Steering control method and system for distributed driving automobile

Abstract

The invention discloses a steering control method and system for a distributed driving automobile, and relates to the technical field of automobile control. And a complex hierarchical structure in a traditional control architecture further reduces the control efficiency. Constructing a steering motor model of the distributed driving automobile, and a two-degree-of-freedom model comprising a yaw velocity and a side slip angle; performing decoupling processing on the two-degree-of-freedom model, and establishing a plurality of single-wheel dynamic models; a steering motor model is combined to construct a steering integrated error model; performing state space reconstruction on the error model to obtain a controllable state equation; designing a fractional order terminal sliding mode surface according to the equation, and generating a virtual control law; establishing a full-order terminal sliding mode surface according to the virtual control law, and generating an actual control law; and on the basis of the virtual control law and the actual control law, a four-wheel steering angle control signal and a yaw moment control signal are generated so as to perform steering control on the distributed driving automobile. The dynamic response speed and the anti-interference capability of vehicle steering control can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automobile control, in particular to a steering control method and system for a distributed drive automobile. BACKGROUND

[0002] The development of new energy vehicles is an important way to alleviate energy crisis and environmental pollution. Four-wheel independent steering / distributed drive electric vehicles have shown significant advantages in improving vehicle handling and stability due to their high degree of control flexibility. Four-wheel independent steering and direct yaw moment control are key functional modules for realizing vehicle stability in extreme steering conditions, and their performance directly affects the safety and dynamic response capability of the vehicle.

[0003] In the prior art, the control system of four-wheel independent steering usually adopts a hierarchical control architecture, which models the vehicle dynamics model and the steering motor model separately and performs cascade control. This multi-layer control method limits the dynamic response of the vehicle yaw moment to the steering motor servo system bandwidth, and the outer loop control performance is subject to the response speed of the inner loop. Especially in extreme steering conditions, system response delay significantly increases the risk of vehicle instability and even rollover. In addition, the complex hierarchical structure in the traditional control architecture further reduces the control efficiency, making it difficult to meet the real-time requirements in high dynamic scenarios.

[0004] At present, how to break through the limitations of the existing hierarchical control architecture and improve the dynamic response speed and stability of the four-wheel independent steering control system has become a technical problem to be solved. SUMMARY

[0005] The purpose of the present application is to solve the problem of traditional control architecture, in extreme steering conditions, system response delay significantly increases the risk of vehicle instability and even rollover, in addition, the complex hierarchical structure in the traditional control architecture further reduces the control efficiency, a steering control method and system for a distributed drive automobile are proposed.

[0006] A steering control method for a distributed drive automobile, the method comprising the following contents:

[0007] Step 1, constructing a steering motor model of a distributed drive automobile, constructing a two-degree-of-freedom model including yaw rate and center side slip angle, and constructing an angular velocity error model of each wheel motor;

[0008] Step 2, decoupling the two-degree-of-freedom model to establish a plurality of single-wheel dynamic models;

[0009] Step 3: Based on multiple single-wheel dynamic models, the preset ideal values ​​of the center of gravity sideslip angle and the ideal values ​​of the yaw rate of each wheel, establish a steering control error model. Based on the steering control error model, the steering motor model of the distributed drive vehicle, and the angular velocity error model of each wheel motor, establish an integrated steering error model.

[0010] Step 4: Decouple and reconstruct the state error variables in the integrated steering error model in sequence to obtain the controllable state equation;

[0011] Step 5: Design a fractional-order terminal sliding surface based on the controllable state equation and generate a virtual control law to compensate for unmatched disturbances;

[0012] Step 6: Establish a full-order terminal sliding surface based on the error between the virtual control law and the actual output, and design an actual control law to make the error converge on the full-order terminal sliding surface.

[0013] Step 7: Merge the virtual control law and the actual control law to generate four-wheel steering angle control signals and yaw moment control signals to perform steering control on the distributed drive vehicle.

[0014] Preferably, in step 3, the error model for steering control is expressed as:

[0015] ,

[0016] In the formula, Indicates the first The ideal value of the sideslip angle of the center of gravity of each wheel; Indicates the first Ideal value of the yaw rate of each wheel; Indicates the first The error of the center of gravity sideslip angle of each wheel; Indicates the first The derivative of the sideslip angle error of the center of gravity of each wheel; Indicates the first The yaw rate error of each wheel; Indicates the first The derivative of the yaw rate error of each wheel; Indicates the first The active steering angle of the hub motor of each wheel; Indicates the first Steering torque compensation input for each wheel;

[0017] The steering integration error model is expressed as:

[0018] ,

[0019] In the formula, denotes the derivative of the steering compensation angle of the i-th wheel; denotes the vehicle side stiffness.

[0020] Preferably, in step 4, the state error variables in the steering integration error model are sequentially decoupled and reconstructed to obtain a controllable state equation, and the specific process is as follows:

[0021] Step 41, decompose the steering integration error model into a non-matching subsystem containing load disturbance and a matching subsystem not containing load disturbance;

[0022] Step 42, decouple the state error vector in the non-matching subsystem to obtain a state error vector containing non-matching disturbance and a state error vector not containing non-matching disturbance, and simultaneously decouple the state error vector containing non-matching disturbance and the state error vector not containing non-matching disturbance by using a non-singular transformation matrix to obtain a decoupled state error vector of non-matching disturbance, and use the decoupled state error vector of non-matching disturbance to form a new non-matching subsystem;

[0023] Step 43, reconstruct the matching disturbance model and the new non-matching disturbance model to generate a controllable state equation.

[0024] Preferably, in step 5, a fractional order terminal sliding mode surface is designed according to the controllable state equation, and a virtual control law compensating for non-matching disturbance is generated, and the specific process is as follows:

[0025] Step 51, design a fractional order terminal sliding mode surface based on the controllable state equation;

[0026] Step 52, generate a virtual control law compensating for non-matching disturbance through the fractional order terminal sliding mode surface to constrain the convergence path and drive the error variables of the steering integration error model to converge.

[0027] A steering control system of a distributed drive vehicle, comprising:

[0028] A model construction unit is configured to construct a steering motor model of the distributed drive vehicle and a two-degree-of-freedom model including a yaw rate and a mass center side slip angle.

[0029] A decoupling unit is configured to decouple the two-degree-of-freedom model to establish a plurality of single wheel dynamic models.

[0030] An integration construction unit is configured to establish an error model of steering control according to the plurality of single wheel dynamic models, preset ideal values of the mass center side slip angle of each wheel, and preset ideal values of the yaw rate of each wheel, and establish a steering integration error model according to the error model of steering control, the steering motor model of the distributed drive vehicle, and the established angular velocity error model of each wheel motor.

[0031] a state space reconstruction unit, configured to sequentially decouple and reconstruct state error variables in the integrated steering error model to obtain a controllable state equation;

[0032] a control law design unit, configured to:

[0033] design a fractional order terminal sliding mode surface according to the controllable state equation, and generate a virtual control law for compensating non-matching disturbances;

[0034] establish a full order terminal sliding mode surface according to an error between the virtual control law and an actual output, and design an actual control law for converging the error on the full order terminal sliding mode surface;

[0035] a steering control unit, configured to fuse the virtual control law and the actual control law to generate a four-wheel steering angle control signal and a yaw moment control signal for steering control of the distributed drive vehicle.

[0036] The present application has the following advantages:

[0037] The steering control method for the distributed drive vehicle provided by the present application realizes unified modeling of the electromechanical coupling system by constructing a steering motor model and a two-degree-of-freedom vehicle dynamics model containing a yaw rate and a mass center side slip angle, and breaks through the limitation of mutual restriction of response speeds of inner and outer loops in the traditional hierarchical control architecture. By decoupling the four-wheel steering system into multiple single-wheel dynamic models, the complex signal transmission link in the traditional hierarchical control is eliminated, and the dynamic response speed of the control system is significantly improved, especially in high-frequency steering conditions, the control delay can still be effectively shortened.

[0038] The state space reconstruction method based on the integrated steering error model fuses the vehicle dynamics characteristics and the motor servo characteristics into a unified controllable state equation, solving the problem of low control efficiency caused by model separation in the traditional architecture. Through the design of the fractional order terminal sliding mode surface, combined with the virtual control law for compensating non-matching disturbances and the actual control law for eliminating matching disturbances, the limitation of the servo system bandwidth on the tracking accuracy of the yaw rate in the traditional method is effectively overcome, and stable tracking control of the vehicle yaw moment can still be maintained in the limit conditions such as tire lateral force saturation or road adhesion mutation. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 The figure is a method flowchart of the steering control method for the distributed drive vehicle according to the embodiment of the present application;

[0040] Figure 2 The figure is a schematic diagram of a four-wheel steering system dynamics model according to the embodiment of the present application;

[0041] Figure 3A single wheel force analysis schematic diagram shown in an embodiment of the present application;

[0042] Figure 4 A comparison schematic diagram of the dynamic response of the centroid side slip angle under sinusoidal working conditions shown in an embodiment of the present application;

[0043] Figure 5 A comparison schematic diagram of the dynamic response of the yaw rate under sinusoidal working conditions shown in an embodiment of the present application;

[0044] Figure 6 A comparison schematic diagram of the dynamic response of the centroid side slip angle under double line shifting working conditions shown in an embodiment of the present application;

[0045] Figure 7 A comparison schematic diagram of the dynamic response of the yaw rate under double line shifting working conditions shown in an embodiment of the present application;

[0046] Figure 8 A module structure schematic diagram of the steering control system of the distributed drive vehicle shown in an embodiment of the present application. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0048] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be further described below with reference to the drawings and specific embodiments, but is not limited to the present application.

[0049] Embodiment 1:

[0050] Reference Figure 1 , Figure 1 A method flow schematic diagram of a steering control method of a distributed drive vehicle. The steering control method of the distributed drive vehicle comprises:

[0051] Step 1, constructing a steering motor model of the distributed drive vehicle, constructing a two-degree-of-freedom model containing a yaw rate and a centroid side slip angle, and constructing an angular velocity error model of each wheel motor;

[0052] In applications, when constructing a steering motor model for a distributed drive vehicle, a mathematical model reflecting the dynamic behavior of the motor can be established by combining the electromechanical characteristics of the steering actuator. The steering motor uses a permanent magnet synchronous motor (PMSM), and its electromagnetic torque model describes the motor's output characteristics using parameters such as the stator current component on the q-axis, angular velocity error, and load torque. Simultaneously, the voltage equation characterizes the relationship between the motor's input voltage and current, inductance, and resistance, forming a complete electromechanical system model. This model accurately expresses the transmission relationship between the mechanical rotation angle of the steering motor shaft and the actual steering angle of the wheel, providing a foundation for subsequent integrated control.

[0053] When constructing a two-degree-of-freedom model incorporating yaw rate and sideslip angle, the vehicle can be simplified into a dynamic system containing only yaw and lateral motion. Yaw rate characterizes the vehicle's rotational state about its vertical axis, while sideslip angle reflects the degree to which the vehicle's lateral motion deviates from the expected path. Assuming a constant longitudinal velocity and neglecting the longitudinal degree of freedom, lateral force balance equations and yaw moment balance equations are established based on Newtonian mechanics principles. The lateral force equations, through the relationship between tire sidestiffness and wheel steering angle, superimpose the lateral forces of the four wheels onto the vehicle's center of gravity. The yaw moment equations, combining wheel position parameters and lateral force distribution, describe the vehicle's rotational dynamics about its center of gravity. By integrating structural parameters such as vehicle mass, moment of inertia, and wheelbase, the two-degree-of-freedom model effectively characterizes the vehicle's stability under low- to medium-speed conditions, providing a theoretical basis for error analysis and dynamic response design in steering control.

[0054] See Figure 2 , Figure 2 This is a schematic diagram of the dynamics model of a four-wheel steering system; Figure 2 The parameters of the vehicle are expressed as follows: The sideslip angle is the angle between the vehicle's center of gravity and its body. The yaw rate of the vehicle; and These are the distances from the front axle and rear axle to the vehicle's center of gravity, respectively. It is the distance between the left and right tires of a vehicle; Or 4, representing the left front wheel, right front wheel, left rear wheel, and right rear wheel in that order; For the first The longitudinal force of each tire; For the first Lateral force of each tire; For the first The steering angle of each wheel.

[0055] The dynamic modeling of a four-wheel independent steering system comprises two key components: the 4WIS DDEV (four-wheel independent steering distributed drive electric vehicle) vehicle dynamics model and the steering actuator model; the steering actuator model (steering motor model) describes the independent motion characteristics of each steering wheel through electromechanical system equations; such as Figure 2 As shown, the vehicle dynamics model, combined with tire characteristic equations, is used to characterize the interaction between the vehicle and the road surface.

[0056] Combination Figure 2 The variables shown can be used to derive the following dynamic equations:

[0057] Vehicle longitudinal dynamics equations:

[0058] (1)

[0059] Vehicle lateral dynamics equations:

[0060] (2)

[0061] In the formula:

[0062]

[0063] The equation for the yaw moment of a vehicle:

[0064] (3)

[0065] in, For the first The lateral force on each wheel; For the first The lateral force on each wheel; For the overall vehicle weight; Longitudinal velocity; For lateral velocity; The derivative of the longitudinal velocity; The derivative of the lateral velocity; M is the vehicle's moment of inertia. z Add yaw moment to the entire vehicle; Indicates the vehicle's yaw rate; The derivative of the vehicle's yaw rate. For the first Additional yaw control drive torque for each wheel; .

[0066] During vehicle operation, the wheel dynamics model under driving conditions is analyzed. For the force analysis of a single wheel, please refer to [link / reference needed]. Figure 3 ; Figure 3 This is a schematic diagram of the force analysis of a single wheel.

[0067] The motion of a single wheel can be composed of two parts: the vertical load equation and the wheel torque equation.

[0068] The equation of motion for a single wheel's rotation:

[0069] (4)

[0070] In the formula, This refers to the driving torque of the wheels; The moment of inertia of the wheel; It is the rotational angular acceleration; The radius of the wheel; The longitudinal frictional force is in the direction of forward movement; This is the braking torque.

[0071] The vertical load on the wheels has a significant impact on the longitudinal and lateral dynamic performance of the vehicle, and is also an important basis for subsequent torque distribution. In this application, the effects of air resistance and road slope on the vertical load of the tires can be neglected, and the vertical load equations for each tire can be expressed as follows:

[0072] (5)

[0073] In the formula, For the first Vertical load on the wheel; The height of the center of mass above the ground; This refers to the wheelbase; It is lateral acceleration; It is longitudinal acceleration; It accounts for 1 / 4 of the total vehicle weight; It is the acceleration due to gravity; The wheelbase is the distance between the left and right wheels of a vehicle.

[0074] Tire models are a core component of vehicle dynamics, accurately reflecting the characteristics of ground forces acting on the vehicle under different road adhesion conditions and dynamic states. The overall performance of a vehicle is determined by its structural parameters and mechanical properties, with the nonlinear characteristics of the tires playing a crucial role in the vehicle's stability under extreme conditions. The accuracy of the tire model directly determines the accuracy of the vehicle dynamics simulation; therefore, constructing a high-precision tire model is of great significance for vehicle dynamics research and control strategy development. The mathematical expression of the tire model is shown below:

[0075] (6)

[0076] In the formula, y represents the longitudinal or lateral force exerted by the ground on the tire; x characterizes the tire's slip ratio or sideslip angle; B1, C, D, E, S hand S v All of these are characteristic parameters.

[0077] Taking lateral force as an example, the slip angle of each tire can be expressed as follows:

[0078] (7)

[0079] In the formula, , , and The tire slip angles are represented sequentially for the left front wheel, right front wheel, left rear wheel, and right rear wheel.

[0080] Let y represent the lateral force of the tire, and x represent the tire slip angle. Then it will have the following forms of expression:

[0081] (8)

[0082] In the formula, the characteristic parameters are expressed as follows:

[0083]

[0084] in, , , , , , , , , , and All of these are vehicle parameters set manually.

[0085] The two-degree-of-freedom model is a simplified dynamic model based on the single-mass rigid body model. It assumes that the vehicle's velocity in the longitudinal direction remains constant and its direction of travel does not change, while neglecting the longitudinal degree of freedom of the vehicle. This model simplifies the vehicle into a two-degree-of-freedom system containing only yaw and lateral motion, using yaw rate and sideslip angle to reflect the stability of the vehicle during operation.

[0086] Using a 2-DOF (Two-Degree-of-Freedom) vehicle reference model with additional yaw moment, the vehicle dynamics model based on yaw stability can be summarized as follows:

[0087] (9)

[0088] In the formula, The derivative of the vehicle's sideslip angle; The derivative representing the vehicle's yaw rate; For the overall vehicle weight; v x l1 represents the longitudinal velocity; l2 and l3 represent the distances from the front axle and rear axle to the vehicle's center of gravity, respectively. For the first Lateral force of each tire; , , and These represent the lateral forces of the left front wheel, right front wheel, left rear wheel, and right rear wheel, respectively. , , and These represent the additional yaw control drive torques for the left front wheel, right front wheel, left rear wheel, and right rear wheel of the vehicle, respectively. The yaw rate of the vehicle; It is the distance between the left and right tires of a vehicle; Let represent the vehicle's moment of inertia.

[0089] Step 2: Decouple the two-degree-of-freedom model and establish multiple single-wheel dynamic models.

[0090] In applications, when decoupling the two-degree-of-freedom model, the dynamic relationship between the overall vehicle's yaw rate and sideslip angle can be decomposed into dynamic components under the independent action of each wheel. Assuming that the contributions of the lateral forces of each wheel to the vehicle's sideslip angle and yaw rate have a linear superposition characteristic, by introducing local dynamic response components when the lateral forces of each wheel act individually, the composite dynamic equation can be decomposed into four independent single-wheel dynamic models. Each single-wheel dynamic model includes the mapping relationship between the wheel's steering angle and lateral force, while also considering the leverage effect of wheel position parameters on the yaw moment, ensuring that the decomposed model can independently characterize the impact of each wheel on the overall vehicle stability.

[0091] By decomposing the wheel steering angle into a composite of the driver's input steering angle and the active compensation steering angle, the single-wheel dynamic model can simultaneously reflect the superposition of human control intentions and active control strategies. The lateral force dynamics of each wheel are expressed by coupling its steering angle change rate, vehicle longitudinal velocity, and center of gravity motion state, forming a subsystem with independent inputs and outputs.

[0092] The decoupled single-wheel dynamic model, through coordinate transformation and parameter reorganization, transforms the coupled state variables in the original two-degree-of-freedom model into local state variables that can be independently observed by each wheel. This process eliminates cross-interference between the dynamics of different wheels, allowing the steering control of each wheel to make independent decisions based on its local state. By transforming the vehicle stability control problem into a coordinated control problem of four distributed subsystems, a model foundation is provided for the subsequent design of a non-cascaded control architecture, thereby solving the problem of limited response speed in traditional hierarchical control.

[0093] Specifically, it can be set , The order is number 1 Lateral force of each wheel When acting alone, the sideslip angle and yaw rate generated at the vehicle's center of gravity satisfy the following conditions: , Then formula (9) can be rewritten as:

[0094] (10)

[0095] In the formula, Indicates the first The distance between the axle of each wheel and the center of gravity of the vehicle; Indicates the first The derivative of the sideslip angle of the center of mass of each wheel; Indicates the first The derivative of the yaw rate of each wheel; For the first The definition parameters for each wheel.

[0096] in,

[0097] The sideslip angles of the four wheels in formula (7) can be approximated as:

[0098] (11)

[0099] In the formula, , , , For the four corners of the vehicle's wheels; , , , These refer to the four corners of the vehicle's wheels; , , and The steering wheel angles given by the driver for the left front wheel, right front wheel, left rear wheel, and right rear wheel, respectively, are calculated after the steering ratio. , , , The active steering angles of the hub motors for the left front wheel, right front wheel, left rear wheel, and right rear wheel are, in order.

[0100] Meanwhile, when the vehicle's lateral acceleration is controlled within 0.4g, the tire's lateral stiffness is within the linear range, and the tire's slip angle is small, the lateral force of each wheel can be simplified as follows:

[0101] (12)

[0102] In the formula, For the first Tire lateral stiffness of each wheel.

[0103] Substituting equations (11) and (12) into equation (10), we can obtain the single-wheel dynamic model of 4WIS and DYC drive-by-wire:

[0104] (13)

[0105] In the formula, Indicates the first The derivative of the sideslip angle of the center of mass of each wheel; Indicates the first The derivative of the yaw rate of each wheel; Indicates the first The distance between the axle of each wheel and the center of gravity of the vehicle; The vehicle's moment of inertia; and The numbers represent the order of the numbers. The sideslip angle and yaw rate generated at the center of gravity of the vehicle when the lateral force of each wheel acts alone; For the overall vehicle weight; Longitudinal velocity; For the first Tire lateral stiffness of each wheel; Indicates the first The steering angle of each wheel; Indicates the first Additional yaw control drive torque for each wheel.

[0106] Step 3: Based on multiple single-wheel dynamic models, the preset ideal values ​​of the center of gravity sideslip angle and the ideal values ​​of the yaw rate of each wheel, establish a steering control error model. Based on the steering control error model, the steering motor model of the distributed drive vehicle, and the angular velocity error model of each wheel motor, establish an integrated steering error model.

[0107] In applications, when constructing an integrated steering error model, the dynamic characteristics of the steering motor can be coupled with the error variables of the dynamic models of each individual wheel for analysis. By defining the steering angle tracking error of each wheel as the deviation between the desired steering angle and the actual steering angle, and combining the transmission relationship between the mechanical rotation angle of the motor shaft and the steering angle of the wheel, the electromechanical error of the steering actuator is dynamically introduced into the vehicle stability control framework.

[0108] The construction of the integrated steering error model requires integrating the decoupled dynamic equations of each individual wheel with the motor control equations to form a composite state equation that includes vehicle stability error and steering execution error. By introducing derivative terms and coupling disturbance terms of the error variables, the model can simultaneously describe the dynamic relationship between the vehicle's center of gravity motion and the steering actions of the four wheels.

[0109] Specifically, based on the dynamic model of formula (13), the first... The tracking error of each subsystem is Establish an error model for steering control:

[0110] (14)

[0111] In the formula, Indicates the first The ideal value of the sideslip angle of the center of gravity of each wheel; Indicates the first Ideal value of the yaw rate of each wheel; Indicates the first The error of the center of gravity sideslip angle of each wheel; Indicates the first The derivative of the sideslip angle error of the center of gravity of each wheel; Indicates the first The yaw rate error of each wheel; Indicates the first The derivative of the yaw rate error of each wheel; Indicates the first The active steering angle of the hub motor of each wheel; Indicates the first Steering torque compensation input for each wheel.

[0112] In applications, a permanent magnet synchronous motor (PMSM) can be used as the steering actuator; the electromagnetic torque error model is as follows:

[0113] (15)

[0114] In the formula, Representing the The stator current of each wheel is Current components of the shaft, Representing the The angular velocity error of the motor of each wheel Indicates the first Speed ​​error of the steering motor of each wheel; Represents load torque. Represents the number of pole pairs of the motor. Represents permanent magnet flux linkage. Represents the moment of inertia.

[0115] Furthermore, surface-mounted permanent magnet synchronous motors (SPMSMs) have the advantages of relatively simple control and low cost, and have great application potential in low-speed and high-power-density scenarios. The voltage equation for an SPMSM is:

[0116] (16)

[0117] In the formula, Indicates the first Each wheel steering motor The derivative of the shaft current signal; For the first Voltage signal for each wheel steering motor For the stator resistance of the motor, For motor Shaft stator inductance.

[0118] Combining the steering control error model (14) and the steering motor model (16), and considering the mechanical transmission relationship of the steering system, the integrated steering error model can be constructed as follows:

[0119] (17)

[0120] In the formula, Indicates the first The derivative of the steering compensation angle of each wheel; This indicates the vehicle's lateral stiffness.

[0121] Step 4: Decouple and reconstruct the state error variables in the integrated steering error model in sequence to obtain the controllable state equation.

[0122] In applications, during the state-space reconstruction of the integrated error model, the coupled composite error equations can be decomposed into matched and unmatched subsystems. By analyzing the dynamic correlations of the state variables in the model, matched disturbance components directly affected by the control input and unmatched disturbance components that need to be transmitted through intermediate states are identified. To address the strong coupling characteristics in the unmatched subsystems, non-singular transformation matrices can be used to transform the original state variables, eliminating cross-coupling terms between different state components and converting the higher-order coupled equations into lower-order decoupled independent dynamic equations.

[0123] During the reconfiguration process, the mechanical constraints of the steering actuator and the dynamic characteristics of the motor can be combined to recombine state variables such as motor angular velocity error, steering angle tracking error, and centroid sideslip angle error into an observable and controllable state vector. By introducing a pseudo-inverse matrix to normalize the transformed subsystem, it is ensured that the reconfigured state equations satisfy the controllability condition, that is, any state of the system can be adjusted to the target value through finite control inputs.

[0124] In this embodiment, the state error variables within the integrated steering error model are sequentially decoupled and reconstructed to obtain a controllable state equation, including:

[0125] Step 41: Decompose the steering integration error model into a non-matched subsystem with load disturbances and a matched subsystem without load disturbances;

[0126] Step 42: Decouple the state error vector in the unmatched subsystem to obtain the state error vector containing the unmatched disturbance and the state error vector without the unmatched disturbance. Use a non-singular transformation matrix to simultaneously decouple the state error vector containing the unmatched disturbance and the state error vector without the unmatched disturbance to obtain the decoupled state error vector with the unmatched disturbance. Use the decoupled state error vector with the unmatched disturbance to form a new unmatched subsystem.

[0127] Step 43: Reconstruct the matched perturbation model and the new unmatched perturbation model to generate controllable state equations.

[0128] Specifically, the aforementioned integrated steering error model has a high dimension, complex coupling of state variables, and cannot be directly controlled. Therefore, it is necessary to decompose the system (17) to obtain a non-matched subsystem containing load disturbances and a subsystem without disturbances. Then, a non-singular transformation is performed on the strongly coupled non-matched subsystem to remove coupling, so that the error variables can converge one by one according to the control input. The specific process is as follows.

[0129] First, the steering integration error model (17) is rewritten as a non-matching subsystem (18) and a matching subsystem (19) after line breaks:

[0130] (18)

[0131] (19)

[0132] In the formula:

[0133] (20)

[0134] In the formula, The derivative of the state error vector of the unmatched subsystem; and The first state gain matrix and the second state gain matrix, which contain state coupling, are represented sequentially. This represents the state error vector of the non-matching subsystem; This represents the state error vector of the matching subsystem; This represents the first non-matching perturbation vector; The derivative of the state error vector of the matching subsystem; and The third and fourth state gain matrices, which contain state coupling, are represented sequentially. Represents the input gain matrix; Indicates system status input; This represents the first unmatched perturbation function; This represents a human-defined upper bound parameter for the perturbation.

[0135] in, ≥0, ≥0 is a known constant. , , ;

[0136] ;

[0137] Secondly, in order to achieve decoupled control of the steering integration model, the non-matching subsystem (18) can be decoupled. Decomposed into and The decomposed system:

[0138] (twenty one)

[0139] In the formula:

[0140]

[0141] in, This represents the state error vector of the non-matched subsystem containing non-matched disturbances; This indicates that the non-matched subsystem does not contain a state error vector containing non-matched disturbances; The derivative of the state error vector of the unmatched subsystem that does not contain unmatched disturbances; The derivative of the state error vector of the unmatched subsystem containing the unmatched disturbance; express and The coupling gain matrix between them; express and The coupling gain matrix between them; express and The coupling gain matrix between them; express and The coupling gain matrix between them; express Matching state error vector The coupling gain matrix between them; This represents the second unmatched perturbation function.

[0142] The uncertainty vector function f2 satisfies:

[0143] (twenty two)

[0144] In the formula, This represents a (2-α)th order fractional calculus; Indicates the upper bound gain of the perturbation. ;D u This represents the upper bound of the defined non-matching perturbation.

[0145] Furthermore, in order to eliminate state coupling in mismatched subsystems, a non-singular transformation matrix is ​​defined. By changing the rules: This achieves decoupling. The transformed unmatched subsystem is:

[0146] (twenty three)

[0147] In the formula, It represents the derivative of the state error vector of the unmatched subsystem containing the unmatched perturbation after non-singular transformation; It represents the derivative of the state error vector of the unmatched subsystem after non-singular transformation, which does not contain unmatched perturbations; This indicates that the unmatched subsystem after non-singular transformation does not contain the state error vector of the unmatched perturbation; This represents the state error vector of the unmatched subsystem after non-singular transformation, which contains unmatched perturbations. express and The coupling gain matrix between them; express and The coupling gain matrix between them; express and The coupling gain matrix between them.

[0148] State transition matrix ,in , pseudo-inverse , ,

[0149]

[0150] In the formula, Represents the identity matrix; This represents manually defined elements in the state transition matrix; Representation matrix The right pseudo-inverse matrix; Representation matrix Transpose of; Representation matrix The reverse.

[0151] Furthermore, by substituting the decoupled unmatched subsystems into the original systems (18) and (19), the controllable state equations of the decoupled system can be expressed in the following form:

[0152] (twenty four)

[0153] In the formula, express The derivative; express The derivative; express The derivative; This indicates that the unmatched subsystem after non-singular transformation does not contain the state error vector of the unmatched perturbation; This represents the state error vector of the unmatched subsystem after non-singular transformation, which contains unmatched perturbations. This represents the state error vector of the non-matching subsystem; This represents the state error vector of the matching subsystem; express and The coupling gain matrix between them; This represents the state error vector of the non-matched subsystem containing non-matched disturbances; This indicates that the non-matched subsystem does not contain a state error vector containing non-matched disturbances; for The derivative; and The first state gain matrix and the second state gain matrix, which contain state coupling, are represented sequentially. and The third and fourth state gain matrices, which contain state coupling, are represented sequentially. Represents the input gain matrix; Indicates system status input; express The coupling gain matrix between x2 and x2; express The derivative of .

[0154] Because the unmatched subsystem has been decoupled, the error variable can be realized. Convergence, error variable It can automatically converge to 0, and the control law in subsequent steps can be designed as a second-order system by considering the second and third rows of system (24), which greatly reduces the difficulty of control.

[0155] Step 5: Design a fractional-order terminal sliding surface based on the controllable state equation and generate a virtual control law to compensate for unmatched disturbances.

[0156] In applications, when designing fractional-order terminal sliding surfaces, the mathematical form of the sliding surface can be defined based on the decoupling error variables in the controllable state equations, enabling it to guide the system state to converge to the equilibrium point along a preset path. By using linear or nonlinear combinations of yaw rate error, centroid sideslip angle error, and motor angular velocity error as sliding surface functions, a terminal sliding surface containing fractional-order differential terms can be constructed to accelerate the convergence speed of error variables and suppress steady-state chattering.

[0157] Because the disturbance and the control signal are of different orders, they exhibit a mismatched state, known as a mismatched disturbance, which requires compensation using a virtual control law. Mismatched disturbances mainly originate from road surface friction, mechanical transmission structure friction, etc., and are uniformly manifested as load disturbances in the steering motor. The role of the virtual control law is to drive the state error of the mismatched subsystem to the sliding surface and maintain the ideal dynamic characteristics of the system on the sliding surface. During the design process, the disturbance estimate can be embedded into the control law, and the asymptotic stability of the closed-loop system can be verified using Lyapunov stability theory. By constructing a positive definite Lyapunov function containing the sliding surface variables and proving that its derivative satisfies the negative definiteness condition under the action of the control law, it is ensured that the error variable can still converge to zero when a disturbance exists.

[0158] By limiting the impact of mismatched disturbances to the dynamic range of the sliding surface, the virtual control law effectively isolates the negative impact of uncertainties such as road surface disturbances and motor parameter perturbations on vehicle stability, thereby improving the robustness of the integrated steering control system.

[0159] In step 5, a fractional-order terminal sliding surface is designed based on the controllable state equation, and a virtual control law to compensate for unmatched disturbances is generated, including:

[0160] Step 51: Design a fractional-order terminal sliding surface based on the controllable state equation;

[0161] Step 52: By constraining the convergence path through the fractional-order terminal sliding surface, a virtual control law to compensate for the mismatched disturbance is generated to drive the convergence of the error variables of the steering integrated error model.

[0162] In applications, the purpose of designing the control law is to converge the error variable to 0. The mechanism of the sliding mode algorithm is to allow the state point to slide from the sliding surface to zero. Therefore, designing the sliding surface based on the error variable can ultimately ensure that the error continuously converges to 0, achieving disturbance rejection control. For the decoupled system described above, in order to make the error variable z1 of the unmatched subsystem converge to 0 and resist unmatched load disturbances, the fractional-order terminal sliding surface of the virtual control law (virtual control sliding surface) can be designed as follows:

[0163] (25)

[0164] In the formula, The sliding surface representing the virtual control law; Table non-matching subsystem Power of; It is a constant matrix. It is a constant, and 0 > <1, For fractional calculus operators, For (2- ( )th order differential.

[0165] Define virtual control law Tracking error vector .error Converging to zero, It will converge to 0, and then It will also reach 0, and Depend on and Composition, therefore It will converge to zero. Design the following fractional-order control law:

[0166] (26)

[0167] (27)

[0168] (28)

[0169] In the formula, Represents a virtual control law; Represents a virtual control law; Represents the equivalent control part of a virtual control law; This represents the switching control section of the virtual control law; Representation matrix The right pseudo-inverse matrix; Represents a constant matrix of Power; Represents the error vector of Power of; This indicates the controller gain.

[0170] Furthermore, the stability of the virtual control law is proven as follows:

[0171] First of all Substituting into formula (24), we get:

[0172] (29)

[0173] In the formula, This represents the error between the error state vector of the error matching subsystem and the virtual control law; Process variables are used to simplify the expression.

[0174] Combine (29) with Substituting the fractional-order terminal sliding surface (25), we have:

[0175] (30)

[0176] In the formula, express The derivative of .

[0177] Get Lyapunov function Substituting into equation (30) and simplifying, we get:

[0178] (31)

[0179] Substituting the switching control law (26) into the above equation, when have:

[0180] (32)

[0181] in, This represents the Lyapunov function designed to prove the stability of the virtual control law; Represents the sliding surface Transpose of; Represents the sliding surface The derivative; The derivative of the Lyapunov function is represented; This indicates that the parameters were set manually. This represents the Lyapunov function raised to the power of 1 / 2.

[0182] Under the control of the fractional integral sliding mode control law, the subsystem (24) will be generated from any initial state. In a limited time Inner reach fractional-order terminal sliding surface ,in And maintain ideal sliding mode on the sliding surface. .then, It will converge to 0 in a finite time, thus making all state variables of system (24)... The output variables in the unmatched uncertain system, namely the decoupled centroid sideslip angle and yaw rate error, will eventually converge to zero, thus achieving stability control of the 4WIS DDEV.

[0183] Step 6: Establish a full-order terminal sliding surface based on the error between the virtual control law and the actual output, and design an actual control law to make the error converge on the full-order terminal sliding surface.

[0184] In applications, when establishing a full-order terminal sliding surface, an error vector can be defined based on the tracking error between the reference trajectory output by the virtual control law and the actual output. By using the composite deviation of yaw rate error, centroid sideslip angle error, and motor angular velocity error as state variables, a full-order terminal sliding surface is constructed. Its mathematical form is defined through a nonlinear combination of fractional differential terms and the error vector, ensuring that the system state converges to the equilibrium point within a finite time.

[0185] Matching disturbances mainly originate from perturbations in the motor's internal parameters and the d-axis current component. Since the disturbance and the control signal have the same order, the actual control law can directly apply to this disturbance. The actual control law drives the error vector to converge rapidly along the sliding surface, while compensating for uncertainties in matching disturbances such as changes in motor parameters and sensor noise. By using the output of the virtual control law as a reference input, the actual control law adjusts the motor current or voltage signal in real time, gradually eliminating the tracking errors of the wheel steering angle and yaw moment. During the design process, the global stability of the closed-loop system can be verified using Lyapunov stability theory. By constructing an energy function that includes sliding surface variables, it can be proven that its derivative satisfies the negative definite condition under control action, ensuring that the system maintains robust stability even when disturbances exist.

[0186] Ultimately, the actual control law and the virtual control law work together to form a layered disturbance rejection architecture, which effectively isolates the coupling effects of internal and external disturbances such as road excitation and load changes on vehicle stability, thus ensuring the rapid response and steady-state accuracy of the four-wheel independent steering system.

[0187] In applications, to eliminate the error between the virtual control vector and the actual output... It is possible to design full-order terminal sliding surfaces (actual control sliding surfaces) for actual control laws, thereby realizing full-order sliding modes with a relative order of 0 for the system. The full-order terminal sliding surface is based on error... Design, full-order terminal sliding surface for:

[0188] (33)

[0189] In the formula, Indicates error The derivative; Indicates error The error vector is q / p raised to the power of p. ,matrix Let be a constant matrix, where q and p are both positive odd numbers, and satisfy 0.

[0190] Furthermore, an actual control law is designed to ensure that the error e2 can continuously converge to 0 on the terminal sliding surface of all orders. The actual control law u is:

[0191] (34)

[0192] (35)

[0193] (36)

[0194] In the formula, Represents the equivalent control component of the actual control law; This represents the switching control section of the actual control law; express The derivative; This represents the controller gain parameter.

[0195] To verify the effectiveness of the actual control law, its stability is proven as follows:

[0196] First, the error vector Substituting equation (24) into the full-order terminal sliding surface (33), we have:

[0197] (37)

[0198] Substituting the actual control law (34) into the above equation, we have:

[0199] (38)

[0200] Let the Lyapunov function V2 = 0.5s 22 T s 22 Combining this with the above equation, its differential is:

[0201] (39)

[0202] Substituting the switching control law (36) into the above equation, and considering the boundary conditions for matching uncertainty, we have:

[0203] (40) ​

[0204] When V2≠0, and At that time, there were:

[0205] (41)

[0206] Where V2 represents the Lyapunov function used to prove the actual control law; Represents the sliding surface Transpose of; The derivative of the Lyapunov function is represented; Indicates a non-matching perturbation The derivative; Indicates the upper bound of the disturbance; This indicates that the parameters were set manually. This represents the Lyapunov function raised to the power of 1 / 2.

[0207] The error state trajectory will start from any initial state In a limited time Inner reach of full-order terminal sliding surface ,in And maintain the sliding mode motion on the sliding surface, so that the error vector It converges to zero within a finite time; thus, through precise control of the steering actuator motor, effective regulation of vehicle stability is directly achieved.

[0208] Step 7: Merge the virtual control law and the actual control law to generate four-wheel steering angle control signals and yaw moment control signals to perform steering control on the distributed drive vehicle.

[0209] In applications, when generating four-wheel steering angle control signals and yaw moment control signals, the outputs of the virtual control law and the actual control law can be dynamically fused. The reference trajectory provided by the virtual control law is combined with the motor input adjustment of the actual control law through error state feedback to form the control command for the steering angle of each wheel. The steering angle command for each wheel is calculated in real time based on the error state of its single-wheel dynamic model, and the steering motor drives the rack and pinion transmission mechanism through voltage or current signals to achieve precise tracking of the wheel steering angle. At the same time, the yaw moment control signal is generated based on the combined state of the vehicle's yaw rate error and center of gravity sideslip angle error; in addition, the longitudinal force distribution of each wheel is dynamically adjusted through the torque distribution strategy of the drive motor, and the lever effect of wheel position parameters on yaw moment is used to form an additional yaw moment to compensate for vehicle stability deviations.

[0210] The coordination between the steering angle control signal and the yaw moment control signal can be adjusted in real time based on the vehicle's dynamic state. During vehicle steering, by monitoring the real-time changes in the center of gravity sideslip angle and yaw rate, the active compensation components of the steering angle of each wheel and the torque output ratio of the drive motor can be dynamically corrected.

[0211] The generated steering angle control signal and yaw moment control signal are then distributed to the execution units of each steering motor and drive motor via the distributed controller. The steering angle control signal achieves precise execution of wheel steering actions through motor angular position adjustment, while the yaw moment control signal achieves dynamic distribution of longitudinal force through drive motor torque adjustment. These two signals work synergistically during vehicle movement to offset the effects of road surface disturbances, load disturbances, and model uncertainties on vehicle stability in real time. This ensures that the yaw rate and sideslip angle of the vehicle converge rapidly to the desired trajectory under extreme steering conditions, thereby improving the steering stability and dynamic response performance of the distributed drive vehicle.

[0212] To verify the effectiveness of the steering control method provided by this invention, the experimental conditions selected were the front wheel sinusoidal steering angle input and dual lane change test scenarios commonly used in four-wheel steering control research. To evaluate the performance of the steering control method provided by this invention, a comparative experiment was designed, including no control and traditional proportional-integral-derivative (PID) control, as a benchmark reference.

[0213] See Figure 4 and Figure 5 , Figure 4 and Figure 5 The dynamic responses of vehicle yaw rate and sideslip angle under three upper-level control strategies (no control, proportional-integral-derivative control, and the fractional-order terminal sliding mode control of this invention) under sinusoidal steering input conditions are presented to compare the performance differences of different steering control methods in terms of stability control. Under the no-control strategy, the peak sideslip angle reaches 2.42 degrees, and the peak yaw rate reaches 22.33 degrees / s. With proportional-integral-derivative control, the sideslip angle decreases significantly, with the peak value dropping to 1.85 degrees, and the peak yaw rate to 21.58 degrees / s. With the fractional-order terminal sliding mode control of this invention, the peak sideslip angle further decreases to 1.45 degrees, a 40.08% reduction compared to no control, and the peak yaw rate further decreases to 20.30 degrees / s, a 9.09% reduction compared to no control. Therefore, the fractional-order terminal sliding mode control method provided by this invention can effectively improve vehicle stability under sinusoidal conditions.

[0214] See Figure 6 and Figure 7 , Figure 6 and Figure 7The dynamic responses of the vehicle's sideslip angle and yaw rate under three upper-level control strategies in the double lane change condition are presented. As can be seen from the figures, the peak sideslip angle of the vehicle without control reaches as high as 1.52 degrees, and the peak yaw rate reaches as high as 21.41 degrees / s. After adopting the proportional-integral-derivative (PID) control strategy, the peak sideslip angle is significantly reduced to 1.16 degrees, and the peak yaw rate decreases to 20.3 degrees / s. After adopting the fractional-order terminal sliding mode control of this invention, the peak sideslip angle further decreases to 0.97 degrees, a reduction of 23.68% compared to without control, and the peak yaw rate further decreases to 19.65 degrees / s, a reduction of 8.22% compared to without control. It is evident that the control algorithm provided by this invention has good tracking performance for both yaw rate and sideslip angle, effectively improving the stability of the vehicle during driving.

[0215] Example 2:

[0216] Corresponding to Example 1, this example also provides a steering control system for a distributed drive vehicle.

[0217] Please see Figure 8 , Figure 8 This is a schematic diagram of the modular structure of a steering control system for a distributed drive vehicle. The steering control system of a distributed drive vehicle includes:

[0218] The model building unit is used to build a steering motor model for a distributed drive vehicle, a two-degree-of-freedom model including yaw rate and sideslip angle, and an angular velocity error model for each wheel motor.

[0219] The decoupling unit is used to decouple the two-degree-of-freedom model and establish multiple single-wheel dynamic models.

[0220] An integrated construction unit is used to establish a steering control error model based on multiple single-wheel dynamic models, preset ideal values ​​of the center of gravity sideslip angle and the yaw rate of each wheel, and to establish an integrated steering error model based on the steering control error model, the steering motor model of the distributed drive vehicle and the angular velocity error model of each wheel motor.

[0221] The state space reconstruction unit is used to decouple and reconstruct the state error variables in the integrated steering error model in sequence to obtain the controllable state equation.

[0222] Control law design unit, used for:

[0223] Based on the controllable state equation, a fractional-order terminal sliding surface is designed, and a virtual control law to compensate for unmatched disturbances is generated.

[0224] Based on the error between the virtual control law and the actual output, a full-order terminal sliding surface is established, and an actual control law is designed to make the error converge on the full-order terminal sliding surface.

[0225] The steering control unit is used to fuse the virtual control law and the actual control law to generate four-wheel steering angle control signals and yaw moment control signals to perform steering control on the distributed drive vehicle.

[0226] In Example 2, the state error variables within the integrated steering error model are sequentially decoupled and reconstructed to obtain the controllable state equations. The specific process is as follows:

[0227] The steering integration error model is decomposed into a non-matched subsystem with load disturbances and a matched subsystem without load disturbances;

[0228] The state error vector in the unmatched subsystem is decoupled to obtain a state error vector containing unmatched disturbances and a state error vector without unmatched disturbances. A non-singular transformation matrix is ​​used to simultaneously decouple the state error vector containing unmatched disturbances and the state error vector without unmatched disturbances to obtain a decoupled state error vector of unmatched disturbances. A new unmatched subsystem is formed using the decoupled state error vector of unmatched disturbances.

[0229] The matched perturbation model and the new unmatched perturbation model are reconstructed to generate controllable state equations.

[0230] In Example 2, a fractional-order terminal sliding surface is designed based on the controllable state equation, and a virtual control law to compensate for unmatched disturbances is generated. The specific process is as follows:

[0231] Design of fractional-order terminal sliding surface based on controllable state equations;

[0232] By constraining the convergence path through a fractional-order terminal sliding surface, a virtual control law is generated to compensate for mismatched disturbances, thereby driving the convergence of error variables in the steering integrated error model.

[0233] Regarding the system in the above embodiments, the specific manner in which each unit module performs operations has been described in detail in the embodiments related to the method, and will not be elaborated further here.

[0234] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A steering control method for a distributed drive vehicle, characterized in that, The method includes the following: Step 1: Construct a steering motor model for a distributed drive vehicle, construct a two-degree-of-freedom model including yaw rate and sideslip angle, and construct an angular velocity error model for each wheel motor. Step 2: Decouple the two-degree-of-freedom model and establish multiple single-wheel dynamic models; Step 3: Based on multiple single-wheel dynamic models, the preset ideal values ​​of the center of gravity sideslip angle and the ideal values ​​of the yaw rate of each wheel, establish a steering control error model. Based on the steering control error model, the steering motor model of the distributed drive vehicle, and the angular velocity error model of each wheel motor, establish an integrated steering error model. Step 4: Decouple and reconstruct the state error variables in the integrated steering error model in sequence to obtain the controllable state equation; Step 5: Design a fractional-order terminal sliding surface based on the controllable state equation and generate a virtual control law to compensate for unmatched disturbances; Step 6: Establish a full-order terminal sliding surface based on the error between the virtual control law and the actual output, and design an actual control law to make the error converge on the full-order terminal sliding surface. Step 7: Merge the virtual control law and the actual control law to generate four-wheel steering angle control signals and yaw moment control signals to perform steering control on the distributed drive vehicle.

2. The steering control method for a distributed drive vehicle according to claim 1, characterized in that, The two-degree-of-freedom model is expressed as: ， In the formula, The derivative of the vehicle's sideslip angle; The derivative representing the vehicle's yaw rate; For the overall vehicle weight; Longitudinal velocity; and These are the distances from the front axle and rear axle to the vehicle's center of gravity, respectively. For the first Lateral force of each tire; ; , , and These represent the lateral forces of the left front wheel, right front wheel, left rear wheel, and right rear wheel, respectively. , , and These represent the additional yaw control drive torques for the left front wheel, right front wheel, left rear wheel, and right rear wheel of the vehicle, respectively. The yaw rate of the vehicle; It is the distance between the left and right tires of a vehicle; The vehicle's moment of inertia; The steering motor model of a distributed drive vehicle is represented as follows: ， In the formula, Indicates the first Each wheel steering motor The derivative of the shaft current signal; For the first Voltage signal for each wheel steering motor For the stator resistance of the motor, For motor Shaft stator inductance, Representing the The stator current of each wheel is Current components of the shaft, Representing the The angular velocity error of each wheel motor Represents permanent magnet flux linkage; If each wheel motor uses a permanent magnet synchronous motor, then the angular velocity error model of each wheel motor is expressed as follows: ， In the formula, Representing the The derivative of the angular velocity error of each wheel motor Represents load torque. Represents the number of pole pairs of the motor. Represents the moment of inertia.

3. The steering control method for a distributed drive vehicle according to claim 2, characterized in that, The dynamic model of a single wheel is represented as follows: Single-wheel dynamic model: ， In the formula, and The numbers represent the order of the numbers. The sideslip angle and yaw rate generated at the center of gravity of the vehicle when the lateral force of each wheel acts alone; Indicates the first The steering angle of each wheel; Indicates the first Additional yaw control drive torque for each wheel; Indicates the first The distance between the axle of each wheel and the vehicle's center of gravity; I z The vehicle's moment of inertia; Indicates the first The derivative of the sideslip angle of the center of mass of each wheel; Indicates the first The derivative of the yaw rate of each wheel; m is the mass of the entire vehicle; v x k represents the longitudinal velocity. i Let be the tire lateral stiffness of the i-th wheel; ; .

4. The steering control method for a distributed drive vehicle according to claim 3, characterized in that, In step 3, the error model for steering control is expressed as: ， In the formula, Indicates the first The ideal value of the sideslip angle of the center of gravity of each wheel; Indicates the first Ideal value of the yaw rate of each wheel; Indicates the first The error of the center of gravity sideslip angle of each wheel; Indicates the first The derivative of the sideslip angle error of the center of gravity of each wheel; Indicates the first The yaw rate error of each wheel; Indicates the first The derivative of the yaw rate error of each wheel; Indicates the first The active steering angle of the hub motor of each wheel; Indicates the first Steering torque compensation input for each wheel; The steering integration error model is expressed as: ， In the formula, The derivative of the steering compensation angle of the i-th wheel; This indicates the vehicle's lateral stiffness.

5. A steering control method for a distributed drive vehicle according to claim 1 or 4, characterized in that, In step 4, the state error variables within the integrated steering error model are sequentially decoupled and reconstructed to obtain the controllable state equations. The specific process is as follows: Step 41: Decompose the steering integration error model into a non-matched subsystem with load disturbances and a matched subsystem without load disturbances; Step 42: Decouple the state error vector in the unmatched subsystem to obtain the state error vector containing the unmatched disturbance and the state error vector without the unmatched disturbance. Use a non-singular transformation matrix to simultaneously decouple the state error vector containing the unmatched disturbance and the state error vector without the unmatched disturbance to obtain the decoupled state error vector with the unmatched disturbance. Use the decoupled state error vector with the unmatched disturbance to form a new unmatched subsystem. Step 43: Reconstruct the matched perturbation model and the new unmatched perturbation model to generate controllable state equations.

6. The steering control method for a distributed drive vehicle according to claim 5, characterized in that, In step 41, the non-matching subsystem is represented as: ， The matching subsystem is represented as: ， In the formula, , The derivative of the state error vector of the unmatched subsystem; and The first state gain matrix and the second state gain matrix, which contain state coupling, are represented sequentially. This represents the state error vector of the non-matching subsystem; This represents the state error vector of the matching subsystem; This represents the first non-matching perturbation vector; The derivative of the state error vector of the matching subsystem; and The third and fourth state gain matrices, which contain state coupling, are represented sequentially. Represents the input gain matrix; Indicates system status input; This represents the first unmatched perturbation function; This represents a manually defined upper bound parameter for the perturbation. = [ , , ] T , =[ , ] T , =[ , ] T ; ； In step 42, the state error vector containing the mismatched disturbance and the state error vector without the mismatched disturbance are represented as follows: ， In the formula, , , This represents the state error vector of the non-matched subsystem containing non-matched disturbances; This indicates that the non-matched subsystem does not contain a state error vector containing non-matched disturbances; The derivative of the state error vector of the unmatched subsystem that does not contain unmatched disturbances; The derivative of the state error vector of the unmatched subsystem containing the unmatched disturbance; express and The coupling gain matrix between them; express and The coupling gain matrix between them; express and The coupling gain matrix between them; express and The coupling gain matrix between them; express Matching state error vector The coupling gain matrix between them; This represents the second unmatched perturbation function. , express( Fractional calculus of order 1; Indicates the upper bound gain of the perturbation. ; This represents the upper bound of the defined non-matching perturbation; The state error vector of the decoupled unmatched disturbance is represented as: ， In the formula, It represents the derivative of the state error vector of the unmatched subsystem containing the unmatched perturbation after non-singular transformation; It represents the derivative of the state error vector of the unmatched subsystem after non-singular transformation, which does not contain unmatched perturbations; This indicates that the unmatched subsystem after non-singular transformation does not contain the state error vector of the unmatched perturbation; This represents the state error vector of the unmatched subsystem after non-singular transformation, which contains unmatched perturbations. express and The coupling gain matrix between them; express and The coupling gain matrix between them; , , The state transformation matrix is... , pseudo-inverse , Represents the identity matrix; This represents manually defined elements in the state transition matrix; Representation matrix The right pseudo-inverse matrix; Representation matrix Transpose of; Representation matrix The reverse; The controllable state equation is expressed as: ， In the formula, express The derivative; express The derivative; express The derivative; This represents the state error vector of the non-matching subsystem; This represents the state error vector of the matching subsystem; express and The coupling gain matrix between them; This represents the state error vector of the non-matched subsystem containing non-matched disturbances; This indicates that the non-matched subsystem does not contain a state error vector containing non-matched disturbances.

7. A steering control method for a distributed drive vehicle according to claim 1 or 6, characterized in that, In step 5, a fractional-order terminal sliding surface is designed based on the controllable state equation, and a virtual control law to compensate for unmatched disturbances is generated. The specific process is as follows: Step 51: Design a fractional-order terminal sliding surface based on the controllable state equation; Step 52: By constraining the convergence path through the fractional-order terminal sliding surface, a virtual control law to compensate for the mismatched disturbance is generated to drive the convergence of the error variables of the steering integrated error model.

8. A steering control method for a distributed drive vehicle according to claim 7, characterized in that, In step 5, specifically step 51, the fractional-order terminal sliding surface is represented as: ， In the formula, The sliding surface representing the virtual control law; Table non-matching subsystem Power of; It is a constant matrix. It is a constant, and 0 < <1, For fractional calculus operators, For (2- )-th order differential; In step 52, the virtual control law is expressed as: ， In the formula, , , Represents a virtual control law; Represents the equivalent control part of a virtual control law; This represents the switching control section of the virtual control law; Representation matrix The right pseudo-inverse matrix; Represents a constant matrix of Power; Represents the error vector of Power of; This indicates the controller gain.

9. A steering control method for a distributed drive vehicle according to claim 1 or 8, characterized in that, In step 6, the full-order terminal sliding surface is represented as: ， In the formula, Indicates the full-order terminal sliding surface. Indicates error The derivative; Indicates error The error vector is q / p raised to the power of p. ,matrix It is a constant matrix. and All are positive odd numbers, and satisfy 0 < <1; The actual control law is expressed as: ， In the formula, , , Represents the equivalent control component of the actual control law; This represents the switching control section of the actual control law; express The derivative; This represents the controller gain parameter.

10. A steering control system for a distributed drive vehicle, characterized in that, include: The model building unit is used to build a steering motor model for a distributed drive vehicle, as well as a two-degree-of-freedom model including yaw rate and sideslip angle. The decoupling unit is used to decouple the two-degree-of-freedom model and establish multiple single-wheel dynamic models. An integrated construction unit is used to establish a steering control error model based on multiple single-wheel dynamic models, preset ideal values ​​of the center of gravity sideslip angle and the yaw rate of each wheel, and to establish an integrated steering error model based on the steering control error model, the steering motor model of the distributed drive vehicle, and the constructed angular velocity error model of each wheel motor. The state space reconstruction unit is used to decouple and reconstruct the state error variables in the integrated steering error model in sequence to obtain the controllable state equation. Control law design unit, used for: Based on the controllable state equation, a fractional-order terminal sliding surface is designed, and a virtual control law to compensate for unmatched disturbances is generated. Based on the error between the virtual control law and the actual output, a full-order terminal sliding surface is established, and an actual control law is designed to make the error converge on the full-order terminal sliding surface. The steering control unit is used to fuse the virtual control law and the actual control law to generate four-wheel steering angle control signals and yaw moment control signals to perform steering control on the distributed drive vehicle.

Images