A dual-motor steer-by-wire system and its active fault-tolerant control method

Abstract

This invention provides a dual-motor steer-by-wire system and its active fault-tolerant control method. The method includes: a front wheel steering angle tracking controller, a vehicle stability controller, and a nonlinear model control method based on a variable parameter (LPV) model. The front wheel steering angle tracking controller and the stability controller are designed based on a linear quadratic regulator (LQR) and an optimal control algorithm, including: solving for the ideal front wheel steering angle; establishing a dynamic model of the dual-motor steering actuation system; and designing the steering angle tracking controller and the stability controller based on the ideal front wheel steering angle and the steering system dynamic model. The dual-motor active fault-tolerant control includes: comparing the control torque output by the dual-motor torque sensors with that output by the front wheel steering angle tracking controller, calculating the fault coefficient, and optimizing the steering actuator model into a variable parameter (LPV) nonlinear structure containing motor fault parameters. This invention can ensure vehicle safety after a dual-motor failure.

Description

Technical Field

[0001] This invention belongs to the field of vehicle steering systems, specifically relating to a dual-motor steer-by-wire system and its active fault-tolerant control method. Background Technology

[0002] As a key component of vehicle active safety, the steer-by-wire system is primarily responsible for controlling the vehicle's active steering and yaw stability. Compared to traditional vehicle steering, the steer-by-wire system eliminates some of the mechanical structures between the steering column and the wheels, and can freely design different vehicle gear ratios according to different drivers and road conditions, thus simultaneously meeting the requirements of low-speed agility and high-speed stability.

[0003] In the design of the steering actuator, to avoid steering actuator failure due to a single motor malfunction, a dual-motor structure is adopted to achieve hardware redundancy, improving system response characteristics and vehicle safety. Furthermore, the rated torque required by the steering system can be shared by both motors, reducing the workload of individual motors and improving the overall efficiency and service life of the steering actuator.

[0004] There are few existing methods for fault-tolerant control of dual motors, especially active fault-tolerant control methods for dual-motor steer-by-wire systems. For example, Chinese invention patent application CN201611018430.1 discloses an actuator fault detection method based on a Kalman filter observer; Chinese invention patent application CN201910136329.3 discloses a dual-motor dual-power steer-by-wire system and its fault-tolerant control method, which studies the dual-motor operating mode under power failure.

[0005] Most of the fault diagnosis methods mentioned in the existing technology are based on linear models of vehicle dynamics or fault-tolerant control methods under fixed parameters. Moreover, most vehicle stability control methods focus on the vehicle's yaw rate and sideslip angle, while control methods that take the vehicle's front wheel steering angle as the research object are relatively few. Summary of the Invention

[0006] Objective of the Invention: The technical problem to be solved by the present invention is to address the shortcomings of the prior art by providing a dual-motor steer-by-wire system and its active fault-tolerant control method, which can track the front wheel steering angle signal of the vehicle in real time with high tracking error accuracy; to address the interference of the return torque feedback signal and the ideal steering angle output signal, a robust controller is designed to improve the anti-interference performance of the system; for the active fault-tolerant control of the dual motors, the fault coefficient of the control signal is designed according to the fault mode, and the linear system is optimized using a variable parameter (LPV) model.

[0007] The system includes an information acquisition module, a front wheel steering angle tracking control module, a variable parameter module, and a steering execution module;

[0008] The information acquisition module includes a steering wheel (1), a steering wheel angle sensor (2), a vehicle speed sensor (6), a left motor torque and speed sensor (7), a right motor torque and speed sensor (12), and a control module (4);

[0009] The steering wheel angle sensor (2) is fixedly connected to the steering column. The steering wheel angle sensor (2) is responsible for collecting the steering wheel angle signal input by the driver and sending the collected signal to the control module (4).

[0010] The vehicle speed sensor (6) is installed in the front wheel of the vehicle to obtain the real-time vehicle speed and send it to the control module (4);

[0011] The steering actuation module includes an actuation motor M1 (5), an actuation motor M2 (13), a left reducer (8), a right reducer (11), a left motor torque and speed sensor (7), a right motor torque and speed sensor (12), a left transmission gear (9), a right transmission gear (10), and a steering tie rod;

[0012] The left motor torque and speed sensor (7) is connected to the output shaft of the actuator motor M1 (5), and the right motor torque and speed sensor (12) is connected to the output shaft of the actuator motor M2 (13). The left motor torque and speed sensor (7) and the right motor torque and speed sensor (12) receive the speed signals of the corresponding motors and send the speed signals of the two motors to the control module (4).

[0013] The front wheel steering angle tracking control module receives signals from the control module, including steering wheel angle signals and front wheel speed signals. It designs a vehicle transmission ratio that varies with different vehicle speeds and tracks and feeds back the front wheel steering angle using an optimal control method and a linear quadratic regulator (LQR). It calculates the optimal control rate of the dual motors, i.e., the output torque of the dual motors. Simultaneously, in the event of a dual motor failure, it receives the dual motor failure coefficients from a variable parameter model and designs a robust control method based on the front wheel feedback from the linear quadratic regulator (LQR) to reduce interference from internal system errors on the front wheel steering angle tracking feedback.

[0014] The variable parameter module is designed based on the linear parameter variation (LPV) model. It is used to receive the dual motor rotation angle signal from the control module (4), calculate the fault coefficient of each motor by comparing it with the output control torque of the linear quadratic regulator (LQR) controller, and optimize the system into a nonlinear model that includes fault parameters.

[0015] The steering execution module receives the dual-motor control torque calculated by the LQR controller in the front wheel angle tracking control module through the dynamic model of the dual-motor steering execution system, and calculates the actual front wheel angle obtained by the steering execution system under the control signal.

[0016] The left motor torque and speed sensor (7) is connected to the output shaft of the actuator motor M1 (5), and the right motor torque and speed sensor (12) is connected to the output shaft of the actuator motor M2 (13). The actuator motor M1 (5) is connected to the left transmission gear (9) through the left reducer (8), and the actuator motor M2 (13) is connected to the right transmission gear (10) through the right reducer (11). The left transmission gear (9) and the right transmission gear (10) mesh with the rack. The rack is fixed to the steering tie rod. The steering tie rod is connected to the left and right front wheels (6) respectively.

[0017] The present invention also provides an active fault-tolerant control method for a dual-motor steer-by-wire system, comprising the following steps:

[0018] Step 1: Establish the relationship between the ideal front wheel steering angle and the steering wheel steering angle, design the variable transmission ratio, and solve for the ideal front wheel steering angle;

[0019] Step 2: Establish the vehicle's dual-motor steering system and its dynamic model;

[0020] Step 3: Design an angle tracking controller based on the ideal front wheel steering angle and the dual-motor steering execution system model;

[0021] Step 4: Based on the motor torque sensor signal and the control signal output by the linear quadratic regulator LQR, design the dual-motor fault coefficient and complete the motor fault-tolerant control.

[0022] Further, step 1 includes: during vehicle operation, when the driver turns the steering wheel, the steering angle sensor (2) collects the steering angle signal θ. sw The ideal front wheel steering angle signal is obtained by collecting the longitudinal vehicle speed signal u through the front wheel speed sensor (6). With steering wheel angle signal θ sw The relationship is as follows:

[0023]

[0024] Among them, i s K represents the vehicle's gear ratio, which varies depending on the vehicle's speed. u For insufficient turning coefficient, a is the wheelbase from the center of gravity to the front axle; b is the wheelbase from the center of gravity to the rear axle; L is the wheelbase between the front and rear axles; m is the vehicle mass; k1 and k2 are the lateral stiffness of the front and rear wheels, respectively; K s This is the yaw rate coefficient, ranging from 0.12 to 0.37 s. -1 .

[0025] Furthermore, step 2 specifically includes:

[0026] The dynamic model of the dual-motor steering system is as follows:

[0027]

[0028] The dynamic model of the dual-motor steering system is based on the vehicle's rack and pinion transmission mechanism and is a dynamic model based on the actual force conditions.

[0029] Where x r This represents the lateral displacement of the rack. and x r The first and second derivatives; θ s For the rotation angle of the pinion, δ f For the front wheel steering angle, m r B is the mass of the rack; r f is the rack damping coefficient; r The lateral frictional resistance of the rack; sgn(x) represents the sign function; r p T is the radius of the pinion; g1 To execute the control torque of motor M1, T g2 To execute the control torque of motor M2, T eq This is the sum of the control torques of the two motors; F R For equivalent yaw resistance; T R M is the equivalent yaw resistance torque of the rack; z For the restoring torque; K t η is the electromagnetic torque coefficient; G1 is the reduction ratio of the actuator motor; η is the transmission efficiency of the actuator motor; G2 is the transmission ratio from the rack to the tire.

[0030] Based on the dynamic model of the dual-motor steering system of the vehicle, the state-space equations concerning the front wheel steering angle and the control torque of the dual motors are derived. The state variables of the state-space model are then selected. The first-order differential signal of the state variable is in, and These are the first-order and second-order differential signals of the front wheel steering angle, respectively; the system input is the dual-motor control torque u = [T]. g1 T g2 ] T System interference The system output is the actual front wheel steering angle y = δ f The state-space model of the dual-motor steer-by-wire system is established as follows:

[0031]

[0032] Where A, B1, B2, and C are state-space matrices. C = [1 0];

[0033] The actual front wheel steering angle signal δ obtained from the dynamic model of the above vehicle dual-motor steering system is used to calculate the actual front wheel steering angle signal δ. f Using the longitudinal vehicle speed signal u as input, and employing a two-degree-of-freedom vehicle model, the vehicle's state parameters, including yaw rate ω, are calculated. r The sideslip angle β of the center of gravity and the sideslip angle α of the front wheels f The differential equation of motion for the two degrees of freedom of the vehicle is as follows:

[0034]

[0035] In the formula, I z The moment of inertia of the vehicle is 1983.8 kgm^2. and These are the first-order differential signals corresponding to the parameters;

[0036] The tire model is simplified to a linear model, and the tire self-aligning torque M is calculated. z :

[0037] Among them, t p With t m These are the trail caused by the kingpin itself and the trail caused by the tire itself, respectively. Their sum is the tire trail, which is generally 0.05m.

[0038] The tire's self-centering torque is converted from the front wheel into equivalent yaw resistance through the mechanical structure and acts on the rack.

[0039] Furthermore, in step 3, the steering angle tracking controller includes: a front wheel steering angle tracking controller and a vehicle stability controller;

[0040] The front wheel steering angle tracking controller is designed using the following method:

[0041] Based on the linear quadratic regulator LQR control algorithm, combined with the ideal front wheel steering angle signal in step 1. The actual output signal y = δ obtained from the dynamic model of the vehicle's dual-motor steering system in step 2. f The residual is The design process for the front wheel steering angle tracking controller is as follows:

[0042] The linear quadratic cost function J is designed as follows: Simplified to:

[0043] Where Q is the weighting coefficient matrix of the residual e; R is the weighting coefficient matrix of the dual-motor control quantity; and dt is the integral term.

[0044] By selecting an appropriate weight coefficient matrix Q, the linear quadratic cost function J of the system can be minimized within the feasible region, thus satisfying the design requirement of minimizing the front wheel steering angle residual value e.

[0045] Introducing the Lagrange parameter γ, the Hamiltonian function H of the cost function J is taken as follows:

[0046] H=((y r -Cx) T Q(y r -Cx)+u T Ru)+γ T (Ax+Bu)

[0047] By the minimum value theorem

[0048] The calculated optimal control rate of the closed-loop system is u = -R. -1 B T γ

[0049] Wherein, the Lagrange parameter γ is about Let λ = Px - ξ be the solution to the non-homogeneous linear system of equations, where Px is the general solution to the corresponding homogeneous system of equations and ξ is a particular solution to the non-homogeneous system of equations, and P and ξ satisfy the Riccati equations as follows:

[0050]

[0051] Therefore, the optimal control law can also be expressed as: u = -R -1 B T (Px-ξ)

[0052] In step 3, the vehicle stability controller is designed using the following method:

[0053] The state equations of the augmented system are constructed as follows:

[0054]

[0055] Among them, the augmented state vector The corresponding first-order differential vector Augmented output vector Control matrix of augmented system Input matrix Output matrix ρ is the coefficient matrix of the error vector;

[0056] Constructing the cost function of the augmented system

[0057]

[0058] Augmented system state vector The weight coefficient matrix is Dual motor control signal u * Weight coefficient matrix

[0059] Gain coefficient matrix of the corresponding augmented system Where K1 is the corresponding state variable The feedback matrix; K2 is the corresponding error quantity. The feedback matrix;

[0060] It still satisfies the Riccati equation: It is a symmetric positive definite solution that satisfies the Ricctia equation.

[0061] Enhanced system dual-motor control rate

[0062] Step 4 includes:

[0063] Common motor failure modes mainly include the following:

[0064] Partial motor failure refers to insufficient motor steering power and reduced power, described as follows:

[0065] Motor interruption refers to a failure of the steering system, in which the motor is unable to execute commands, described as follows:

[0066] λ represents the torque sensor input signal of the i-th motor; i The value of u represents the fault coefficient of the i-th motor, satisfying 0 ≤ λ ≤ 1; i This represents the control signal for the i-th motor;

[0067] The control torque of the dual motors can be summarized as follows: Use u FThis represents the actual torque signal of the two motors after the actuator fails; where λ1 and λ2 both satisfy 0≤λ≤1;

[0068] Based on the measurement signals from the left and right motor torque sensors, respectively denoted as... The dual-motor control signals output by the linear quadratic (LQR) controller are u1 and u2, respectively;

[0069] The formula for calculating the failure coefficient of the i-th motor is:

[0070] Therefore, the dynamic model of the dual-motor steer-by-wire system based on the variable parameter (LPV) module is simplified as follows:

[0071]

[0072] Simplifying, we get:

[0073] in,

[0074] The failure factors λ1 and λ2 of the dual steering motors are selected as variable parameters to model a multi-cell variable parameter (LPV) system for the steering actuator. The multi-cell model with four vertices is composed of the variable parameters λ1 and λ2, and the coordinates of the four vertices are as follows:

[0075] Q1=(0,0); Q2=(1,0); Q3=(0,1); Q4=(1,1);

[0076] Linear time-varying system matrix and input control matrix Updated to:

[0077]

[0078] Where φ0 represents the set of constant terms of the system matrix and the control matrix, including A a0 B ua0 B wa0 φ1 represents the set of terms of the matrix with respect to λ1, including A a1 B ua1 B wa1 φ2 represents the set of terms of the matrix with respect to λ2, including A a2 B ua2 B wa2 ;

[0079] In the formula,

[0080] Substitute the dual-motor failure coefficients λ1 and λ2 of the four vertices Q1, Q2, Q3, and Q4 of the multi-cell model into the updated system matrix. and input control matrix The local state matrices of the four vertices are obtained as follows:

[0081]

[0082] In the formula, the local system matrix corresponding to the j-th vertex is A. bj The local input control matrix is ​​B. ubj B wbj ;

[0083] The sampling time t is set, and the system model of the state space at each vertex is discretized using the Euler method. The weight coefficient α at each vertex is... j The calculation formula is:

[0084]

[0085] The system's discretized model simplifies to:

[0086] The formulas for calculating the system matrix and input matrix of the new model are as follows:

[0087] Among them, A bj B is the local system matrix corresponding to the j-th vertex; ubj With B wbj ρ1 and ρ2 are the local input control matrix; ρ1 and ρ2 are the weighting coefficients for the two motor fault parameters; α j represents the weight coefficients of each vertex; I is the identity matrix;

[0088] In step 4, the discretized mathematical model of the steering actuator based on the variable parameter module (LPV) is finally obtained as shown below:

[0089]

[0090] The system state matrix is ​​as follows:

[0091]

[0092] For a new discrete system with variable parameters λ1 and λ2, the optimal control law in the current feasible region is calculated using a linear quadratic (LQR) controller, as follows:

[0093] Still introducing Lagrange parameters Using the Hamiltonian function and the minimum principle, the optimal control rate for the two motors in the event of a motor failure is:

[0094]

[0095] in, It is the unique positive definite solution that satisfies the following Riccati equations:

[0096]

[0097] The beneficial effects of this invention are as follows:

[0098] This invention leverages the advantages of the Linear Quadratic Optimal Controller (LQR) algorithm in output tracking feedback systems. By studying external disturbances to the vehicle, a corresponding closed-loop system is designed, and robust control of the closed-loop system is implemented, improving vehicle steering stability. Simultaneously, this invention designs a variable parameter model (LPV) for motor fault parameters and compensates for control signals to both motors, ensuring vehicle driving safety even after a motor failure.

[0099] The method of this invention is simple and broadens the research ideas for solving the problem of automobile steering system control. Attached Figure Description

[0100] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, and the advantages of the other aspects of the present invention will become clearer.

[0101] Figure 1 This is a schematic diagram of the steering system structure of the present invention.

[0102] Figure 2 This is a schematic diagram of the dual-motor angle tracking control principle of the present invention, which includes a variable parameter model (LPV). Detailed Implementation

[0103] Example

[0104] This embodiment provides a dual-motor steer-by-wire system and its active fault-tolerant control method. The system is as follows: Figure 1 As shown, 1 is the steering wheel, 2 is the steering wheel angle sensor, 3 is the road feel motor feedback torque sensor, 4 is the control module, 5 is the left M1 motor, 6 is the front tire, 7 is the left motor angle torque sensor, 8 is the reducer 1, 9 is the transmission gear 1, 10 is the transmission gear 2, 11 is the reducer 2, 12 is the right motor angle torque sensor, 13 is the right M2 motor, and 14 is the road feel motor.

[0105] The method flow in this embodiment is as follows: Figure 2 As shown, for a front-wheel active steering vehicle, a steering system dynamic model is established. Simultaneously, using information from onboard wheel speed sensors and steering wheel angle sensors, the external inputs and observations of the steering angle tracking control system are established. The specific implementation is as follows:

[0106] Step 1: Acquire the steering angle signal θ using the steering wheel angle sensor. swThe wheel speed sensor collects the longitudinal vehicle speed signal u at this moment, which is 2π rad and 20 m / s respectively. This signal can be obtained through the simulation software Carsim.

[0107] Design the gear ratio for an active steering car:

[0108]

[0109] In the formula, K u For insufficient turning coefficient, a is the wheelbase from the center of gravity to the front axle, 0.983m; b is the wheelbase from the center of gravity to the rear axle, 1.596m; L is the vehicle wheelbase, 2.579m; m is the vehicle weight, 1283.9kg; k1 and k2 are the lateral stiffness of the front and rear tires, respectively -95000N / rad and -90000N / rad in this embodiment; K s This is the yaw rate coefficient, ranging from 0.12 to 0.37 s. -1 .

[0110] Step 2: Establish a dynamic model of the vehicle's dual-motor steering system;

[0111] The dynamic model of the vehicle's dual-motor steering system is as follows:

[0112]

[0113] Based on the above dynamic model, the state-space equations for the vehicle's front wheel steering angle and the dual-motor control current are derived. The state variables of the state-space model are taken as follows: The system input is the dual-motor output torque u = [T] g1 T g2 ] T The system interference is The system output is the actual front wheel steering angle y = δ f The state-space model of the dual-motor steer-by-wire system is established as follows:

[0114]

[0115] in,

[0116] In the formula, m r B is the mass of the rack, 2 kg. r The rack damping coefficient is 90 N / m; r p The radius of the pinion is 0.007367 m; F R For equivalent yaw resistance; f r For frictional resistance; M z For the restoring torque; K tη is the electromagnetic torque coefficient, 0.065 Nm / A; G1 is the reduction ratio of the actuator motor, 8; η is the transmission efficiency of the actuator motor, 98%; G2 is the transmission ratio from rack to tire, 0.14.

[0117] The actual front wheel steering angle signal δ obtained from the dynamic model of the above vehicle dual-motor steering system is used to calculate the actual front wheel steering angle signal δ. f Using the longitudinal vehicle speed signal u as input, and employing a two-degree-of-freedom vehicle model, the vehicle's state parameters, including yaw rate ω, are calculated. r The sideslip angle β of the center of gravity and the sideslip angle α of the front wheels f The differential equation of motion for the two degrees of freedom of the vehicle is as follows:

[0118]

[0119] In the formula, I z The moment of inertia of the vehicle is 1983.8 kgm^2.

[0120] The tire model is simplified to a linear model, and the tire self-aligning torque M is calculated. z :

[0121] In the formula, (t) p +t m () represents tire trail, 0.05m;

[0122] The tire's self-centering torque is converted from the front wheel into equivalent yaw resistance through the mechanical structure and acts on the rack.

[0123] Step 3: Design the front wheel steering angle tracking controller and stability controller;

[0124] Furthermore, the front wheel steering angle tracking controller in step 3 specifically includes:

[0125] Based on the dynamic model of the vehicle dual-motor steering system described in step 2, it can be expressed as follows:

[0126]

[0127] In the formula, x is the state variable, u is the control variable, w is the system noise vector, and y is the system measurement output.

[0128] Design a quadratic performance cost function, expressed as:

[0129]

[0130] This study investigates the tracking problem of a system over an infinite time domain, assuming a terminal time t. f =∞, error vector Where the expected output vector Actual output vector y = δf ;

[0131] Simplified to:

[0132] By selecting an appropriate gain matrix, the corresponding quadratic cost function of the system can be minimized, thus satisfying the design requirement of minimizing the front wheel steering angle error.

[0133] Choose an appropriate gain matrix Q as a positive semi-definite matrix and R as a positive definite matrix, with Q = 100.

[0134] The observability and controllability matrix of the above steering system error model shows that the system is observable and controllable. Introducing the Lagrange parameter γ, the Hamiltonian function of the above cost function J is taken as follows:

[0135] H=((y r -Cx) T Q(y r -Cx)+u T Ru)+γ T (Ax+Bu)

[0136] Based on the principle of minimum value The optimal control rate for the closed-loop system is u = -R. -1 B T λ=-R -1 B T (Px-ξ), the process is as follows:

[0137] By the minimum value theorem

[0138] The calculated optimal control rate of the closed-loop system is u = -R. -1 B T γ

[0139] Wherein, the Lagrange parameter γ is about Let λ = Px - ξ be the solution to the non-homogeneous linear system of equations, where Px is the general solution to the corresponding homogeneous system of equations and ξ is a particular solution to the non-homogeneous system of equations, and P and ξ satisfy the Riccati equations as follows:

[0140]

[0141] Therefore, the optimal control law can also be expressed as: u = -R -1 B T (Px-ξ)

[0142] Furthermore, the vehicle stability control in step 3 specifically includes:

[0143] The tracking control law designed with a linear quadratic (LQR) controller can basically track a given ideal front wheel steering angle signal. However, due to the presence of disturbance terms such as self-aligning torque, it is necessary to study the stability of the system to eliminate the disturbance vector. The impact on the system was investigated, and a corresponding optimal robust controller was designed accordingly.

[0144] The control force required to design the optimal robust controller is provided by two parts: a servo compensator and a stabilizing compensator. All states of the system are observable, and the control force provided by the stabilizing compensator can be represented by the system states. The specific formula simplifies to:

[0145] u(t)=K1ξ(t)+K2n(t)=K1ξ(t)+K2x(t)

[0146] Construct the state equations of the augmented system as follows:

[0147]

[0148] Among them, the augmented state vector Augmented output vector ρ is the coefficient matrix of the error vector;

[0149] Construct the cost function of the augmented system:

[0150]

[0151] Selecting an appropriate augmented gain matrix

[0152] It still satisfies the Riccati equation: It is a symmetric positive definite solution satisfying the Ricctia equation, corresponding to the gain coefficient matrix of the augmented system.

[0153] Control rate

[0154] The specific implementation of the dual-motor steer-by-wire system model containing variable-parameter LPV in this invention is as follows:

[0155] Since motor failures are mainly classified into three types—partial failure, interruption, and jamming—the system control is quantified into the following form for partial failure and interruption faults:

[0156] Partial motor failure refers to insufficient motor steering power and reduced power, described as follows:

[0157] Motor interruption refers to a failure of the steering system, in which the motor is unable to execute commands, described as follows:

[0158] Therefore, the torque control of the dual motors can be summarized as follows:

[0159] In the formula, both λ1 and λ2 satisfy 0≤λ≤1;

[0160] Based on the measurement signals from the left motor torque sensor and the right motor sensor, respectively denoted as... The dual-motor control signals output by the linear quadratic (LQR) controller are u1 and u2, respectively;

[0161] The formula for calculating the failure coefficient of the i-th motor is:

[0162] Therefore, the model of the dual-motor steer-by-wire system based on the variable parameter model (LPV) is simplified as follows:

[0163] Simplifying, we get:

[0164] in, The failure factors λ1 and λ2 of the dual steering motors are selected as variable parameters to model a multi-cell variable parameter (LPV) system for the steering actuator. The multi-cell model with four vertices is composed of the variable parameters λ1 and λ2, and the coordinates of the four vertices are as follows:

[0165] Q1=(0,0); Q2=(1,0); Q3=(0,1); Q4=(1,1).

[0166] Furthermore, the linear time-varying system matrix and input control matrix are as follows:

[0167]

[0168] In the formula,

[0169]

[0170] Substituting the dual-motor fault coefficients λ1 and λ2 at the four vertices of the multi-cell model into the above equation, the local state matrices of the four vertices are obtained as follows:

[0171]

[0172] The sampling time is t = 0.1s, and the Euler method is used to discretize the system model of the state space at each vertex.

[0173] The weight coefficient α at each vertex j The calculation formula is:

[0174]

[0175] The system's discretized model simplifies to: x(k+1)=Ax(k)+B u u(k)+B w w(k)

[0176] in

[0177] The final discretized mathematical model of the steering actuator based on the variable parameter module (LPV) is shown below:

[0178]

[0179] The system state matrix is ​​as follows:

[0180]

[0181] For the new discrete system with variable parameters λ1 and λ2, the optimal control law in the current feasible region is calculated using a linear quadratic (LQR) controller, as follows:

[0182] Still introducing Lagrange parameters Calculated using the Hamiltonian function and the minimum principle, the optimal control rate for the two motors in the event of a motor failure is:

[0183]

[0184] in, It is the unique positive definite solution that satisfies the following Riccati equations:

[0185]

[0186] In its specific implementation, this application provides a computer storage medium and a corresponding data processing unit. The computer storage medium is capable of storing a computer program, which, when executed by the data processing unit, can run the invention's content regarding a dual-motor steer-by-wire system and its active fault-tolerant control method, as well as some or all of the steps in various embodiments. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0187] Those skilled in the art will clearly understand that the technical solutions in the embodiments of the present invention can be implemented using computer programs and their corresponding general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of computer programs, i.e., software products. These computer program software products can be stored in a storage medium and include several instructions to cause a device containing a data processing unit (which may be a personal computer, server, microcontroller, MUU, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

[0188] This invention provides a dual-motor steer-by-wire system and its active fault-tolerant control method. Many methods and approaches exist for implementing this technical solution; the above description is merely a preferred embodiment of the invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications should also be considered within the scope of protection of this invention. All components not explicitly stated in this embodiment can be implemented using existing technologies.

Claims

1. A dual-motor steer-by-wire system, characterized in that, It includes an information acquisition module, a front wheel steering angle tracking control module, a variable parameter module, and a steering execution module; The information acquisition module includes a steering wheel (1), a steering wheel angle sensor (2), a vehicle speed sensor (6), a left motor torque and speed sensor (7), a right motor torque and speed sensor (12), and a control module (4); The steering wheel angle sensor (2) is fixedly connected to the steering column. The steering wheel angle sensor (2) is responsible for collecting the steering wheel angle signal input by the driver and sending the collected signal to the control module (4). The vehicle speed sensor (6) is installed in the front wheel of the vehicle to obtain the real-time vehicle speed and send it to the control module (4); The steering actuation module includes an actuation motor M1 (5), an actuation motor M2 (13), a left reducer (8), a right reducer (11), a left motor torque and speed sensor (7), a right motor torque and speed sensor (12), a left transmission gear (9), a right transmission gear (10), and a steering tie rod; The left motor torque and speed sensor (7) is connected to the output shaft of the actuator motor M1 (5), and the right motor torque and speed sensor (12) is connected to the output shaft of the actuator motor M2 (13). The left motor torque and speed sensor (7) and the right motor torque and speed sensor (12) receive the speed signals of the corresponding motors and send the speed signals of the two motors to the control module (4). The front wheel steering angle tracking control module receives signals from the control module (4), including steering wheel angle signals and front wheel speed signals. It designs a vehicle transmission ratio that varies with different vehicle speeds. It tracks and feeds back the front wheel steering angle using the optimal control method and the linear quadratic regulator LQR. It calculates the optimal control rate of the dual motors, i.e., the output torque of the dual motors. At the same time, after the dual motors fail, it receives the dual motor failure coefficients from the variable parameter model and designs a robust control method based on the front wheel feedback of the linear quadratic regulator LQR. The variable parameter module is designed based on a linear variable parameter model. It is used to receive the dual motor rotation angle signal from the control module (4), calculate the fault coefficient of each motor by comparing it with the output control torque of the linear quadratic regulator controller, and optimize the system into a nonlinear model containing fault parameters.

2. The system according to claim 1, characterized in that, The steering execution module receives the dual-motor control torque calculated by the LQR controller in the front wheel angle tracking control module through the dynamic model of the dual-motor steering execution system, and calculates the actual front wheel angle obtained by the steering execution system under the control signal.

3. The system according to claim 2, characterized in that, The left motor torque and speed sensor (7) is connected to the output shaft of the actuator motor M1 (5), and the right motor torque and speed sensor (12) is connected to the output shaft of the actuator motor M2 (13). The actuator motor M1 (5) is connected to the left transmission gear (9) through the left reducer (8), and the actuator motor M2 (13) is connected to the right transmission gear (10) through the right reducer (11). The left transmission gear (9) and the right transmission gear (10) mesh with the rack. The rack is fixed to the steering tie rod. The steering tie rod is connected to the left and right front wheels (6) respectively.

4. An active fault-tolerant control method for a dual-motor steer-by-wire system, characterized in that, Includes the following steps: Step 1: Establish the relationship between the ideal front wheel steering angle and the steering wheel steering angle, design the variable transmission ratio, and solve for the ideal front wheel steering angle; Step 2: Establish the vehicle's dual-motor steering system and its dynamic model; Step 3: Design an angle tracking controller based on the ideal front wheel steering angle and the dynamic model of the dual-motor steering system; Step 4: Based on the motor torque sensor signal and the control signal output by the linear quadratic regulator LQR, design the dual-motor fault coefficient and complete the motor fault-tolerant control.

5. The method according to claim 4, characterized in that, Step 1 includes: During vehicle operation, when the driver turns the steering wheel, the steering angle sensor (2) collects the steering angle signal θ. sw The ideal front wheel steering angle signal is obtained by collecting the longitudinal vehicle speed signal u through the front wheel speed sensor (6). With steering wheel angle signal θ sw The relationship is as follows: Where i s K represents the vehicle's gear ratio, which varies depending on the vehicle's speed. u For insufficient turning coefficient, a is the wheelbase from the center of gravity to the front axle; b is the wheelbase from the center of gravity to the rear axle; L is the wheelbase between the front and rear axles; m is the vehicle mass; k1 and k2 are the front wheel lateral stiffness and the rear wheel lateral stiffness, respectively; K s This is the yaw rate coefficient.

6. The method according to claim 5, characterized in that, Step 2 includes: The dynamic model of the dual-motor steering system is as follows: The dynamic model of the dual-motor steering system is based on the vehicle's rack and pinion transmission mechanism and is a dynamic model based on the actual force conditions. Where x r This represents the lateral displacement of the rack. and x r The first and second derivatives; θ s For the rotation angle of the pinion, δ f For the front wheel steering angle, m r B is the mass of the rack; r f is the rack damping coefficient; r The lateral frictional resistance of the rack; sgn represents the sign function; r p T is the radius of the pinion; g1 To execute the control torque of motor M1, T g2 To execute the motor M2 control torque, T eq This is the sum of the control torques of the two motors; F R For equivalent yaw resistance; T R M is the equivalent yaw resistance torque of the rack; z For the restoring torque; K t η is the electromagnetic torque coefficient; G1 is the reduction ratio of the actuator motor; η is the transmission efficiency of the actuator motor; G2 is the transmission ratio from the rack to the tire. Based on the dynamic model of the dual-motor steering system of the vehicle, the state-space equations concerning the front wheel steering angle and the control torque of the dual motors are derived. The state variables of the state-space model are then selected. The first-order differential signal of the state variable is in, and These are the first-order and second-order differential signals of the front wheel steering angle, respectively; the system input is the dual-motor control torque u = [T]. g1 T g2 ] T System interference The system output is the actual front wheel steering angle y = δ f The state-space model of the dual-motor steer-by-wire system is established as follows: Where A, B1, B2, and C are state-space matrices. C = [1 0]; The actual front wheel steering angle signal δ obtained from the dynamic model of the above vehicle dual-motor steering system is used to calculate the actual front wheel steering angle signal δ. f Using the longitudinal vehicle speed signal u as input, and employing a two-degree-of-freedom vehicle model, the vehicle's state parameters, including yaw rate ω, are calculated. r The sideslip angle β of the center of gravity and the sideslip angle α of the front wheels f The differential equation of motion for the two degrees of freedom of the vehicle is as follows: In the formula, I z The vehicle's moment of inertia; and These are the first-order differential signals corresponding to the parameters; The tire model is simplified to a linear model, and the tire self-aligning torque M is calculated. z : Among them, t p With t m These are the trail caused by the kingpin itself and the trail caused by the tire itself, respectively, and their sum is the tire trail.

7. The method according to claim 6, characterized in that, In step 3, the steering angle tracking controller includes: a front wheel steering angle tracking controller and a vehicle stability controller; The front wheel steering angle tracking controller is designed using the following method: Based on the linear quadratic regulator LQR control algorithm, combined with the ideal front wheel steering angle signal in step 1. The actual output signal y = δ obtained from the dynamic model of the vehicle's dual-motor steering system in step 2. f The residual is The design process for the front wheel steering angle tracking controller is as follows: Design the linear quadratic cost function J as follows: Simplified to: Where Q is the weighting coefficient matrix of the residual e; R is the weighting coefficient matrix of the dual-motor control quantity; and dt is the integral term. By selecting an appropriate weight coefficient matrix Q, the linear quadratic cost function J of the system can be minimized within the feasible region, thus satisfying the design requirement of minimizing the front wheel steering angle residual value e. Introducing the Lagrange parameter γ, the Hamiltonian function H of the cost function J is taken as follows: H=((y r -Cx) T Q(y r -Cx)+u T Ru)+γ T (Ax+B) By the minimum value theorem The calculated optimal control rate of the closed-loop system is u = -R. -1 B T γ Wherein, the Lagrange parameter γ is about Let λ = Px - ξ be the solution to the non-homogeneous linear system of equations, where Px is the general solution to the corresponding homogeneous system of equations and ξ is a particular solution to the non-homogeneous system of equations, and P and ξ satisfy the Riccati equations as follows: Therefore, the optimal control law is expressed as: u = -R -1 B T (Px-ξ).

8. The method according to claim 7, characterized in that, In step 3, the vehicle stability controller is designed using the following method: The state equations of the augmented system are constructed as follows: Among them, the augmented state vector The corresponding first-order differential vector Augmented output vector Control matrix of augmented system Input matrix Output matrix ρ is the coefficient matrix of the error vector; Constructing the cost function of the augmented system Augmented system state vector The weight coefficient matrix is Dual motor control signal u * Weight coefficient matrix Gain coefficient matrix of the corresponding augmented system Where K1 is the corresponding state variable The feedback matrix; K2 is the corresponding error quantity. The feedback matrix; It still satisfies the Riccati equation: It is a symmetric positive definite solution that satisfies the Ricctia equation; Enhanced system dual-motor control rate 9. The method according to claim 8, characterized in that, Step 4 includes: Motor failure modes include the following: Partial motor failure refers to insufficient motor steering power and reduced power, described as follows: Motor interruption refers to a failure of the steering system, in which the motor is unable to execute commands, described as follows: λ represents the torque sensor input signal of the i-th motor; i The value of u represents the fault coefficient of the i-th motor, satisfying 0 ≤ λ ≤ 1; i This represents the control signal for the i-th motor; The control torque of the dual motors can be summarized as follows: Use u F This represents the actual torque signal of the two motors after the actuator fails; where λ1 and λ2 both satisfy 0≤λ≤1; Based on the measurement signals from the left and right motor torque sensors, respectively denoted as... The dual-motor control signals output by the linear quadratic LQR controller are u1 and u2, respectively; The formula for calculating the failure coefficient of the i-th motor is: Therefore, the dynamic model of the dual-motor steer-by-wire system based on the variable parameter LPV module can be simplified as follows: Simplifying, we get: in, The failure coefficients λ1 and λ2 of the dual steering motors are selected as variable parameters to model a multi-cell variable parameter LPV system for the steering actuator. The multi-cell model with four vertices is composed of the variable parameters λ1 and λ2, and the coordinates of the four vertices are as follows: Q1=(0,0); Q2=(1,0); Q3=(0,1); Q4=(1,1); Linear time-varying system matrix and input control matrix Updated to: Where φ0 represents the set of constant terms of the system matrix and the control matrix, including A a0 B ua0 B wa0 φ1 represents the set of terms of the matrix with respect to λ1, including A a1 B ua1 B wa1 φ2 represents the set of terms of the matrix with respect to λ2, including A a2 B ua2 B wa2 ; In the formula, Substitute the dual-motor failure coefficients λ1 and λ2 of the four vertices Q1, Q2, Q3, and Q4 of the multi-cell model into the updated system matrix. and input control matrix The local state matrices of the four vertices are obtained as follows: In the formula, the local system matrix corresponding to the j-th vertex is A. bj The local input control matrix is ​​B. ubj B wbj ; The sampling time t is set, and the system model of the state space at each vertex is discretized using the Euler method. The weight coefficient α at each vertex is... j The calculation formula is: The system's discretized model simplifies to: The formulas for calculating the system matrix and input matrix of the new model are as follows: Among them, A bj B is the local system matrix corresponding to the j-th vertex; ubj With B wbj ρ1 and ρ2 are the local input control matrix; ρ1 and ρ2 are the weighting coefficients for the two motor fault parameters; α j denoted as the weight coefficients of each vertex; I is the identity matrix.

10. The method according to claim 9, characterized in that, In step 4, the discretized mathematical model of the steering actuator based on the variable parameter module LPV is finally obtained as shown below: The system state matrix is ​​as follows: For a new discrete system with variable parameters λ1 and λ2, the optimal control law in the current feasible region is calculated using a linear quadratic LQR controller, as follows: Introducing Lagrange parameters Using the Hamiltonian function and the minimum principle, the optimal control rate for the two motors in the event of a motor failure is: in, It is the unique positive definite solution that satisfies the following Riccati equations:

Images