A vehicle platoon longitudinal and lateral control method considering motor fault tolerance

Abstract

This invention relates to the field of electric vehicle platoon control technology, specifically to a vehicle platoon longitudinal and lateral control method considering motor fault tolerance. First, a longitudinal controller for the vehicle platoon is designed based on a model predictive control (MMC) algorithm, and a slip ratio controller is also designed based on MMC to ensure the longitudinal stability of following vehicles. Second, a lateral controller for following vehicles is designed based on MMC and sliding mode control (SMC) algorithms, where the designed SMC controller ensures the lateral stability of following vehicles. Finally, a motor fault-tolerant controller for following vehicles is designed based on SMC, rationally distributing the additional yaw torque output from the second-order SMC controller and the torque output from the longitudinal controller. Based on the distribution result, torque is output to the four motors of the vehicle, ensuring stability when the following vehicle motor fails. This improves the longitudinal and lateral control accuracy of the electric vehicle platoon and guarantees the stability of following vehicles.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle queuing control technology, and more specifically to a vehicle queuing longitudinal and lateral control method that takes into account motor fault tolerance. Background Technology

[0002] With the continuous increase in the number of cars, transportation systems face enormous challenges. Safe, efficient, and energy-saving driving has become a focal point of concern. Vehicle platooning control, as a form of formation control, currently sees most research focused on longitudinal control. The main evaluation metrics for longitudinal platooning control are the distance between adjacent vehicles and the speed of the following vehicle. However, in actual control, model-based controllers are susceptible to changes in vehicle mass, tire lateral stiffness, and other parameters, as well as interference from external conditions such as crosswinds and road surface unevenness. These factors all affect the control accuracy of the desired distance and speed. Furthermore, following vehicles may lose control due to slippage during longitudinal movement, leading to rear-end collisions. Lateral control of the vehicle platoon can be achieved by having following vehicles track the real-time trajectory of the lead vehicle. However, when following vehicles are following a path, sideslip may occur, and motor failure can cause instability and serious traffic accidents. Summary of the Invention

[0003] The purpose of this invention is to provide a vehicle platoon longitudinal and lateral control method that takes into account motor fault tolerance, so as to reduce the risk of longitudinal slippage and lateral sideslip of the platoon following vehicles, and improve the control accuracy and stability of the vehicle in both lateral and longitudinal directions.

[0004] To achieve the above objectives, the present invention provides a vehicle platoon longitudinal and lateral control method considering motor fault tolerance, comprising the following steps:

[0005] Design of a longitudinal controller for vehicle platooning based on model predictive control algorithm;

[0006] Design a lateral controller for following a vehicle based on model predictive control algorithm and sliding mode control algorithm;

[0007] Design a fault-tolerant motor controller for following a vehicle based on a sliding mode control algorithm.

[0008] The vehicle platoon longitudinal and lateral control method considering motor fault tolerance adopts a leading vehicle and navigator following communication topology and a fixed vehicle platoon spacing strategy to establish a vehicle platoon longitudinal model.

[0009] The vehicle platoon longitudinal controller includes a robust model prediction longitudinal controller and a model prediction slip ratio controller.

[0010] In the process of designing a vehicle platoon longitudinal controller based on the model predictive control algorithm, the error between the state values ​​at time k and time k-1 is used as feedback to form a robust model predictive longitudinal controller, which outputs the desired acceleration. Then, the desired acceleration is converted into torque to output the desired torque to control the vehicle.

[0011] In the process of designing a vehicle platoon longitudinal controller based on model predictive control algorithm, a slip ratio controller based on model predictive control algorithm was designed on the basis of robust model predictive longitudinal controller. Specifically, a tire slip ratio model of the following vehicle was established, and then it was transformed into the form of linear state space equation. According to the control requirements, the corresponding objective function and constraints were designed, and finally the driving torque and braking torque were output to ensure that the actual tire slip ratio of the following vehicle can track the set reference safe slip ratio.

[0012] In the process of designing a lateral controller for following a vehicle based on model predictive control and sliding mode control algorithms, a vehicle planar dynamics model for lateral control and lateral stability control is established according to the characteristics of lateral, longitudinal, and yaw motions.

[0013] Specifically, the design process of the lateral controller for the following vehicle involves establishing a linear state-space equation based on the established planar dynamics model of the following vehicle, then designing the corresponding cost function and constraints, using the real-time position and heading angle of the lead vehicle as reference values, and ensuring that the following vehicle can travel along the path of the lead vehicle by outputting the desired front wheel steering angle.

[0014] In the process of establishing the vehicle planar dynamics model for lateral control and lateral stability control, a hierarchical sliding mode lateral stability controller with a disturbance observer was designed. The upper controller controls the yaw rate and sideslip angle of the following vehicle, and at the same time, a disturbance observer is added to observe the internal and external disturbances of the vehicle, and finally outputs the desired additional yaw torque. The lower controller outputs the driving torque and braking torque through a dynamic load distribution method.

[0015] In the process of designing a fault-tolerant controller for a vehicle-following motor based on a sliding mode control algorithm, the additional yaw torque output by the second-order sliding mode controller and the torque output by the longitudinal controller are reasonably allocated according to the fault-tolerant control allocation algorithm. Based on the allocation result, the torque is output to the four motors of the vehicle respectively.

[0016] This invention provides a vehicle platoon longitudinal and lateral control method considering motor fault tolerance. First, a longitudinal controller for the vehicle platoon is designed based on a model predictive control (MMC) algorithm, and a slip ratio controller is also designed based on MMC to ensure the longitudinal stability of following vehicles. Second, a lateral controller for following vehicles is designed based on MMC and sliding mode control (SMC) algorithms, where the designed SMC controller ensures the lateral stability of following vehicles. Finally, a motor fault-tolerant controller for following vehicles is designed based on the SMC algorithm. According to the fault-tolerant control allocation algorithm, the additional yaw torque output by the second-order SMC controller and the torque output by the longitudinal controller are rationally allocated. Based on the allocation result, torque is output to the four motors of the vehicle, thereby ensuring stability when the following vehicle motor fails. This improves the longitudinal and lateral control accuracy of the electric vehicle platoon and ensures the stability of following vehicles. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating a vehicle queuing longitudinal and lateral control method that considers motor fault tolerance according to the present invention.

[0019] Figure 2 This is a schematic diagram of the vehicle queue longitudinal controller framework of the present invention.

[0020] Figure 3 This is a schematic diagram of the longitudinal model of the vehicle queue of the present invention.

[0021] Figure 4 This is a schematic diagram of the longitudinal dynamics model of the following vehicle according to the present invention.

[0022] Figure 5 This is a schematic diagram of the planar dynamics model of the following vehicle according to the present invention.

[0023] Figure 6 This is a schematic diagram of the lateral stability controller framework of the present invention.

[0024] Figure 7 This is a schematic diagram of the sliding mode fault-tolerant controller framework of the present invention. Detailed Implementation

[0025] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0026] Please see Figure 1 This invention proposes a vehicle platoon longitudinal and lateral control method considering motor fault tolerance, comprising the following steps:

[0027] S1: Design of a longitudinal controller for a vehicle platoon based on a model predictive control algorithm;

[0028] S2: Design of a lateral controller for following a vehicle based on model predictive control algorithm and sliding mode control algorithm;

[0029] S3: Design of a fault-tolerant motor controller for following a vehicle based on sliding mode control algorithm.

[0030] The following describes the specific implementation steps of the present invention:

[0031] S1: Design of a longitudinal controller for a vehicle platoon based on a model predictive control algorithm;

[0032] Please see Figure 2 The provided longitudinal controller framework consists of a robust model prediction controller and a model prediction slip ratio controller. Specifically:

[0033] A longitudinal model of the vehicle platoon is established using a lead-leader following communication topology and a fixed inter-vehicle platoon spacing strategy. For details, please refer to [link to relevant documentation]. Figure 3 The provided vehicle queue longitudinal model is as follows:

[0034] The expected distance error between the two vehicles is expressed as:

[0035] d error =x1-x2-L des -l

[0036] Where x1 and x2 represent the distances between the front and rear vehicles and the inertial reference point, respectively, and l represents the length of the lead vehicle. des This indicates the expected distance between the two vehicles.

[0037] Replace the expected distance between the two vehicles with a strategy based on a fixed workshop time interval:

[0038] L des =τv2+d safe

[0039] The spacing error between the replaced vehicle queues is as follows:

[0040] d error =x1-x2-(τv2+d safe )-l

[0041] The spacing error between vehicle queues is rewritten as:

[0042] d error =LL des

[0043] The relative speed between two adjacent vehicles is expressed as:

[0044] Δv=v1-v2

[0045] The desired longitudinal acceleration of a vehicle is usually represented by a first-order inertial element:

[0046]

[0047] In the formula: a2 is the actual longitudinal acceleration of the following vehicle; k is the system gain; τ′ is the time constant; a des It is the desired longitudinal acceleration output by the controller;

[0048] By differentiating the vehicle platoon spacing error, the relative velocity between adjacent vehicles, and the expected longitudinal acceleration of the vehicles, the differential equation of the longitudinal model of the vehicle platoon can be obtained, which is expressed as:

[0049]

[0050] A longitudinal controller for vehicle platooning was designed based on a model predictive control algorithm, and a longitudinal controller based on a robust model predictive control algorithm was also designed; specifically:

[0051] The differential equations of the longitudinal model of the vehicle queue are rewritten in state-space form:

[0052]

[0053] Where Γ1=[d error Δv a2] T , as a state variable; w1 = a1, As a control input.

[0054] The above state equations represent a continuous system. However, in practical control systems, model predictive control requires a discrete control model. Therefore, the continuous equations are discretized using the forward Euler method, and can be expressed as follows:

[0055]

[0056] in: T t1 Indicates the sampling step size.

[0057] To reduce or eliminate the static error of the control system and effectively constrain the increment of the control system, the control input quantity is... Transform into control increment make

[0058] The discrete state equations in incremental form are expressed as follows:

[0059]

[0060] in: in, It is N u ×N x 0-dimensional matrix It is N u 3D identity matrix It is N u dimensional column matrix;

[0061] Considering potential vehicle parameter errors and control errors caused by internal and external disturbances during vehicle modeling, the predicted state of the system at time k+1 is corrected by the error between the state value Γ1(k) at time k and the state value Γ1(k-1) at time (k-1). The prediction error equation at time k can be expressed as:

[0062] ΔΓ e (k)=Γ1(k)-Γ1(k-1)

[0063] The prediction error ΔΓ in the prediction error equation e (k) Considering the discrete state equations, an improved state equation can be obtained, expressed as:

[0064]

[0065] In the formula: Z is the gain matrix, Z = diag(z1,z2,z3).

[0066] Next, through the prediction derivation of the improved state equation, the prediction output equation can be obtained, expressed as:

[0067] Y1(k)=Ω1Γ1(k)+Θ1ΔU1(k)+G1E1(k)+D1

[0068] In the formula: (N c1 To control the step size, N p1 (For predicting step size)

[0069] To control the state variables of a vehicle queuing control model while seeking the optimal control variables and control increments, a cost function is designed, expressed as:

[0070]

[0071] Among them, Y 1ref For reference value, Y 1ref =(d error_ref ,Δv ref ,a1); Q1, R1, S1 represent the weight matrices of the system's state variables, control increments, and control variables, respectively; ω1 represents the relaxation factor; ρ1 represents the weight parameter of the relaxation factor.

[0072] The physical meaning of the cost function expression is explained as follows. The first term indicates that by adjusting the weight of Q1, the spacing of the vehicle queue can be ensured to be within a safe range, and the following vehicles can keep up with the speed of the lead vehicle. The second term outputs an optimal expected acceleration increment, and by adjusting R1, the control quantity of the system can be ensured to be as smooth as possible without excessive jitter. The third term outputs an optimal expected acceleration, and by adjusting S1, the control quantity of the system can be ensured to be limited within a certain range and not too large. The last term avoids the situation of having no solution by appropriately adjusting ρ1 and ω1.

[0073] To facilitate computation in a computer, the cost function expression is converted into a standard quadratic form as follows:

[0074]

[0075] Among them: ζ=Ω1Γ1(k)+G1E1(k)+D1-Y 1ref .

[0076] To ensure that the control input and control increment of the system do not change too much, the following constraints are added to the system:

[0077] U 1min ≤U1≤U 1max

[0078] ΔU 1max ≤ΔU1≤ΔU 1min

[0079] The standard quadratic cost function, combined with system constraints, can then be transformed into the following quadratic programming problem:

[0080]

[0081]

[0082] The quadratic programming problem is solved using a quadratic programming function. The first variable in the series of optimal control increments obtained from the solution is then applied to the system, completing the dynamic optimization process.

[0083]

[0084]

[0085] Finally, for ease of calculation, this section does not consider the motor model of the following vehicle. The output 'a' of the designed model predicts the longitudinal controller after solving. des The direct conversion to torque is as follows:

[0086]

[0087] In the formula: m2 represents the mass of the following vehicle; R t2 This indicates the effective radius of the tires of the following vehicle.

[0088] A longitudinal controller for vehicle platooning was designed based on a model predictive control algorithm, and a slip ratio controller was also designed based on a model predictive control algorithm; specifically:

[0089] The influence of tire slip ratio on the longitudinal dynamic performance of a vehicle is studied by establishing a longitudinal dynamics model that includes the longitudinal motion of the vehicle and the rotational motion of the four wheels. For details, please refer to [link / reference needed]. Figure 4 The provided longitudinal dynamic model, through force analysis, yields the dynamic equations, which are expressed as follows:

[0090]

[0091]

[0092] F fi =u λi F Ni

[0093] Among them: J wi The moment of inertia of the tire; w i It is the angular velocity of the tire; F fi It is the friction force of the tire; F Ni It is the vertical load on a single wheel of the vehicle; u λi Represents the slip ratio λ i The relevant coefficient of friction;

[0094] Among them, u λi Can be achieved through slip ratio λ i It can be expressed as the following equation:

[0095]

[0096] Where α1, α2, and α3 are constants, taking different values ​​depending on the road surface adhesion coefficient, and the vehicle tire slip ratio λ i It can be expressed as:

[0097]

[0098] Where: λ i The value range is [0,1].

[0099] Differentiating the slip ratio of the vehicle tires yields λ. i The expression for the differential equation is:

[0100]

[0101] The differential equation for the slip ratio after deformation can be expressed as:

[0102]

[0103] The vertical load F on a single wheel of the vehicle in the differential equation of slip ratio Ni It can be expressed as follows:

[0104]

[0105]

[0106]

[0107]

[0108] Where: a f2 This indicates the distance from the vehicle's center of gravity to the front axle; b r2 Indicates the distance from the vehicle's center of gravity to the rear axle; h g2 Indicates the height of the center of gravity of the following vehicle; a y2 d represents the lateral acceleration following the vehicle. f2 d r2 These represent the wheel track from the vehicle's center of gravity to the front and rear axles, respectively.

[0109] Transforming the differential equation of slip ratio yields the nonlinear dynamic equation, which is expressed as: The state variables are Γ2=[λ1,λ2,λ3,λ4], and the driving / braking torque is... To control the input. Since nonlinear controllers designed with nonlinear equations require complex computational processing, increasing the computational burden and making it difficult to guarantee the controller's real-time performance and stability, linear time-varying models are linearized to take advantage of their ease of prediction and low computational cost. At the current operating point... Performing a first-order Taylor expansion on the state equations yields a linear time-varying equation, expressed as:

[0110]

[0111] in:

[0112] Discretizing the linear time-varying equation using the forward Euler method, we can derive the incremental equation, which is expressed as:

[0113]

[0114] in: (I represents the identity matrix, T) t2 (Represents the sampling step size).

[0115] Similarly, the control quantity in the linear time-varying equation Transforming it into control increments yields an incremental equation, expressed as:

[0116]

[0117] In the formula: I m This represents a matrix with m rows of 1s and 0s. m This represents a matrix with m columns of 0. m×n This represents a matrix with m rows and m columns of 0.

[0118] By deriving the prediction from the incremental equation, we can obtain the following prediction output equation, expressed as:

[0119] Y2(k)=Ω2Γ2(k)+Θ2ΔU2(k)+G2E2(k)+D2

[0120] Using the slip ratio of the four wheels of the following vehicle as the control objective, a cost function is designed as follows:

[0121]

[0122] The cost function is transformed into a quadratic form, and constraints on control quantity and control increment are added. Finally, the quadratic programming problem is solved using MATLAB.

[0123] S2: Design of a lateral controller for following a vehicle based on model predictive control algorithm and sliding mode control algorithm;

[0124] Specifically, a lateral controller for following a vehicle was designed based on model predictive control and sliding mode control algorithms, including establishing a vehicle planar dynamics model for lateral control and lateral stability control based on the characteristics of lateral, longitudinal, and yaw motions.

[0125] Specifically, when a vehicle is in motion, it exhibits lateral, longitudinal, and yaw motion characteristics. Vehicle planar dynamics models accurately reflect these motion states of a real vehicle and are therefore frequently analyzed when designing vehicle control algorithms. Please refer to... Figure 5 The provided planar dynamics model of the vehicle, through force analysis, yields the following differential equations:

[0126]

[0127]

[0128]

[0129] Where ΔM z It can be expressed as follows:

[0130]

[0131] in, Indicates the lateral speed of the vehicle. This indicates the vehicle's actual yaw rate. This represents the longitudinal force of the tire. δ represents the lateral force of the tire. f2 Indicates the steering angle of the vehicle's front wheels, I z2 The moment of inertia is represented by ΔM. z Indicates the additional yaw moment. This refers to the interference of uncertainties inside the vehicle and external disturbances.

[0132] Assuming the tires following the vehicle are within the linear range and the front wheel steering angle is small, the vehicle's planar dynamics model can be simplified to the following equations:

[0133]

[0134] Where, χ 2v y represents the longitudinal displacement of the vehicle. 2v C represents the lateral displacement of the vehicle. lf2 C represents the longitudinal stiffness of the front wheel. lr2 C represents the longitudinal stiffness of the rear wheel. cf2 Indicates the lateral stiffness of the front wheel, s f2 The slip ratio of the front wheels, s r2 χ represents the slip ratio of the rear wheels. 2g The y-coordinate represents the longitudinal displacement of the vehicle in the inertial coordinate system.2g This represents the lateral displacement of the vehicle in the inertial coordinate system.

[0135] A lateral controller for following a vehicle was designed based on model predictive control and sliding mode control algorithms. It also includes establishing a vehicle planar dynamics model for lateral control and lateral stability control based on the characteristics of lateral, longitudinal, and yaw motions.

[0136] Select the state variable as Front wheel steering angle δ f2 To control the input, a nonlinear dynamic equation was established. Then the nonlinear dynamic equations are linearized, and finally at the current operating point... A first-order Taylor expansion of the vehicle's planar dynamics equations yields the following linear time-varying equations:

[0137]

[0138] in:

[0139]

[0140]

[0141] After discretizing the linear time-varying equation, it is transformed into the form of control increments, and the final expression of the prediction output equation can be obtained as follows:

[0142] Y3(k)=Ω3Γ3(k)+Θ3ΔU3(k)+G3E3(k)+D3

[0143] Design a cost function as follows:

[0144]

[0145] Where: Y 3ref For reference value, Y 3ref =[θ 1ref ,y 1ref Let Q3, R3, and S3 represent the system's state variables, control increments, and the weight matrix of the control variables, respectively; ω3 represents the relaxation factor; and ρ3 represents the weight parameter of the relaxation factor. By adjusting the weight of Q3, the tracking accuracy of lateral displacement and heading angle in vehicle path tracking is ensured; by adjusting R3, large jitter is prevented; and by adjusting S3, the smoothness of vehicle turning angles is maximized. The cost function is transformed into a quadratic form, and constraints on the control variables and control increments are added. Then, a quadratic programming problem is solved using MATLAB.

[0146] A lateral controller for following a vehicle was designed based on model predictive control and sliding mode control algorithms. It also includes a lateral stability controller with a disturbance observer designed based on sliding mode control algorithms. Specifically:

[0147] A hierarchical sliding mode lateral stability controller with a perturbation observer was designed. The specific controller framework is detailed below. For more details, please refer to [link / reference needed]. Figure 6 As shown:

[0148] The upper-level controller controls the yaw rate and sideslip angle of the following vehicle, and adds a disturbance observer to observe the internal and external disturbances of the vehicle, and finally outputs the desired additional yaw moment.

[0149] The design process specifically involves the design of a second-order sliding mode controller: A sliding surface is designed as follows:

[0150] s=γ-γ df +k(β-β df )

[0151] Where k is a weighting factor and satisfies k > 0, representing the influence of the centroid sideslip angle. Differentiating the sliding mode formula yields the differential equation, expressed as:

[0152]

[0153] In the formula: It should be noted that and It is bounded, and there exists a constant. satisfy

[0154] The differential equation of the sliding surface above can be written in the following simplified sliding differential equation form:

[0155]

[0156] Where: u1 is the control variable, u1 = ΔM z b1 = 1 / I z2 The expression for a1 is as follows:

[0157]

[0158] Here, a1 is at least locally bounded, therefore there exists a positive real number. satisfy Then u1 can be defined as follows:

[0159]

[0160] in, β1′>2 (let β2′=β2 / b1, β1′=β1′ / b1).

[0161] Design of the disturbance observation controller: Let s = y1, The simplified sliding mode differential equation can then be written as the following equation:

[0162]

[0163] Let y = [y1, y2] T Then equation (3-29) can be transformed into the following equation:

[0164]

[0165] Where: g1 = [0, b1] T g2 = [0,1] T f(y) = [y2, 0] T .

[0166] The perturbation observer can then be designed as follows:

[0167]

[0168] In the formula: Let K represent the estimated value of a1, where K = [l1, l2].

[0169] Final additional yaw moment ΔM Z It can be expressed as the following equation:

[0170]

[0171] The lower-level controller outputs drive / braking torque through a dynamic load distribution method to ensure the lateral stability of the following vehicle, specifically:

[0172] The lower-level controller uses a dynamic load distribution method to convert the desired yaw torque output from the upper-level SMC controller into different torques and distribute them to the four wheels of the following vehicle. The relationship between the motor drive / braking torque and the tire longitudinal force is as follows:

[0173]

[0174] Wherein: T i Indicates driving / braking torque.

[0175] Combining the relationship between the motor drive / braking torque and the tire longitudinal force, we can obtain four expressions for the drive / braking torque as follows:

[0176]

[0177]

[0178]

[0179] Wherein: F N It is the total vertical load on the four wheels of the vehicle.

[0180] S3: Design of a fault-tolerant motor controller for following a vehicle based on sliding mode control algorithm.

[0181] Specifically, to address the instability issue that may occur when a following vehicle experiences motor failure, a sliding mode fault-tolerant controller was designed based on the lateral stability controller and combined with the optimal allocation algorithm to ensure the stability of the following vehicle when the motor fails.

[0182] For details, please refer to Figure 7 The diagram shows the framework structure of a sliding mode fault-tolerant controller. The sliding mode fault-tolerant controller consists of a second-order sliding mode controller and a fault-tolerant control distribution controller. The fault-tolerant control distribution controller assigns the motor failure conditions to a failure factor σ. i Taking this into account (assuming the failure factor is known), then consider the torque T output by the longitudinal controller. total The additional yaw torque output from the upper-level second-order sliding mode controller is rationally distributed, ultimately outputting torque to the four motors of the vehicle, thereby ensuring stability in the event of motor failure. Specifically, an optimal allocation algorithm was used to design the FTC allocation controller.

[0183] The ratio of the force currently acting on a tire to the maximum torque it can provide describes tire utilization and is expressed as:

[0184]

[0185] Where: λ i This represents tire utilization rate, with a value range of [0,1].

[0186] The longitudinal torque output by the lateral model predictive controller and the additional yaw moment output by the second-order sliding mode controller are used as the control targets, i.e.:

[0187] v=(T total ,ΔM Z ) T

[0188] The torque allocated to the four wheels by the fault-tolerant distribution controller is used as the control input, and a motor failure factor σ is introduced. i Then the following expression must be satisfied:

[0189]

[0190] in: σi The value range is [0,1].

[0191] Next, we will consider the utilization rate of four tires (λ). i The sum of squares of the motor failure factor σ i Establish a cost function, taking into account the lateral force of the tire. Since it is uncontrollable, it can be ignored. Therefore, the optimization objective function J4 can be defined as the following expression:

[0192]

[0193] Where: μ i This represents the tire-road adhesion coefficient.

[0194] Furthermore, optimizing the objective function requires satisfying the physical constraints of the equation and the motor output; therefore, the optimization problem is expressed as follows:

[0195]

[0196]

[0197] Where: W represents the adjustable weight matrix, W = diag(W Texp W ΔM );

[0198]

[0199] The physical meaning expressed in the constraints of the objective function is as follows: The first term represents adjusting W. ΔM To ensure optimal total output torque; the second item indicates adjustment of W. Texp To ensure vehicle stability in the event of motor failure, while maintaining the vehicle's longitudinal speed.

[0200] The above description discloses only one preferred embodiment of the present invention, and should not be construed as limiting the scope of the present invention. Those skilled in the art will understand that all or part of the processes of the above embodiments can be implemented, and equivalent changes made in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A vehicle platoon longitudinal and lateral control method considering motor fault tolerance, characterized in that, comprising the following steps: designing a vehicle platoon longitudinal controller based on a model predictive control algorithm; designing a lateral controller of the following vehicle based on a model predictive control algorithm and a sliding mode control algorithm; in the process of designing the lateral controller of the following vehicle based on the model predictive control algorithm and the sliding mode control algorithm, a vehicle planar dynamics model for lateral control and lateral stability control is established according to the characteristics of lateral and longitudinal and yaw motion; in the process of establishing the vehicle planar dynamics model for lateral control and lateral stability control, a hierarchical sliding mode lateral stability controller with a disturbance observer is designed, the upper controller controls the yaw rate and the center of mass side slip angle of the following vehicle, and a disturbance observer is added to observe the internal and external disturbances of the vehicle, and finally an expected additional yaw moment is output; the lower controller outputs the driving torque and the braking torque through a dynamic load distribution method; in the process of designing the motor fault-tolerant controller of the following vehicle based on the sliding mode control algorithm, the additional yaw moment output by the second-order sliding mode controller and the torque output by the longitudinal controller are reasonably distributed according to the fault-tolerant control distribution algorithm, and the torque is output to the four motors of the vehicle according to the distribution result; designing the motor fault-tolerant controller of the following vehicle based on the sliding mode control algorithm.

2. The vehicle platoon longitudinal and lateral control method considering motor fault tolerance according to claim 1, wherein the vehicle platoon longitudinal model is established by using a front vehicle and leader follower type communication topology and a fixed inter-vehicle distance platoon spacing strategy.

3. The vehicle platoon longitudinal and lateral control method considering motor fault tolerance according to claim 1, wherein the vehicle platoon longitudinal controller comprises a robust model predictive longitudinal controller and a model predictive slip ratio controller.

4. The vehicle platoon longitudinal and lateral control method considering motor fault tolerance according to claim 3, wherein in the process of designing the vehicle platoon longitudinal controller based on the model predictive control algorithm, a robust model predictive longitudinal controller is constituted by taking the error of the state values at time k and time k-1 as feedback through the model predictive control algorithm, an expected acceleration is output, and then an expected torque is output through torque conversion to control the vehicle.

5. The vehicle platoon longitudinal and lateral control method considering motor fault tolerance according to claim 3, wherein in the process of designing the vehicle platoon longitudinal controller based on the model predictive control algorithm, a slip ratio controller based on the model predictive control algorithm is designed on the basis of the robust model predictive longitudinal controller, the tire slip ratio model of the following vehicle is established, then it is converted into a linear state space equation form, and a corresponding objective function and constraint are designed according to the control requirement, and finally the driving torque and the braking torque are output to ensure that the actual tire slip ratio of the following vehicle can track the set reference safety slip ratio.

6. The vehicle platoon longitudinal and lateral control method considering motor fault tolerance according to claim 1, wherein ​ ​ ​ ​ The design process of the lateral controller of the following vehicle, specifically, a linear state space equation is established in combination with the established planar dynamics model of the following vehicle, then a corresponding cost function and constraint are designed, the real-time position and heading angle of the leading vehicle are taken as reference values, and the following vehicle can travel along the path of the leading vehicle by outputting the expected front wheel steering angle.

Images