Intelligent control method for electric vehicle
Through the design of constraint control and unknown control directions, the nonlinear correlation function and fuzzy state estimator are used to solve the stability problem of electric vehicles when the lateral speed and yaw angular velocity are unmeasurable, and the efficient handling and stability of the vehicle are achieved promote.
Patent Information
- Application Number
- CN202411085906.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-08
- Publication Date
- 2024-12-06
AI Technical Summary
The prior art is difficult to effectively control the electric vehicle when the lateral speed and yaw angular velocity are unmeasurable, especially under the influence of the longitudinal speed of the vehicle and the steering angle of the front wheel, resulting in the impact of the vehicle's stability and the lack of unknown control directions. Intelligent control method.
Using constrained control and unknown control direction design, the vehicle lateral speed and yaw angular velocity are limited within a certain range through nonlinear correlation functions and fuzzy state estimator. The controller is designed to avoid the influence of control direction on stability, and through the Lyapunov function and recursive methods achieve tracking goals for a limited time.
It is realized that the car's lateral speed and yaw angular velocity are tracked to the expected value within a limited time, reducing the impact of longitudinal speed changes and front wheel steering angle on stability, and improving the car's handling and stable performance.
Smart Images

Figure CN119078951A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of automobile stability control, and in particular to an intelligent control method for an electric vehicle. Background Art
[0002] Cars can be used as production and transportation products, as well as consumer products such as transportation, leisure, and tourism. With the development of society and economy, cars have gradually become people's inseparable partners. As autonomous driving technology gradually improves and new energy vehicles develop rapidly, people's requirements for cars are not only speed and capacity, but also require improved energy efficiency, vehicle control, comfort, and safety. Therefore, in recent years, research on autonomous electric vehicle lateral stability control systems has received widespread attention.
[0003] The lateral control of a car is mainly aimed at the control of the vehicle's steering wheel. Since the car will have a certain offset angle when turning, if the speed is too high or the angle of the steering wheel is incorrect, it may slip, causing the vehicle to be unable to maintain stability. At the same time, during the driving process of the vehicle, the steering stiffness of the wheels on different road surfaces is different, which brings about nonlinear changes in tire force. If the speed is not well controlled or encounters force majeure, the lateral stability of the vehicle will be seriously affected. In addition, the nonlinear dynamics of the tires and the changes in the longitudinal speed of the vehicle bring about parameter uncertainty.
[0004] The constraint control of a car with unknown control direction means that during the driving process of the car, due to changes in factors such as the road surface and weather, the driving direction of the car is unpredictable, which brings great challenges to the control of the car. At present, researchers have proposed some methods to solve this problem. For example, a backstepping adaptive control method based on the obstacle Lyapunov function. This method can ensure that the system state always remains within the constraint range during operation. However, no one has proposed an intelligent control method for a car that takes into account the lateral stability of the vehicle and the unknown control direction, so the present invention has certain innovation. Summary of the invention
[0005] Purpose of the invention: In order to solve the above problems, the present invention provides an intelligent control method for electric vehicles. When the lateral speed and yaw rate of the vehicle cannot be measured, the influence of the front wheel steering angle on the control torque is considered to ensure that the lateral speed and yaw rate of the vehicle can be limited to a certain range; under the influence of the longitudinal speed change of the vehicle and the front wheel steering angle, the influence of the control direction on the stability of the vehicle is avoided; considering the unknown parameters and external influences of the vehicle, the lateral speed of the vehicle can be well tracked to the expected value within a limited time; thereby improving the control and stability performance of the vehicle.
[0006] Technical solution: The present invention discloses an intelligent control method for electric vehicles, including constraint control and unknown control direction design; the constraint control uses a nonlinear correlation function to process the restrictions on the lateral speed and yaw rate of the vehicle, and the unknown control direction design avoids the influence of the control direction on the stability of the vehicle; the method includes the following steps:
[0007] Step 1: Establish the state equation of the vehicle lateral dynamic control model based on the vehicle system model;
[0008] Step 2: Considering the influence of the front wheel steering angle on the control torque, a nonlinear correlation function is constructed, and a fuzzy state estimator is designed based on it to estimate the system function after the nonlinear function transformation. A new time-varying constraint method is used to change the vehicle's lateral speed and yaw rate within a certain range;
[0009] Step 3: Introduce unknown control direction design to avoid the influence of control direction on vehicle stability, construct error variables according to the state equation and control target, and use recursive design method to complete controller design;
[0010] Step 4: Verify the designed controller, design the Lyapunov function V of the integrated closed-loop system and use the finite-time stability criterion To prove that the intelligent control method of electric vehicles can make the lateral speed and yaw rate of the vehicle reach the expected value within a limited time. is the derivative of the Lyapunov function V, where ρ1, ρ2>0, β∈(0,1).
[0011] Furthermore, in step 1, the system model of the car is:
[0012]
[0013] in, m is the total mass of the vehicle, γ is the yaw rate, is the yaw angular acceleration, F f is the lateral force of the front wheel, F γ is the lateral force on the rear wheel, v x is the longitudinal velocity of the vehicle, v y is the lateral speed of the car, is the lateral acceleration, l f is the distance from the vehicle's center of mass to the front axle, l γ is the distance from the center of mass of the vehicle to the rear axle, I is the moment of inertia of the vehicle body, M is the yaw moment of the vehicle provided by the outside, C f is the cornering stiffness of the front wheel, C γ is the cornering stiffness of the rear wheel; α f is the front wheel slip angle, α γ is the rear wheel slip angle, δ fOutput for active front wheel steering;
[0014] Define state variable x1 = v y ,x2=γ, the state equation of the system model is expressed as follows:
[0015]
[0016] Where x1 = v y is the lateral velocity of the car, x2=γ is the yaw rate, x=[x1,x2] T Restricted to In the range, R is a real number set, k i1 ,k i2 Indicates the constraint boundary of the vehicle's lateral velocity and yaw rate due to the influence of the road surface and crosswind; x i (t) represents the function of the vehicle's lateral velocity and yaw rate changing with time, which is abbreviated as x i ;
[0017]
[0018] l1 and l2 are gain coefficients. The choice of L makes A a strict Hurwitz matrix, Q = Q T >0 is a given positive definite matrix, find a positive definite matrix P = P T >0 satisfies A T P+PA=-2Q.
[0019] Furthermore, the nonlinear correlation function constructed in step 2 is as follows:
[0020]
[0021] Among them, ξ i (t) is a new variable defined by the nonlinear correlation function, abbreviated as ξ i , completely dependent on the state x i , k i1 (t), k i2 (t) represents the time-varying constraint boundary of the vehicle’s lateral velocity and yaw rate due to the influence of the road surface and crosswind, abbreviated as k i1 ,k i2 , if -k i1 <x i (t)<k i2 , then i It is bounded and can satisfy the constraints of yaw rate and lateral velocity. According to equation (3), the state equation (2) is transformed into:
[0022]
[0023] in,
[0024]
[0025] Construct a fuzzy state estimator to obtain the yaw rate and lateral velocity:
[0026] F i (ξ i ) There are unknown parameters in the system function, which are processed by fuzzy logic system. in yes The estimated value of Yes F i (ξ i ), Yes i The estimate, for The estimated value of is the adaptive law, is the fuzzy basis function; define the variable error It is a function Fuzzy approximation processing, is the estimated value after approximation; is the optimal parameter vector, and defines ε' i =ε i +D i ,δ' i =δ i +D i , there exists a normal number Make holds, where i=1,2;
[0027] The fuzzy state estimator is designed as follows:
[0028]
[0029] in, is an estimate of Y, yes The first derivative of .
[0030] Furthermore, the control objectives in step 3 include:
[0031] Goal 1: The system output y can track the expected signal y d ;
[0032] Goal 2: The lateral speed and yaw rate of the vehicle are controlled within a certain range;
[0033] Objective 3: Solve the problem of controlling the direction of the vehicle due to the influence of longitudinal speed and rotational inertia;
[0034] Objective 4: The vehicle remains stable despite disturbances in the steering angle of the front wheels.
[0035] Furthermore, the unknown control direction design introduced in step 3 is specifically as follows:
[0036] Define function N i (S) The following conditions are met:
[0037]
[0038] Among them, i=1,2, this function is used to deal with the problem of unknown control direction.
[0039] Furthermore, the controller designed in step 3 to meet the control objective is:
[0040]
[0041] Among them, α2 is the virtual control, S i is a smooth function on [0,T], υ i is the compensation error, z i is the error variable, M is the yaw moment, l2 is the gain coefficient to be designed, is the design parameter, i=1,2, is the derivative of the tracking signal after nonlinear transformation, ξ1=z1+ξ d .
[0042] Furthermore, the Lyapunov function V of the integrated closed-loop system designed in step 4 is as follows:
[0043] The Lyapunov function of the integrated closed-loop system is
[0044] Among them, i is the compensation error, P is a positive definite matrix, P = P T >0, All represent error values. for The estimated value of is the optimal parameter vector.
[0045] Beneficial effects:
[0046] The present invention aims at an electric vehicle model with unknown control direction and unknown lateral speed and yaw angular velocity, and proposes an intelligent control method with unknown control direction. The designed controller realizes the estimation of the lateral speed and yaw angular velocity of the vehicle, and at the same time reduces the influence of the front wheel steering angle on the yaw moment; reduces the influence of the longitudinal speed change on the control direction; for the influence of the road surface and crosswind, the designed controller can well limit the lateral speed and yaw angular velocity within a certain range; the proposed control scheme makes the lateral speed and yaw angular velocity tracking errors of the vehicle tend to zero domain within a limited time; the present invention realizes the constraint control of the vehicle and improves the stability of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] Figure 1 This is the two-degree-of-freedom model diagram of the car;
[0048] Figure 2 This is a flow chart of the adaptive control strategy described in an embodiment of the present invention;
[0049] Figure 3 This is a specific design thinking diagram for the automobile control of the present invention;
[0050] Figure 4 This is the overall control structure diagram of the automobile of the present invention. DETAILED DESCRIPTION
[0051] The technical solution of the present invention is described in detail below.
[0052] The present invention discloses an intelligent control method for electric vehicles, which targets electric vehicle models with unknown control direction, lateral velocity and yaw rate, including constraint control and unknown control direction design; constraint control uses nonlinear correlation functions to process the lateral velocity and yaw rate of the vehicle being restricted, and unknown control direction design avoids the influence of the control direction on the stability of the vehicle. The control method comprises the following steps:
[0053] Step 1: The system model of the car is:
[0054]
[0055] in m is the total mass of the vehicle, γ is the yaw angular velocity, is the yaw angular acceleration, F f is the lateral force of the front wheel, F γ is the lateral force on the rear wheel, v x is the longitudinal velocity of the car, is the lateral acceleration, l f is the distance from the vehicle's center of mass to the front axle, l γ is the distance from the center of mass of the vehicle to the rear axle, I is the moment of inertia of the vehicle body, M is the yaw moment of the vehicle provided by the outside, Cf is the cornering stiffness of the front wheel, C γ is the cornering stiffness of the rear wheel; α f is the front wheel slip angle, α γ is the rear wheel slip angle, δ f Active front wheel steering output.
[0056] Define state variable x1 = v y ,x2=γThe state equation of the system model can be expressed as follows:
[0057]
[0058] Where x1 = v y is the lateral velocity of the car, x2=γ is the yaw rate, x=[x1,x2] T Restricted to In the range, R is a real number set, k i1 ,k i2 It represents a constraint boundary of the lateral velocity and yaw rate of the vehicle due to the influence of the road surface and crosswind. L=[l1,l2] T ,B i =[0,1] T , C=[1,0],
[0059]
[0060] L is chosen so that A is a strict Hurwitz matrix, Q = Q T > 0 is a given positive definite matrix, we can find a positive definite matrix P = P T >0 satisfies A T P+PA=-2Q.
[0061] Step 2: To handle the problem that the lateral velocity and yaw rate of the car are constrained, a nonlinear correlation function is constructed as follows:
[0062]
[0063] From formula (3), we can know that the variable ξ i Completely dependent on state x i , if -k i1 <x i (t)<k i2 , then i It is bounded and can satisfy the constraints of yaw rate and lateral velocity. According to equation (3), equation (2) is transformed into:
[0064]
[0065] in,
[0066] Since the lateral velocity and yaw rate of the vehicle cannot be measured, a fuzzy state estimator needs to be constructed to obtain the yaw rate and lateral velocity. i (ξ i ) function has unknown parameters, which are processed by fuzzy logic system. in Yes i The estimate of variable error where θ * is the optimal parameter vector, for The estimated value of is the adaptive law, yes The estimated value of Yes F i (ξ i ), It is a function Fuzzy approximation processing, is the estimated value after approximation; define ε' i =ε i +D i ,δ' i =δ i +D i , there exists a normal number Make holds, where i=1,2.
[0067] Then, the fuzzy state estimator is designed as follows:
[0068]
[0069] Step 3 specifically includes:
[0070] Step 3.1: Control objectives include:
[0071] Goal 1: The system output y can track the expected signal y d ;
[0072] Goal 2: The lateral speed and yaw rate of the vehicle are controlled within a certain range;
[0073] Objective 3: Solve the problem of controlling the direction of the vehicle due to the influence of longitudinal speed and rotational inertia;
[0074] Objective 4: The vehicle remains stable despite the disturbance of the front wheel steering angle;
[0075] Step 3.2: Define the function N i (S) The following conditions are met:
[0076]
[0077] Among them, i=1,2, this function is used to deal with the problem of unknown control direction.
[0078] Step 3.3: The controller that meets the control objectives includes:
[0079]
[0080] Among them, α2 is the virtual control, S i is a smooth function on [0,T], υ i is the compensation error, z i is the error variable, for The estimated value of is the adaptive law, Yes i The estimate of ξ i is a new variable defined by a nonlinear correlation function, β∈(0,1), M is the yaw moment, l2 is the gain coefficient to be designed, is the design parameter, i=1,2, is the derivative of the tracking signal after nonlinear transformation, ξ1=z1+ξ d .
[0081] Step 4 verifies the designed controller, designs the Lyapunov function V of the integrated closed-loop system and uses the finite-time stability criterion To prove that the intelligent control method of electric vehicles can make the lateral speed and yaw rate of the vehicle reach the expected value within a limited time. Specifically, the following steps are included:
[0082] The Lyapunov function of the integrated closed-loop system is as follows:
[0083]
[0084] Taking the time derivative of V and integrating the calculations, we get:
[0085]
[0086] in, is a normal number, is the constant to be designed, and H is a bounded constant.
[0087] According to the following lemma: For positive constants ρ1, ρ2>0, β∈(0,1), if the continuous function satisfies When V(t) converges to the equilibrium point in a finite time, T≤t0+1 / [ρ1(1-β)]ln[(ρ1V 1-β (t0)+ρ2)ρ2], we can get T1≤t0+1 / [ηρ1(1-β)]ln[(ρ1V 1-β (t0)+ρ2)ρ2], similarly if Then T2≤t0+1 / [ρ1(1-β)]ln[(ρ1V 1-β (t0)+ηρ2)ηρ2], where t0 is the initial time, parameter η∈(0,1), υ i , It can converge to the desired area within a finite time T2, which can be expressed as the following formula:
[0088]
[0089] Next, we prove that the asymmetric time-varying full-state constraint of the system is not violated. Since υ1 is bounded, according to the command filter design is the compensation variable of command filtering, we can get is continuous and bounded, calling , we can infer that z1 is bounded and ξ1=z1+ξ d ,ξ d is the tracking signal after nonlinear transformation, and the result satisfies the state constraint of x1. Depend on Available is bounded, and because υ2 is bounded, z2 is bounded. Assume that the reference signal y d and its second-order derivative are continuous and bounded, for any k 11 (t)>0,k 22 (t)>0, there exists a positive constant Y0, ,Y1, the reference signal satisfies It can be seen that α2 is bounded. In addition, according to the formula , ψ2 and is bounded, σ2 is the time constant, α2, ψ2 are the input and output of the filter. By expressing ξ2=z2+ψ2, we can get that the boundedness of x2 is limited to -k 21 (t)<x2<k 22 (t) region. Similar to the above analysis, all states are restricted to the constraint range In the above example, no feasibility condition is required -k 11 (t)<α2<k 12 (t).
[0090] Then, we prove that the tracking error of the original system is bounded. According to inequality (10), the tracking error of the transformed system can be described as
Claims
1. An intelligent control method for electric vehicles, It is characterized in that It includes constraint control and unknown control direction design; constraint control uses nonlinear correlation functions to process the restrictions on the lateral speed and yaw rate of the vehicle, and the unknown control direction design avoids the influence of the control direction on the stability of the vehicle; it includes the following steps: Step 1: Establish the state equation of the vehicle lateral dynamic control model based on the vehicle system model; Step 2: Considering the influence of the front wheel steering angle on the control torque, a nonlinear correlation function is constructed, and a fuzzy state estimator is designed based on it to estimate the system function after the nonlinear function transformation. A new time-varying constraint method is used to change the vehicle's lateral speed and yaw rate within a certain range; Step 3: Introduce unknown control direction design to avoid the influence of control direction on vehicle stability, construct error variables according to the state equation and control target, and use recursive design method to complete controller design; Step 4: Verify the designed controller, design the Lyapunov function V of the integrated closed-loop system and use the finite-time stability criterion To prove that the intelligent control method of electric vehicles can make the lateral speed and yaw rate of the vehicle reach the expected value within a limited time. is the derivative of the Lyapunov function V, the positive constant ρ 1 ,ρ 2 >0,β∈(0,1).
2. An intelligent control method for electric vehicles according to claim 1, It is characterized in that In step 1, the system model of the car is: in m is the total mass of the vehicle, γ is the yaw angular velocity, is the yaw angular acceleration, F f is the lateral force of the front wheel, F γ is the lateral force on the rear wheel, v x is the longitudinal velocity of the vehicle, v y is the lateral speed of the car, is the lateral acceleration, l f is the distance from the vehicle's center of mass to the front axle, l γ is the distance from the center of mass of the vehicle to the rear axle, I is the moment of inertia of the vehicle body, M is the yaw moment of the vehicle provided externally, C f is the cornering stiffness of the front wheel, C γ is the cornering stiffness of the rear wheel; α f is the front wheel slip angle, α γ is the rear wheel slip angle, δ f Output for active front wheel steering; Define the state variable x 1 =v y ,x 2 =γ, the state equation of the system model is expressed as follows: y=Cx Among them, x 1 =v y is the lateral speed of the car, x 2 =γ is the yaw rate, x = [x 1 ,x 2 ] T Restricted to In the range, R is a real number set, k i1 ,k i2 Indicates the constraint boundary of the vehicle's lateral velocity and yaw rate due to the influence of the road surface and crosswind; x i (t) represents the function of the vehicle's lateral velocity and yaw rate changing with time, which is abbreviated as x i ; l 1 , l 2 is the gain coefficient, and L is chosen so that A is a strict Hurwitz matrix, Q = Q T >0 is a given positive definite matrix, find a positive definite matrix P = P T >0 satisfies A T P+PA=-2Q.
3. An intelligent control method for electric vehicles according to claim 2, It is characterized in that The nonlinear correlation function constructed in step 2 is as follows: Among them, ξ i (t) is a new variable defined by the nonlinear correlation function, abbreviated as ξ i , completely dependent on the state x i , k i1 (t), k i2 (t) represents the time-varying constraint boundary of the vehicle's lateral velocity and yaw rate due to the influence of the road surface and crosswind, abbreviated as k i1 ,k i2 , if -k i1 <x i (t)<k i2 , then i It is bounded and can satisfy the constraints of yaw rate and lateral velocity. According to equation (3), the state equation (2) is transformed into: Y=Cξ in, Construct a fuzzy state estimator to obtain the yaw rate and lateral velocity: F i (ξ i ) There are unknown parameters in the system function, which are processed by fuzzy logic system. in yes The estimated value of Yes F i (ξ i ), Yes i The estimate, for The estimated value of is the adaptive law, is the fuzzy basis function; define the variable error It is a function Fuzzy approximation processing, is the estimated value after approximation; is the optimal parameter vector, and ε′ is defined i =ε i +D i ,δ′ i =δ i +D i , there exists a normal number Make holds, where i=1,2; The fuzzy state estimator is designed as follows: in, is an estimate of Y, yes The first derivative of .
4. An intelligent control method for electric vehicles according to claim 1, It is characterized in that The control objectives in step 3 include: Goal 1: The system output y can track the expected signal y d ; Goal 2: The lateral speed and yaw rate of the vehicle are controlled within a certain range; Objective 3: Solve the problem of controlling the direction of the vehicle due to the influence of longitudinal speed and rotational inertia; Objective 4: The vehicle remains stable despite disturbances in the steering angle of the front wheels.
5. An intelligent control method for electric vehicles according to claim 3, It is characterized in that The unknown control direction design introduced in step 3 is specifically as follows: Define function N i (S) The following conditions are met: Among them, i=1,2, this function is used to deal with the problem of unknown control direction.
6. An intelligent control method for electric vehicles according to claim 5, It is characterized in that The controller designed in step 3 to meet the control objective is: Among them, α 2 is virtual control, S i is a smooth function on [0,T], υ i is the compensation error, z i is the error variable, M is the yaw moment, l 2 is the gain coefficient to be designed, c i ,θ i ,k i >0 is the design parameter, i=1,2, is the derivative of the tracking signal after nonlinear transformation, ξ 1 =z 1 +ξ d .
7. An intelligent control method for electric vehicles according to claim 1, It is characterized in that The Lyapunov function V of the integrated closed-loop system designed in step 4 is as follows: The Lyapunov function of the integrated closed-loop system is Among them, i is the compensation error, P is a positive definite matrix, P = P T >0, All represent error values. for The estimated value of is the optimal parameter vector.
