; The memory stores program instructions executable by the processor, and the processor invokes the program instructions to execute the trajectory tracking method of the three-degree-of-freedom vehicle model based on the action of the strong wind according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores program instructions, which when executed by the processor, cause the processor to perform the trajectory tracking method of the three-degree-of-freedom vehicle model based on the action of strong wind according to any one of claims 1 to 6.

Images