Commercial vehicle distributed sliding plate chassis and transverse and longitudinal cooperative stability control method thereof in variable load state

Abstract

The invention discloses a commercial vehicle distributed sliding plate chassis and a transverse and longitudinal cooperative stability control method in a variable load state thereof. The commercial vehicle distributed sliding plate chassis comprises wheels, a longitudinal vehicle speed sensor, a four-wheel independent steering module, a differential braking module, a suspension module, a frame, a six-degree-of-freedom inertial measurement unit and an active rollover-sideslip prevention controller. A commercial vehicle distributed sliding plate chassis load state-motion state observer based on an improved volume Kalman filtering algorithm is constructed by introducing diagonalization transformation and a self-adaptive fading factor, vehicle load parameters and motion parameters are obtained, and the accuracy of obtaining the vehicle parameters under variable load and coupling nonlinear conditions is improved; calculating a vehicle motion reference value and constraint based on vehicle parameters, and calculating stability indexes of a four-wheel independent steering unit and a differential braking unit; and a game architecture between a four-wheel independent steering unit and a differential braking unit is constructed, an optimal control strategy is obtained through calculation, and the active stability control capability of the vehicle under the variable load condition is guaranteed.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle steering and braking control, and particularly relates to a distributed skate chassis for commercial vehicles and a transverse and longitudinal collaborative stability control method under variable load conditions. Background Art

[0002] Vehicle sideslip and rollover are very dangerous traffic accidents that can cause serious losses of life and property. Therefore, the coupled stability control of vehicle active anti-sideslip and rollover has become a current research hotspot. Vehicle coupled stability control refers to using specific control strategies to keep the vehicle stable under certain extreme conditions to prevent serious problems such as vehicle sideslip, fishtailing, and rollover. Existing commercial vehicle anti-sideslip and rollover technologies mainly focus on the independent control of each subsystem, including active steering, differential braking, and active suspension. However, with the continuous development of vehicle control technology, the limitations of the single subsystem control method are becoming increasingly apparent. The vehicle collaborative stability technology can effectively integrate the control effects of multiple subsystems, combine the advantages of each subsystem, and has gradually become a research hotspot for commercial vehicle anti-rollover control in recent years. Compared with the traditional single control method, the distributed skate chassis for commercial vehicles combines four-wheel independent steering and differential braking, which not only improves the dynamic response ability of commercial vehicles but also can effectively suppress sideslip and rollover under large load changes, and is the trend of future commercial vehicle anti-rollover control. Therefore, it is necessary to design a control method that can give full play to the flexible loading capacity of the distributed skate chassis for commercial vehicles and effectively reduce the risk of vehicle sideslip and rollover.

[0003] In the existing research on commercial vehicle active anti-rollover control methods, for example: Chinese Patent Application No. CN202111319074.8, titled "A Vehicle Anti-Rollover Control Method for Autopilot", discloses a vehicle anti-rollover control scheme based on differential braking. By calculating the lateral load transfer ratio and keeping it within a fixed range, the roll stability of the vehicle is ensured, but it is difficult to effectively handle scenarios with load changes, and the control effect is difficult to meet the requirements. Chinese Patent Application No. CN202210157766.5, titled "An Anti-Rollover Driving Decision-Making Method for Large Commercial Vehicles Considering Front Obstacles", adopts a combined control scheme of steering and braking. However, the Markov training method requires a large amount of training data, which is difficult to meet the real-time control requirements of vehicles, and at the same time, it does not effectively coordinate the control effects of the steering system and the braking system.

[0004] In addition, the existing control strategies have the following two potential problems:

[0005] First, the existing research on the collaborative stability control algorithm for commercial vehicles mainly focuses on the coordination mechanism among multiple actuators, ignoring the changes in the roll stability domain and yaw stability domain caused by the nonlinear changes in vehicle dynamics characteristics due to the random variation of the vehicle's own load. To ensure the accuracy of the state variables of commercial vehicles during control, the evolution of vehicle dynamics performance caused by variable loads should be combined with the dynamic capture of roll thresholds. However, due to different loading masses, uneven distributions, and severe body shaking during steering conditions of commercial vehicles, parameters such as the sprung mass, moment of inertia, and centroid position of the vehicle will change, leading to parameter perturbations in the vehicle system model and affecting the control effect of the controller.

[0006] Second, the existing vehicle stability controllers only perform vehicle stability control for commercial vehicles with control parameters in a fixed mode, without fully considering the control coupling generated between subsystems. The centralized collaborative control architecture will result in poor control effects.

[0007] In summary, how to give full play to the advantages of actuator redundancy in the distributed skateboard chassis system of commercial vehicles, develop collaborative control methods for subsystems such as steering and braking, effectively improve the anti-roll and anti-yaw performance of commercial vehicles, and ensure the driving stability of commercial vehicles under variable load conditions has become the key factor restricting the large-scale implementation of the commercial distributed skateboard chassis system. Summary of the Invention

[0008] Aiming at the deficiencies of the above-mentioned existing technologies, the purpose of the present invention is to provide a distributed skateboard chassis for commercial vehicles and a longitudinal and lateral collaborative stability control method under variable load conditions, so as to solve the problems that in the existing multi-actuator collaborative control technology, when performing active anti-roll and anti-yaw control under the condition of random variation of the vehicle's own load, due to the nonlinear changes in vehicle dynamics characteristics, the roll stability domain and yaw stability domain of commercial vehicles change, making it difficult to perform effective coupling control among the subsystems of commercial vehicles, resulting in poor coordination control effects of the subsystems of commercial vehicles under extreme conditions and being prone to yaw and roll.

[0009] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0010] A distributed skateboard chassis for commercial vehicles includes: wheels, in-wheel motors, four-wheel independent steering modules, differential braking modules, suspension modules, vehicle frames, six-degree-of-freedom inertial measurement units, longitudinal vehicle speed sensors, and active anti-roll and anti-yaw controllers;

[0011] The output shaft of the built-in rotor of the in-wheel motor is fixedly connected to the wheel to drive the wheel to rotate;

[0012] The longitudinal vehicle speed sensor is fixedly installed at the output shaft position of the built-in rotor of the in-wheel motor for collecting longitudinal vehicle speed signals;

[0013] The suspension module includes: an upper control arm, a lower control arm, and a suspension mechanism assembly; the upper control arm and the lower control arm are fixedly connected to the vehicle frame, and the suspension mechanism assembly is fixedly connected to the lower control arm;

[0014] The four-wheel independent steering module includes: a steering motor controller, a steering motor, a planetary gear reduction mechanism, a steering angle sensor, and a steering shaft; the steering motor controller is fixedly connected to the steering motor, and the steering motor is rotationally connected to the upper control arm through its housing; the planetary gear reduction mechanism is located below the steering motor and is fixedly connected to the steering motor, the output shaft of the planetary gear reduction mechanism is key-connected to the steering shaft, the steering angle sensor is fixedly installed at the position of the output shaft of the planetary gear reduction mechanism, the steering shaft is fixedly connected to the housing of the in-wheel motor to drive the wheel to rotate, and the lower end of the steering shaft is rotationally connected to the lower control arm; the steering motor controller controls the steering motor to drive the planetary gear reduction mechanism to rotate by receiving the steering control signal sent by the active rollover - sideslip controller, so as to drive the rotation of the steering shaft to control the rotation of the wheel; the steering motor controller also receives the steering angle signal collected by the steering angle sensor to adjust and control the steering torque;

[0015] The differential braking module includes: a brake disc, a brake caliper assembly, a brake motor controller, a brake motor, and a ball screw reduction mechanism; the brake disc is fixedly connected to the built-in rotor of the in-wheel motor, the brake caliper assembly is fixedly connected to the housing of the in-wheel motor, the ball screw reduction mechanism is fixedly connected to the brake caliper assembly, the output shaft of the brake motor is connected to the input shaft of the ball screw reduction mechanism, and the brake motor controller is fixedly connected to the brake motor; the brake motor controller receives the braking control signal sent by the active rollover - sideslip controller, controls the brake motor to drive the output shaft of the ball screw reduction mechanism to brake, so that the brake caliper assembly presses against the brake disc to achieve braking;

[0016] The six-degree-of-freedom inertial measurement unit is fixedly installed at the center of gravity position of the commercial vehicle body, and is used to measure the vehicle motion information and generate vehicle motion data, including: longitudinal acceleration, lateral acceleration, yaw angular velocity, and roll angular velocity;

[0017] The active rollover - sideslip controller is fixedly installed on the vehicle frame, and includes: an observer unit, a four-wheel independent steering unit, and a differential braking unit. The observer unit is used to generate vehicle motion parameters and load parameter signals according to the vehicle motion data and the longitudinal vehicle speed signal. The four-wheel independent steering unit is used to generate a steering control signal according to the signals generated by the observer unit and the braking control signal generated by the differential braking unit. The differential braking unit is used to generate a braking control signal according to the signals generated by the observer unit and the steering control signal generated by the four-wheel independent steering unit.

[0018] A method for transverse and longitudinal cooperative stability control under variable load conditions of a distributed skate chassis for commercial vehicles, based on the above chassis, comprises the following steps:

[0019] 1) Establish a vehicle dynamics model for commercial vehicles to construct a vehicle state observation model for commercial vehicles. Through the observation model, a load state-motion state observer for the distributed skate chassis of commercial vehicles based on an improved cubature Kalman filter algorithm is constructed to obtain vehicle load parameters and motion parameters;

[0020] 2) Reconstruct the vehicle dynamics model for commercial vehicles according to the vehicle load parameters and motion parameters calculated in step 1), update the reference values and constraints of the vehicle motion parameters, update the incremental stability constraints and physical execution constraints of the four-wheel independent steering unit and the differential braking unit, and calculate the stability indicators of the four-wheel independent steering unit and the differential braking unit;

[0021] 3) Based on the stability indicators of the four-wheel independent steering unit and the differential braking unit obtained in step 2), calculate the coordination parameters of the four-wheel independent steering unit and the differential braking unit; combine the physical execution constraints, incremental execution constraints of the four-wheel independent steering unit and the differential braking unit obtained in step 2) and the reference values of the vehicle motion parameters to construct a cost function for the four-wheel independent steering unit and the differential braking unit, and combine the coordination parameters of the four-wheel independent steering unit and the differential braking unit to construct a comprehensive cost function with the property of a convex function, calculate the Pareto solution of the execution strategies of the four-wheel independent steering unit and the differential braking unit, obtain the optimal control strategies of the four-wheel independent steering unit and the differential braking unit, perform information interaction between the four-wheel independent steering unit and the differential braking unit and send the optimal control strategies to the steering motor controller and the braking motor controller for execution.

[0022] Further, the establishment of the vehicle dynamics model for commercial vehicles in step 1) includes a vehicle body dynamics model, a magic tire model, and a wheel dynamics model;

[0023] The vehicle body dynamics model is established as follows:

[0024]

[0025]

[0026] In the formula, m s is the sprung mass, and the sprung mass represents the mass borne by the commercial vehicle suspension; m us is the unsprung mass, representing the mass borne by the non-suspension of the commercial vehicle; m is the vehicle mass; h s is the roll height of the vehicle center of mass; K φ is the roll stiffness; C φ is the roll damping; g is the acceleration due to gravity; ΣF x , ∑Fy , ∑M z respectively represent the resultant force of the vehicle in the x-axis direction, the resultant force of the vehicle in the y-axis direction, and the resultant moment of the vehicle about the z-axis; φ is the body roll angle; is the body roll angular velocity; is the derivative of the body roll angular velocity; a x , a y , a ys are respectively the longitudinal acceleration of the vehicle, the lateral acceleration of the vehicle, and the lateral acceleration of the sprung mass; V x , V y are respectively the longitudinal velocity and the lateral velocity of the vehicle; are respectively the derivative of the longitudinal velocity of the vehicle and the derivative of the lateral velocity of the vehicle; ω r is the yaw angular velocity of the vehicle; is the derivative of the yaw angular velocity of the vehicle; I xzs is the product of inertia of the sprung mass about the x and z axes; R xs , R z are respectively the radius of gyration of the sprung mass about the x-axis and the radius of gyration of the whole vehicle about the z-axis; I xs , I z , I xzs are respectively the moment of inertia of the sprung mass about the x-axis, the moment of inertia of the whole vehicle about the z-axis, and the combined moment of inertia of the whole vehicle about the x and z axes; O represents the calculation parameter;

[0027] Through force analysis, the resultant force of the vehicle body dynamics model in the x-axis direction is:

[0028] ∑F x = F x1 cosδ1 + F x2 cosδ2 + F x3 cosδ3 + F x4 cosδ4

[0029] - F y1 sinδ1 - F y2 sinδ2 - F y3 sinδ3 - F y4 sinδ4

[0030] In the formula, F xi is the longitudinal tire force; F yi is the lateral tire force; δ i is the wheel angle, and i = 1, 2, 3, 4 correspond to the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively;

[0031] The resultant force in the y-axis direction is:

[0032] ∑F y = F x1 sinδ1 + Fx2 sinδ2 + F x3 sinδ3 + F x4 sinδ4

[0033] +F y1 cosδ1 + F y2 cosδ2 + F y3 cosδ3 + F y4 cosδ4

[0034] The resultant moment about the z - axis is as follows:

[0035]

[0036] Where a and b are the distances from the center of mass to the front axle and the rear axle respectively, and d is the distance between the tire centers;

[0037] The magic tire formula is used to establish a non - linear magic tire model as follows:

[0038]

[0039] Where Y v is the tire lateral force, tire longitudinal force or tire self - aligning moment; X h is the tire slip angle or tire longitudinal slip ratio; x h , y v are calculation parameters; D t is the peak factor; B t is the stiffness factor; C t is the curve shape factor; E t is the curve curvature factor; S h is the horizontal drift of the curve; S v is the vertical drift of the curve;

[0040]

[0041] Where F z is the vertical load on the tire from the ground; s is the tire longitudinal slip ratio; B x , C x , D x , E x , BCD x are tire longitudinal force calculation parameters; a0, a1, …, a 10 are fitting parameters; α is the tire slip angle; γ is the wheel camber angle; B y , C y , D y , E y , BCD y are tire lateral force calculation parameters; b0, b1, …, b 13 are fitting parameters;

[0042] Under the combined steering and braking conditions, the longitudinal force F of the tire x and the lateral force F y are respectively expressed as:

[0043]

[0044]

[0045] In the formula, ε x , ε y , ε respectively represent the longitudinal force parameter, the lateral force parameter and the comprehensive tire force parameter;

[0046] Establish a wheel dynamics model;

[0047] Establish the vertical load model of each wheel as follows:

[0048]

[0049] In the formula, F zi represents the vertical load of the wheel. i = 1, 2, 3, 4 respectively represent the left front wheel, the right front wheel, the left rear wheel, and the right rear wheel. L represents the wheelbase of the vehicle's front and rear axles; H1 and H2 respectively represent the height of the vehicle's center of mass from the ground and the height of the center of mass of the sprung mass from the ground;

[0050] Establish the side slip angle model of each tire as follows:

[0051]

[0052] In the formula, α i represents the side slip angle of each wheel. i = 1, 2, 3, 4 respectively represent the left front wheel, the right front wheel, the left rear wheel, and the right rear wheel.

[0053] Furthermore, in step 1), a commercial vehicle vehicle state observation model including vehicle load parameters and motion parameters is established based on the commercial vehicle vehicle dynamics model. The load parameters include the sprung mass m s and the roll height h of the vehicle's center of mass s , and the motion parameters include the body roll angle φ, the sideslip angle β of the vehicle's center of mass, and the vehicle yaw rate ω r ; Establish a commercial vehicle vehicle dynamics augmented model including vehicle load parameters and motion parameters as follows:

[0054]

[0055] In the formula, represents the derivative of the sprung mass, represents the derivative of the roll height of the vehicle's center of mass;

[0056] Based on the augmented model of commercial vehicle vehicle dynamics and sensor data, including: longitudinal vehicle speed V x , yaw rate ω r , roll rate longitudinal acceleration a x and lateral acceleration a y , a system continuous state space model is established as follows:

[0057]

[0058] u = [δ1 δ2 δ3 δ4 F x1 F x2 F x3 F x4 T

[0059] In the formula, X represents the state vector, that is, the vector to be observed; represents the derivative of the state vector; Z represents the measurement vector; u represents the system input vector; g(·), h(·) represent the continuous state function and the continuous observation function respectively; w and v represent the system process noise and the measurement noise respectively, w ∼ N(0, Q), v ∼ N(0, R), and both noises are Gaussian white noises;

[0060] The forward Euler method is used to discretize the system continuous state space model, and a system discretized state space model, that is, a commercial vehicle vehicle state observation model, is established as follows:

[0061]

[0062] In the formula, k represents the time; X(k + 1) represents the state vector at time k + 1, X(k) represents the state vector at time k, Z(k) represents the measurement vector at time k, w(k) represents the system process noise at time k, v(k) represents the system measurement noise at time k, u(k) represents the system input at time k, and G(·), H(·) represent the state equation and the observation equation respectively;

[0063] The state equation and output equation of the commercial vehicle vehicle state observation model are expressed as:

[0064] G(X(k), u(k)) = X(k) + g(X(k), u(k)) · T

[0065]

[0066] In the formula, V x (k) represents the longitudinal vehicle speed at time k, ω r (k) represents the yaw rate of the vehicle at time k, represents the roll rate of the vehicle at time k, a x ​(k) represents the longitudinal acceleration of the vehicle at time k, a y (k) represents the lateral acceleration of the vehicle at time k, and T is the sampling time of the observation system.

[0067] Furthermore, in the step 1), an observer for the load state - motion state of the commercial vehicle distributed skateboard chassis based on the improved cubature Kalman filter is constructed through the observation model, and the vehicle load parameters and motion parameters are calculated, specifically including:

[0068] 11) Initialize the system state as follows:

[0069]

[0070] In the formula, X0 is the initial value of the system state, is the initial observed value of the system state, and it satisfies P X,0 is the initial value of the system state error covariance matrix, and E(·) represents the expected value calculation function;

[0071] 12) Generate the cubature point set and calculate the system state cubature point set as follows:

[0072] The cubature Kalman filter algorithm is improved by using diagonalization transformation to prevent the reduction of algorithm stability caused by the loss of positive definiteness of the state error covariance matrix during the iteration process. The following transformation is performed on the error covariance matrix:

[0073] Set A d as an n - order real symmetric matrix, then there is an n - order orthogonal matrix V such that V T A d V = V -1 A d V = D, that is, A d = V T DV, where D is a diagonal matrix with the n eigenvalues of A d as the diagonal elements. Then the error covariance matrix P X,k of the system state is expressed as:

[0074]

[0075] In the formula, V X,k represents the auxiliary matrix for the decomposition of the error covariance matrix, D X,k represents the decomposition matrix of the error covariance matrix, represents the eigenvalues of the error covariance matrix of the system state, i p = 1, 2, …, n, representing the sequence of the eigenvalues of the error covariance matrix of the system state;

[0076] Define the DCP function as:

[0077]

[0078] In the formula, S X,k represents the matrix for diagonalization decomposition of the system state error covariance matrix;

[0079] Calculate the system state cubature point set as follows:

[0080]

[0081] In the formula, is the cubature point set, i d = 1, 2, …, n, m d = 2n, I n×n represents the n-dimensional identity matrix, m d represents the system cubature point set parameter, is the calculation of the system state cubature point set, is the observed value of the system state at time k;

[0082] 13) Update the observer system state as follows:

[0083] 131) Propagate the system state cubature point set as follows:

[0084]

[0085] In the formula, represents the system propagated state cubature point set at time k + 1, u k represents the system input value at time k;

[0086] 132) Update the observer system predicted state

[0087]

[0088] In the formula, represents the predicted value of the system state at time k + 1;

[0089] 133) Update the system state error covariance matrix P X,k+1|k ;

[0090] Calculate the initial value of the system state error covariance matrix as follows:

[0091]

[0092] In the formula, is the initial value of the system state error covariance matrix at time k + 1, Q w,k is the system state noise covariance matrix at time k;

[0093] Generate the initial value of the system measurement cubature point set and calculate the initial value of the system propagated measurement cubature point set as follows:

[0094]

[0095] wherein, is the predicted initial value of the system state error covariance matrix decomposition matrix at time k + 1, is the initial value of the system measurement cubature point set at time k + 1, is the initial value of the system propagated measurement cubature point set at time k + 1;

[0096] Update the initial value of the observer system measurement prediction as follows:

[0097]

[0098] wherein, is the initial value of the system measurement prediction at time k + 1;

[0099] 14) Update the observer system measurement;

[0100] 141) Generate the system measurement cubature point set and calculate the system propagated measurement cubature point set as follows:

[0101]

[0102] wherein, S X,k+1|k is the system error covariance matrix decomposition matrix at time k + 1, is the system measurement cubature point set at time k + 1, is the system propagated measurement cubature point set at time k + 1;

[0103] 142) Update the observer system measurement prediction value as follows:

[0104]

[0105] 143) Update the system measurement error covariance matrix as follows:

[0106]

[0107] wherein, P Z,k+1|k is the system measurement error covariance matrix at time k + 1, R v,k is the system measurement noise covariance matrix at time k;

[0108] 15) Update the observer state and measurement cross-covariance matrix P XZ,k+1|k as follows:

[0109]

[0110] 16) Update the observer gain matrix and the system state error covariance matrix as follows:

[0111]

[0112] where K k+1 is the observer gain matrix at time k+1, is the updated value of the system state at time k+1, and P X,k+1 is the updated value of the system state error covariance matrix at time k+1;

[0113] By observing the sprung mass m s , the vehicle center of mass roll height h s , the vehicle lateral velocity V y , the vehicle longitudinal velocity V x , the yaw rate ω r , and the vehicle body roll angle φ, as follows:

[0114] [V x V y ω r φm s h s T = C ICKF X

[0115] where C ICKF represents the observation selection matrix, and X represents the state vector;

[0116] By the vehicle lateral velocity V y , the vehicle longitudinal velocity V x , the required vehicle center of mass side slip angle β is calculated as follows:

[0117]

[0118] Furthermore, step 2) specifically includes:

[0119] 21) Based on the sprung mass m s and the vehicle center of mass roll height h s obtained in step 1), the established commercial vehicle dynamics model is reconstructed, and the reconstructed vehicle dynamics model is characterized as:

[0120]

[0121] where represents the system state variable, represents the derivative of the system state variable, u1 = [δ1 δ2 δ3 δ4] T represents the four-wheel independent steering unit input variable, u2 = [F x1 F x2 F x3 F x4 ​​T Denote the input variables of the differential braking unit;

[0122] Linearize the reconstructed vehicle dynamics model, and define the state increment Δx of the system at time k k , Δx k = x k - x k-1 , where, x k is the system state quantity at time k, x k-1 is the system state quantity at time k - 1, the control input increment Δu of the four-wheel independent steering unit at time k 1,k , Δu 1,k = u 1,k - u 1,k-1 , where, u 1,k is the control input variable of the four-wheel independent steering unit at time k, u 1,k-1 is the control input variable of the four-wheel independent steering unit at time k - 1, the control input increment Δu of the differential braking unit at time k 2,k , Δu 2,k = u 2,k - u 2,k-1 , where, u 2,k is the control input increment of the differential braking unit at time k, u 2,k-1 is the control input increment of the differential braking unit at time k - 1, and the increment equation is formed as follows:

[0123]

[0124] Discretize the state space equation in incremental form to obtain the incremental recurrence relation as follows:

[0125] Δx k = AΔx k-1 + B1Δu 1,k-1 + B2Δu 2,k-1

[0126] In the formula,

[0127]

[0128] In the formula, A represents the discrete system state matrix; B1 represents the input matrix of the four-wheel independent steering unit; B2 represents the input matrix of the differential braking unit; is the Jacobi matrix, and ΔT is the sampling time;

[0129] 22) Update the reference values and constraints of the vehicle motion parameters based on the commercial vehicle chassis load parameters obtained in step 1):

[0130] The control objectives of the active rollover - sideslip controller are yaw rate, vehicle center - of - mass sideslip angle, and lateral load transfer ratio. The reference value of the yaw rate is selected as the nominal value of the two - degree - of - freedom vehicle model, and the nominal value of the yaw rate is corrected according to the lateral load transfer ratio. The lateral load transfer ratio LTR is limited as follows:

[0131]

[0132] In the formula, LTR max represents the maximum lateral load transfer ratio;

[0133] Considering rollover control, the reference value of the yaw rate ω r_ref is set as:

[0134]

[0135] In the formula, K represents the stability factor of the two - degree - of - freedom model, μ represents the road adhesion coefficient, and δ represents the target front - wheel steering angle;

[0136] The reference value of the vehicle center - of - mass sideslip angle is set as the steady - state response value of the vehicle two - degree - of - freedom model. Considering the limitation of the maximum body sideslip angle by road adhesion, the nominal value of the vehicle center - of - mass sideslip angle β ref is corrected to:

[0137]

[0138] In the formula, K r represents the rear - wheel sideslip stiffness, and L represents the vehicle wheelbase;

[0139] 23) Update the incremental stability constraints and physical execution constraints of the four - wheel independent steering unit and the differential braking unit based on the commercial vehicle chassis load parameters obtained in step 1), including: the physical execution constraints of the four - wheel independent steering unit and the differential braking unit and the execution constraints in incremental form defined in step 21), and calculate the stability indicators of the four - wheel independent steering unit and the differential braking unit;

[0140] For the control system design of four - wheel independent steering, the requirements of vehicle lateral motion stability need to be met. The execution constraints of the four - wheel independent steering unit are expressed as follows:

[0141] Δu 1min ≤Δu1≤Δu 1max ,u 1min ≤u1≤u 1max

[0142]

[0143] In the formula, It represents the maximum and minimum turning angles of the steering motor in the four-wheel independent steering module within a unit time, where k1 and k2 represent the constraint calculation parameters of the four-wheel independent steering unit, and e is a constant; u 1max =-u 1min It represents the physical constraint of the steering motor's turning angle; ψ1 and ψ2 represent coefficients; σ1 represents the vehicle sideslip stability index, It represents the upper bound of the vehicle sideslip stability index, where k3 and k4 represent the stability index calculation parameters of the four-wheel independent steering unit. The lateral movement of the vehicle needs to satisfy the constraint of the vehicle lateral stability region formed by the β-ω r phase plane;

[0144] For the design of the differential braking control system, the constraints include the braking force saturation constraint limited by the actuator power and the constraint to ensure that the vehicle can always operate within the safe region of the vehicle roll motion represented by the phase diagram, which is specifically expressed as:

[0145] Δu 2min ≤Δu2≤Δu 2max ,u 2min ≤u2≤0

[0146]

[0147] In the formula, It represents the maximum and minimum requirements for the increment of the wheel longitudinal force, where k5 and k6 represent the constraint calculation parameters of the differential braking unit; u 2min =-F x,max It represents the minimum requirement for the wheel longitudinal force; It represents a coefficient; σ2 represents the vehicle rollover stability index, It represents the upper bound value of the vehicle critical rollover stability index. Among them, k7 and k8 represent the stability index calculation parameters of the differential braking unit; According to the lateral force and related state variables of the current wheel working point, the maximum single-wheel longitudinal force F under the adhesion ellipse constraint is approximately obtained x,max , and its calculation formula is as follows:

[0148]

[0149] In the formula, ρ is the curvature parameter of the friction adhesion ellipse.

[0150] Furthermore, step 3) specifically includes:

[0151] 31) Based on the vehicle stability indices σ1 and σ2 in step 23), the coordination parameters of the four-wheel independent steering unit and the differential braking unit are calculated respectively. The relationship between the coordination parameters and the stability indices is expressed by the following formula:

[0152]

[0153] In the formula, κ p respectively represent the upper and lower bounds of the coordination parameter κ p ; represents the upper bound of the stability index σ p ;

[0154] 32) According to the reference values of the vehicle motion parameters calculated in step 22), and the coordination parameters and shared information between the units of the four-wheel independent steering unit and the differential braking unit calculated in step 31), construct the cost functions of the four-wheel independent steering unit and the differential braking unit, and construct a comprehensive cost function with the property of a convex function based on the cost functions of the four-wheel independent steering unit and the differential braking unit;

[0155] For the four-wheel independent steering unit, to ensure the yaw stability of the vehicle, the cost function J1(k) is expressed as:

[0156]

[0157] In the formula, x ref = [β ref ω r_ref T represents the reference state vector, C1 is the state selection matrix, N p is the prediction horizon, Q1 represents the process state matrix of the four-wheel independent steering unit, R1 represents the input cost matrix of the four-wheel independent steering unit, and P1 represents the terminal state matrix of the four-wheel independent steering unit;

[0158] For the differential braking unit, to improve the roll stability during the vehicle steering process, the cost function J2(k) is expressed as:

[0159]

[0160] In the formula, x ref = φ ref represents the reference state, C2 is the state selection matrix, Q2 represents the process state matrix of the differential braking unit, R2 represents the input cost matrix of the differential braking unit, and P2 represents the terminal state matrix of the differential braking unit;

[0161] According to the coordination parameters of the four-wheel independent steering unit and the differential braking unit calculated in step 31), design a linear convex combination of the control objective functions as the comprehensive cost function J(U) as follows:

[0162]

[0163] In the formula, J q represents the cost functions of the four-wheel independent steering unit and the differential braking unit;​

[0164] Solve the optimal cost function to obtain the execution strategies of the four-wheel independent steering unit and the differential braking unit, and the Pareto solution strategy U is as follows:

[0165]

[0166] 33) Based on the Pareto solution obtained in step 32), construct the shared strategy of the four-wheel independent steering unit and the differential braking unit at the next moment, which is the optimal control strategy U of the four-wheel independent steering unit and the differential braking unit n,q , as follows:

[0167]

[0168] In the formula, U n,q (k|k - 1) represents the shared strategy of the four-wheel independent steering unit and the differential braking unit for constructing the k-th moment at the (k - 1)-th moment, K x represents the system state feedback gain, x m represents the system terminal state, υ q represents the control input of the four-wheel independent steering unit and the differential braking unit;

[0169] 34) Perform information interaction between the four-wheel independent steering unit and the differential braking unit, and send the steering motor controller of the four-wheel independent steering module and the braking motor controller of the differential braking module for control execution.

[0170] Advantages of the present invention:

[0171] 1. Aiming at the observation error problem caused by the significant dynamic nonlinear characteristics of commercial vehicles under extreme conditions, the present invention proposes a distributed skateboard chassis load state - motion state observer for commercial vehicles based on improved cubature Kalman filter. By introducing an adaptive fading factor into the error covariance matrix, the residual sequence is kept orthogonal to each other, and the accuracy of obtaining vehicle load parameters and motion state parameters under variable load and coupled nonlinear conditions is improved.

[0172] 2. Aiming at the problem of cooperative rollover prevention control of commercial vehicles under variable load conditions, the present invention constructs a game architecture between the four-wheel independent steering unit and the differential braking unit, and proposes a game horizontal and longitudinal cooperative stability control strategy based on four-wheel independent steering and differential braking. Based on the vehicle load parameters and motion parameters obtained by the distributed skateboard chassis load state - motion state observer of commercial vehicles, controllable constraints of the four-wheel independent steering unit and the differential braking unit are designed to ensure the active anti-skid and rollover control performance of commercial vehicles under variable load conditions, and effectively improve the driving safety of commercial vehicles. Description of the Drawings

[0173] Figure 1This is the structural diagram of the distributed skateboard chassis for commercial vehicles of the present invention.

[0174] Figure 2 This is a partial schematic diagram of the wheel position in the distributed skateboard chassis for commercial vehicles of the present invention. Detailed implementation manners

[0175] For the convenience of those skilled in the art to understand, the present invention will be further described below in conjunction with embodiments and the accompanying drawings. The content mentioned in the implementation manners does not limit the present invention.

[0176] Refer to Figure 1 - Figure 2 As shown, a distributed skateboard chassis for commercial vehicles of the present invention includes: wheels 1, in-wheel motors 9, four-wheel independent steering modules 3, differential braking modules 4, suspension modules 5, vehicle frames 6, six-degree-of-freedom inertial measurement units 7, longitudinal vehicle speed sensors 2, and active rollover - sideslip controllers 8;

[0177] The output shaft of the built-in rotor of the in-wheel motor 9 is fixedly connected to the wheel 1 to drive the wheel to rotate;

[0178] The longitudinal vehicle speed sensor 2 is fixedly installed at the output shaft position of the built-in rotor of the in-wheel motor 9 for collecting longitudinal vehicle speed signals;

[0179] The suspension module 5 includes: upper control arms 17, lower control arms 21, and a suspension mechanism assembly 16; the upper control arms 17 and the lower control arms 21 are fixedly connected to the vehicle frame, and the suspension mechanism assembly is fixedly connected to the lower control arm 21;

[0180] The four-wheel independent steering module 3 includes: a steering motor controller 15, a steering motor 10, a planetary gear reduction mechanism 12, an angle sensor 13, and a steering shaft 22; the steering motor controller 15 is fixedly connected to the steering motor 10, and the steering motor 10 is rotatably connected to the upper control arm 17 through its housing; the planetary gear reduction mechanism 12 is located below the steering motor 10 and is fixedly connected to the steering motor 10. The output shaft of the planetary gear reduction mechanism 12 is key-connected to the steering shaft. The angle sensor 13 is fixedly installed at the output shaft position of the planetary gear reduction mechanism 12. The steering shaft is fixedly connected to the housing of the in-wheel motor 9 to drive the wheel to rotate, and the lower end of the steering shaft is rotatably connected to the lower control arm; the steering motor controller controls the steering motor 10 to drive the planetary gear reduction mechanism 12 to rotate by receiving the steering control signal sent by the active rollover - sideslip controller 8, so as to drive the rotation of the steering shaft 22 to control the rotation of the wheel; the steering motor controller 15 also receives the angle signal collected by the angle sensor 13 to adjust and control the steering torque;

[0181] The differential braking module 4 includes: a brake disc 11, a brake caliper assembly 14, a brake motor controller 18, a brake motor 19, and a ball screw reduction mechanism 20; the brake disc 11 is fixedly connected to the built-in rotor of the in-wheel motor 9, the brake caliper assembly 14 is fixedly connected to the housing of the in-wheel motor 9, the ball screw reduction mechanism 20 is fixedly connected to the brake caliper assembly 14, the output shaft of the brake motor 11 is connected to the input shaft of the ball screw reduction mechanism 20, and the brake motor controller 18 is fixedly connected to the brake motor 19; the brake motor controller 18 receives a braking control signal sent by the active rollover - sideslip controller 8, controls the brake motor 19 to drive the output shaft of the ball screw reduction mechanism 20 to perform braking, and makes the brake caliper assembly 14 press against the brake disc 11 to achieve braking;

[0182] The six-degree-of-freedom inertial measurement unit is fixedly installed at the center of gravity position of the commercial vehicle body, and is used for measuring vehicle motion information and generating vehicle motion data, including: longitudinal acceleration, lateral acceleration, yaw angular velocity, and roll angular velocity;

[0183] The active rollover - sideslip controller is fixedly installed on the vehicle frame, and includes: an observer unit, a four-wheel independent steering unit, and a differential braking unit. The observer unit is used for generating vehicle motion parameters and load parameter signals according to vehicle motion data and longitudinal vehicle speed signals. The four-wheel independent steering unit is used for generating a steering control signal according to the signals generated by the observer unit and the braking control signal generated by the differential braking unit. The differential braking unit is used for generating a braking control signal according to the signals generated by the observer unit and the steering control signal generated by the four-wheel independent steering unit.

[0184] A method for controlling the transverse and longitudinal collaborative stability of a commercial vehicle distributed skateboard chassis under variable load conditions according to the present invention is based on the above chassis, and the steps are as follows:

[0185] 1) Establish a commercial vehicle vehicle dynamics model to construct a commercial vehicle vehicle state observation model, and through the observation model, construct a commercial vehicle distributed skateboard chassis load state - motion state observer based on an improved cubature Kalman filter algorithm to obtain vehicle load parameters and motion parameters;

[0186] Among them, the establishment of the commercial vehicle vehicle dynamics model in step 1) includes a vehicle body dynamics model, a magic tire model, and a wheel dynamics model;

[0187] Establish a vehicle body dynamics model as follows:

[0188]

[0189] In the formula, m s is the sprung mass, and the sprung mass represents the mass borne by the commercial vehicle suspension; m usis the unsprung mass, representing the mass borne by the non-suspension of the commercial vehicle; m is the vehicle's total mass; h s is the roll center height of the vehicle's center of mass; K φ is the roll stiffness; C φ is the roll damping; g is the acceleration due to gravity; ∑F x , ∑F y , ∑M z respectively represent the resultant force of the vehicle in the x-axis direction, the resultant force of the vehicle in the y-axis direction, and the resultant moment of the vehicle about the z-axis; φ is the body roll angle; is the body roll angular velocity; is the derivative of the body roll angular velocity; a x , a y , a ys are respectively the vehicle's longitudinal acceleration, lateral acceleration, and the lateral acceleration of the sprung mass; V x , V y are respectively the vehicle's longitudinal velocity and lateral velocity; ∑ are respectively the derivatives of the vehicle's longitudinal velocity and lateral velocity; ω r is the vehicle's yaw angular velocity; is the derivative of the vehicle's yaw angular velocity; I xzs is the product of inertia of the sprung mass about the x and z axes; R xs , R z are respectively the radius of gyration of the sprung mass about the x-axis and the radius of gyration of the vehicle about the z-axis; I xs , I z , I xzs are respectively the moment of inertia of the sprung mass about the x-axis, the moment of inertia of the vehicle about the z-axis, and the combined moment of inertia of the vehicle about the x and z axes; O represents the calculation parameter;

[0190] Through force analysis, it can be obtained that the resultant force of the vehicle body dynamics model in the x-axis direction is:

[0191] ∑F x = F x1 cosδ1 + F x2 cosδ2 + F x3 cosδ3 + F x4 cosδ4 - F y1 sinδ1 - F y2 sinδ2 - F y3 sinδ3 - F y4 sinδ4

[0192] In the formula, F xi is the longitudinal tire force; F yi is the lateral tire force; δ i is the wheel angle, and i = 1, 2, 3, 4 correspond to the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively;

[0193] The resultant force in the y-axis direction is:

[0194] ∑F y = F x1 sinδ1 + F x2 sinδ2 + F x3 sinδ3 + F x4 sinδ4

[0195] + F y1 cosδ1 + F y2 cosδ2 + F y3 cosδ3 + F y4 cosδ4

[0196] The resultant moment about the z-axis is:

[0197]

[0198] In the formula, a and b are the distances from the center of mass to the front axle and the rear axle respectively, and d is the distance between the tire centers;

[0199] The magic tire formula is used to establish a non-linear magic tire model as follows:

[0200]

[0201] In the formula, Y v is the tire lateral force, tire longitudinal force or tire self-aligning moment; X h is the tire slip angle or tire longitudinal slip ratio; x h , y v are calculation parameters; D t is the peak factor; B t is the stiffness factor; C t is the curve shape factor; E t is the curve curvature factor; S h is the curve horizontal drift; S v is the curve vertical drift;

[0202]

[0203] In the formula, F z is the ground vertical load on the tire; s is the tire longitudinal slip ratio; B x , C x , D x , E x , BCD x are tire longitudinal force calculation parameters; a0, a1, …, a 10 are fitting parameters; α is the tire slip angle; γ is the wheel camber angle; B y , Cy , D y , E y , BCD y are the calculation parameters of the tire lateral force; b0, b1, …, b 13 are the fitting parameters;

[0204] Under the combined steering and braking conditions, the longitudinal force F x and the lateral force F y are respectively expressed as:

[0205]

[0206] In the formula, ε x , ε y , ε represent the longitudinal force parameter, the lateral force parameter and the comprehensive tire force parameter respectively;

[0207] Establish a wheel dynamics model;

[0208] Establish the vertical load model of each wheel as follows:

[0209]

[0210] In the formula, F zi represents the vertical load of the wheel, i = 1, 2, 3, 4 respectively represent the left front wheel, the right front wheel, the left rear wheel, and the right rear wheel, L represents the wheelbase of the vehicle's front and rear axles; H1, H2 respectively represent the height of the vehicle's center of mass from the ground and the height of the center of mass of the sprung mass from the ground;

[0211] Establish the sideslip angle model of each tire as follows:

[0212]

[0213] In the formula, α i represents the sideslip angle of each wheel, i = 1, 2, 3, 4 respectively represent the left front wheel, the right front wheel, the left rear wheel, and the right rear wheel.

[0214] Among them, in step 1), a commercial vehicle vehicle state observation model including vehicle load parameters and motion parameters is established based on the commercial vehicle vehicle dynamics model. The load parameters include the sprung mass m s and the roll height h of the vehicle's center of mass s , and the motion parameters include the body roll angle φ, the sideslip angle β of the vehicle's center of mass, and the yaw angular velocity ω r ; establish a commercial vehicle vehicle dynamics augmented model including vehicle load parameters and motion parameters as follows:

[0215]

[0216] In the formula, represents the derivative of the sprung mass, represents the derivative of the vehicle's roll center height;

[0217] Based on the augmented model of commercial vehicle dynamics and sensor data, including: longitudinal vehicle speed V x , yaw rate ω r , roll rate longitudinal acceleration a x and lateral acceleration a y , a system continuous state space model is established as follows:

[0218]

[0219] In the formula, X represents the state vector, that is, the vector to be observed; represents the derivative of the state vector; Z represents the measurement vector; u represents the system input vector; g(·) and h(·) respectively represent the continuous state function and the continuous observation function; w and v respectively represent the system process noise and measurement noise, w ~ N(0, Q), v ~ N(0, R), and both noises are Gaussian white noises;

[0220] The forward Euler method is used to discretize the system continuous state space model to establish a system discretized state space model, that is, the commercial vehicle state observation model, as follows:

[0221]

[0222] In the formula, k represents the time; X(k + 1) represents the state vector at time k + 1, X(k) represents the state vector at time k, Z(k) represents the measurement vector at time k, w(k) represents the system process noise at time k, v(k) represents the system measurement noise at time k, u(k) represents the system input at time k, and G(·) and H(·) respectively represent the state equation and the observation equation;

[0223] The state equation and output equation of the commercial vehicle state observation model are expressed as:

[0224] G(X(k), u(k)) = X(k) + g(X(k), u(k))·T

[0225]

[0226] In the formula, V x (k) represents the longitudinal vehicle speed at time k, ω r (k) represents the yaw rate of the vehicle at time k, represents the roll rate of the vehicle at time k, a x (k) represents the longitudinal acceleration of the vehicle at time k, a y (k) represents the lateral acceleration of the vehicle at time k, and T is the sampling time of the observation system.

[0227] Among them, in the step 1), a vehicle load state - motion state observer based on an improved cubature Kalman filter algorithm is constructed through an observation model, and vehicle load parameters and motion parameters are calculated, specifically including:

[0228] 11) Initialize the system state as follows:

[0229]

[0230] In the formula, \(X_0\) is the initial value of the system state, is the initial observation value of the system state, and it satisfies \(P\) X,0 is the initial value of the system state error covariance matrix, and \(E(\cdot)\) represents the expected value calculation function;

[0231] 12) Generate the cubature point set and calculate the system state cubature point set as follows:

[0232] The cubature Kalman filter algorithm is improved by using diagonalization transformation to prevent the algorithm stability from decreasing due to the loss of positive definiteness of the state error covariance matrix during the iteration process. The following transformation is performed on the error covariance matrix:

[0233] Set \(A\) d as an \(n\) - order real - symmetric matrix, then there exists an \(n\) - order orthogonal matrix \(V\) such that \(V\) T \(A\) d \(V = V\) -1 \(A\) d \(V = D\), that is, \(A\) d \(=\ V\) T \(DV\), where \(D\) is a diagonal matrix with the \(n\) eigenvalues of \(A\) d as diagonal elements. Then the error covariance matrix \(P\) X,k of the system state is expressed as:

[0234]

[0235] In the formula, \(V\) X,k represents the error covariance matrix decomposition auxiliary matrix, \(D\) X,k represents the decomposition matrix of the error covariance matrix, represents the eigenvalues of the error covariance matrix of the system state, \(i\) p \( = 1,2,\cdots,n\) represents the sequence of the eigenvalues of the error covariance matrix of the system state;

[0236] Define the DCP function as:

[0237]

[0238] In the formula, \(S\) X,kThe matrix representing the diagonalization decomposition of the system state error covariance matrix;

[0239] Calculate the system state cubature point set as follows:

[0240]

[0241] Where, is the cubature point set, i d = 1, 2, …, n, m d = 2n, I n×n represents the n-dimensional identity matrix, m d represents the system cubature point set parameter, X id,k is the calculation of the system state cubature point set, is the observed value of the system state at time k;

[0242] 13) Update the observer system state as follows:

[0243] 131) Propagate the system state cubature point set as follows:

[0244]

[0245] Where, represents the system propagated state cubature point set at time k + 1, u k represents the system input value at time k;

[0246] 132) Update the observer system predicted state

[0247]

[0248] Where, represents the predicted value of the system state at time k + 1;

[0249] 133) Update the system state error covariance matrix P X,k+1|k ;

[0250] Calculate the initial value of the system state error covariance matrix as follows:

[0251]

[0252] Where, is the initial value of the system state error covariance matrix at time k + 1, Q w,k is the system state noise covariance matrix at time k;

[0253] Generate the initial value of the system measurement cubature point set and calculate the initial value of the system propagated measurement cubature point set as follows:

[0254]

[0255] In the formula, is the predicted initial value of the system state error covariance matrix decomposition matrix at time k+1, is the initial value of the system measurement cubature point set at time k+1, is the initial value of the system propagated measurement cubature point set at time k+1;

[0256] Update the initial value of the system measurement prediction of the observer as follows:

[0257]

[0258] In the formula, is the initial value of the system measurement prediction at time k+1;

[0259] Introduce an adaptive fading factor to adjust the error covariance matrix online for improvement, ensuring the observation accuracy of the load state - motion state observer of the commercial vehicle distributed skate chassis;

[0260] Calculate the system measurement error as follows:

[0261]

[0262] In the formula, e k+1 is the system measurement error at time k+1, and Z k+1 is the system measurement value at time k+1;

[0263] Calculate the system measurement calculation factor as follows:

[0264]

[0265] In the formula, ι k+1|k is the system measurement calculation factor at time k+1;

[0266] Calculate the system state calculation factor as follows:

[0267]

[0268] In the formula, χ k+1|k is the system state calculation factor at time k+1;

[0269] Calculate the system H calculation factor as follows:

[0270]

[0271] In the formula, H k+1 is the system H calculation factor at time k+1;

[0272] Calculate the system M calculation factor as follows:

[0273]

[0274] Wherein, M k+1 is the system M calculation factor at the (k + 1)-th moment;

[0275] Calculate the system V calculation factor:

[0276]

[0277] Wherein, are respectively the initial system measurement error and the transpose of the initial system measurement error, θ, θ ∈ (0, 1) is the forgetting factor, V k is the system V calculation factor at the k-th moment, V k+1 is the system V calculation factor at the (k + 1)-th moment;

[0278] Calculate the system N calculation factor, as follows:

[0279]

[0280] Wherein, ζ, ζ ≥ 1 is the weakening factor, R v,k is the system measurement noise covariance matrix at the k-th moment, N k+1 is the system N calculation factor at the (k + 1)-th moment;

[0281] Calculate the adaptive fading factor λ k+1 , as follows:

[0282]

[0283] Calculate the system state covariance matrix P X,k+1|k containing the adaptive fading factor, as follows:

[0284] P X,k+1|k = λ k+1 (P k+1|k - Q w,k ) + Q w,k ;

[0285] 14) Update the observer system measurement;

[0286] 141) Generate the system measurement volume point set and calculate the system propagated measurement volume point set, as follows:

[0287]

[0288] Wherein, S X,k+1|k is the system error covariance matrix decomposition matrix at the (k + 1)-th moment, is the system measurement volume point set at the (k + 1)-th moment, is the system propagated measurement volume point set at the (k + 1)-th moment;

[0289] 142) Update the observer system measurement prediction value As follows:

[0290]

[0291] 143) Update the system measurement error covariance matrix as follows:

[0292]

[0293] Where P Z,k+1|k is the system measurement error covariance matrix at time k + 1, and R v,k is the system measurement noise covariance matrix at time k;

[0294] 15) Update the observer state and measurement cross-covariance matrix P XZ,k+1|k as follows:

[0295]

[0296] 16) Update the observer gain matrix and the cross-covariance matrix of the system state error as follows:

[0297]

[0298] Where K k+1 is the observer gain matrix at time k + 1, is the updated value of the system state at time k + 1, and P X,k+1 is the updated value of the system state error covariance matrix at time k + 1;

[0299] The sprung mass m s , the vehicle center of mass roll height h s , the vehicle lateral velocity V y , the vehicle longitudinal velocity V x , the yaw rate ω r , and the vehicle body roll angle φ are obtained through observation as follows:

[0300] [V x V y ω r φm s h s T = C ICKF X

[0301] Where C ICKF represents the observation selection matrix, and X represents the state vector;

[0302] The required vehicle center of mass sideslip angle β is calculated through the vehicle lateral velocity V y and the vehicle longitudinal velocity V x as follows:

[0303] ​

[0304] 2) Reconstruct the vehicle dynamics model of the commercial vehicle based on the vehicle load parameters and motion parameters calculated in step 1), update the reference values and constraints of the vehicle motion parameters, update the incremental stability constraints and physical execution constraints of the four-wheel independent steering unit and the differential braking unit, and calculate the stability indicators of the four-wheel independent steering unit and the differential braking unit;

[0305] Among them, step 2) specifically includes:

[0306] 21) Based on the sprung mass m s and the vehicle body roll height h s obtained in step 1), reconstruct the established vehicle dynamics model of the commercial vehicle, and represent the reconstructed vehicle dynamics model as:

[0307]

[0308] In the formula, represents the system state variable, represents the derivative of the system state variable, u1 = [δ1 δ2 δ3 δ4] T represents the input variable of the four-wheel independent steering unit, u2 = [F x1 F x2 F x3 F x4 T represents the input variable of the differential braking unit;

[0309] Linearize the reconstructed vehicle dynamics model, define the state increment Δx k of the system at time k, Δx k = x k - x k-1 , where x k is the system state quantity at time k, x k-1 is the system state quantity at time k-1, the control input increment Δu 1,k of the four-wheel independent steering unit at time k, Δu 1,k = u 1,k - u 1,k-1 , where u 1,k is the control input variable of the four-wheel independent steering unit at time k, u 1,k-1 is the control input variable of the four-wheel independent steering unit at time k-1, the control input increment Δu 2,k of the differential braking unit at time k, Δu 2,k = u 2,k - u 2,k-1 , where u 2,k is the control input increment of the differential braking unit at time k, u 2,k-1 ​Let the control input increment at time k-1 of the differential braking unit be the input for constructing the increment equation, as follows:

[0310]

[0311] Discretize the state space equation in incremental form to obtain the incremental recurrence relation, as follows:

[0312] Δx k = AΔx k-1 + B1Δu 1,k-1 + B2Δu 2,k-1

[0313] In the formula,

[0314]

[0315] In the formula, A represents the discrete system state matrix; B1 represents the input matrix of the four-wheel independent steering unit; B2 represents the input matrix of the differential braking unit; is the Jacobi matrix, and ΔT is the sampling time;

[0316] 22) Update the reference values and constraints of the vehicle motion parameters based on the commercial vehicle chassis load parameters obtained in step 1):

[0317] The control objectives of the active rollover - sideslip controller are yaw rate, vehicle center of mass sideslip angle, and lateral load transfer ratio. The reference value of the yaw rate is selected as the nominal value of the two-degree-of-freedom vehicle model, and the nominal value of the yaw rate is corrected according to the lateral load transfer ratio. The lateral load transfer ratio LTR is restricted as follows:

[0318]

[0319] In the formula, LTR max represents the maximum lateral load transfer ratio;

[0320] Considering rollover control, the reference value ω of the yaw rate r_ref is set to:

[0321]

[0322] In the formula, K represents the stability factor of the two-degree-of-freedom model, μ represents the road surface adhesion coefficient, and δ represents the target front wheel steering angle;

[0323] The reference value of the vehicle center of mass sideslip angle is set to the steady-state response value of the vehicle two-degree-of-freedom model. Considering the limitation of the maximum value of the body sideslip angle by road surface adhesion, the nominal value β of the vehicle center of mass sideslip angle ref is corrected to:

[0324]

[0325] where K r represents the cornering stiffness of the rear wheels, and L represents the wheelbase of the vehicle;

[0326] 23) Update the incremental stability constraints and physical execution constraints of the four-wheel independent steering unit and the differential braking unit based on the commercial vehicle chassis load parameters obtained in step 1), including: the physical execution constraints of the four-wheel independent steering unit and the differential braking unit and the execution constraints in incremental form defined in step 21), and calculate the stability indexes of the four-wheel independent steering unit and the differential braking unit;

[0327] For the control system design of four-wheel independent steering, the requirements of vehicle lateral motion stability need to be met. The execution constraints of the four-wheel independent steering unit are expressed as follows:

[0328] Δu 1min ≤Δu1≤Δu 1max ,u 1min ≤u1≤u 1max

[0329]

[0330] where represents the maximum and minimum turning angles of the steering motor in the four-wheel independent steering module per unit time. Here, k1 and k2 represent the calculation parameters of the execution constraints of the four-wheel independent steering unit, and e is a constant; u 1max =-u 1min represents the physical constraint of the steering motor turning angle; ψ1 and ψ2 represent coefficients; σ1 represents the vehicle sideslip stability index, represents the upper bound of the vehicle sideslip stability index. Here, k3 and k4 represent the calculation parameters of the stability index of the four-wheel independent steering unit. The vehicle lateral motion needs to meet the constraints of the vehicle lateral stability region formed by the β-ω r phase plane;

[0331] For the differential braking control system design, the constraints include the braking force saturation constraint limited by the actuator power and the constraint to ensure that the vehicle can always operate within the safe region of the vehicle roll motion represented by the phase diagram, which is specifically expressed as:

[0332] Δu 2min ≤Δu2≤Δu 2max ,u 2min ≤u2≤0

[0333]

[0334] where represents the requirements for the maximum and minimum values of the increment of the longitudinal wheel force. Here, k5 and k6 represent the calculation parameters of the execution constraints of the differential braking unit; u2min = -F x,max represents the minimum requirement of the longitudinal wheel force; represents a coefficient; σ2 represents the vehicle rollover stability index, represents the upper bound of the critical rollover stability index of the vehicle, where k7 and k8 represent the calculation parameters of the differential braking unit stability index; based on the lateral force and related state variables of the current wheel operating point, approximately obtain the maximum single-wheel longitudinal force F under the adhesion ellipse constraint x,max , and its calculation formula is as follows:

[0335]

[0336] In the formula, ρ is the curvature parameter of the friction adhesion ellipse.

[0337] 3) Based on the stability indices of the four-wheel independent steering unit and the differential braking unit obtained in step 2), calculate the coordination parameters of the four-wheel independent steering unit and the differential braking unit; combine the physical execution constraints, incremental execution constraints, and reference values of vehicle motion parameters of the four-wheel independent steering unit and the differential braking unit obtained in step 2), construct the cost functions of the four-wheel independent steering unit and the differential braking unit, and combine the coordination parameters of the four-wheel independent steering unit and the differential braking unit to construct a comprehensive cost function with the property of a convex function, calculate the Pareto solution of the execution strategies of the four-wheel independent steering unit and the differential braking unit, obtain the optimal control strategies of the four-wheel independent steering unit and the differential braking unit, perform information interaction between the four-wheel independent steering unit and the differential braking unit, and send the steering motor controller and the braking motor controller to execute the optimal control strategies;

[0338] Among them, the specific steps of step 3) include:

[0339] 31) Based on the vehicle stability indices σ1 and σ2 in step 23), calculate the coordination parameters of the four-wheel independent steering unit and the differential braking unit respectively. The relationship between the coordination parameters and the stability indices is expressed by the following formula:

[0340]

[0341] In the formula, κ p respectively represent the upper and lower bounds of the coordination parameter κ p ; represents the upper bound of the stability index σ p ;

[0342] 32) Based on the reference values of vehicle motion parameters calculated in step 22), and the coordination parameters and shared information between units of the four-wheel independent steering unit and the differential braking unit calculated in step 31), construct the cost functions of the four-wheel independent steering unit and the differential braking unit, and construct a comprehensive cost function with the property of a convex function based on the cost functions of the four-wheel independent steering unit and the differential braking unit;

[0343] For the four-wheel independent steering unit, to ensure the yaw stability of the vehicle, the cost function J1(k) is expressed as:

[0344]

[0345] In the formula, x ref = [β ref ω r_ref T represents the reference state vector, C1 is the state selection matrix, N p is the prediction horizon, Q1 represents the process state matrix of the four-wheel independent steering unit, R1 represents the input cost matrix of the four-wheel independent steering unit, and P1 represents the terminal state matrix of the four-wheel independent steering unit;

[0346] For the differential braking unit, to improve the roll stability during vehicle steering, the cost function J2(k) is expressed as:

[0347]

[0348] In the formula, x ref = φ ref represents the reference state, C2 is the state selection matrix, Q2 represents the process state matrix of the differential braking unit, R2 represents the input cost matrix of the differential braking unit, and P2 represents the terminal state matrix of the differential braking unit;

[0349] Based on the coordination parameters of the four-wheel independent steering unit and the differential braking unit calculated in step 31), design a linear convex combination of the control objective functions as the comprehensive cost function J(U) as follows:

[0350]

[0351] In the formula, J q represents the cost functions of the four-wheel independent steering unit and the differential braking unit;

[0352] Solve the optimal cost function to obtain the execution strategies of the four-wheel independent steering unit and the differential braking unit, and the Pareto solution strategy U is:

[0353]

[0354] 33) Based on the Pareto solutions obtained in step 32), construct the shared strategy for the four-wheel independent steering unit and the differential braking unit at the next moment, which is the optimal control strategy U for the four-wheel independent steering unit and the differential braking unit. n,q , as follows:

[0355]

[0356] Where U n,q (k|k - 1) represents the shared strategy for the four-wheel independent steering unit and the differential braking unit to construct the strategy at moment k based on moment k - 1, and K x represents the system state feedback gain, and x m represents the system terminal state, and υ q represents the control input of the four-wheel independent steering unit and the differential braking unit;

[0357] 34) Conduct information interaction between the four-wheel independent steering unit and the differential braking unit, and send it to the steering motor controller of the four-wheel independent steering module and the braking motor controller of the differential braking module for control execution.

[0358] The specific application scenarios of the present invention are numerous. The above description is only the preferred implementation mode of the present invention. It should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements can be made, and these improvements should also be regarded as the protection scope of the present invention.

Claims

1. A commercial vehicle distributed skateboard chassis, characterized in that: include: Wheels, in-wheel motors, four-wheel independent steering module, differential brake module, suspension module, frame, six-degree-of-freedom inertial measurement unit, longitudinal vehicle speed sensor and active anti-rollover-side slip controller; The built-in rotor output shaft of the hub motor is fixedly connected to the wheel to drive the wheel to rotate; The longitudinal vehicle speed sensor is fixedly mounted on the output shaft of the built-in rotor of the wheel hub motor and is used to collect the longitudinal vehicle speed signal; The suspension module comprises: an upper fork arm, a lower fork arm and a suspension mechanism assembly; the upper fork arm and the lower fork arm are fixedly connected to the vehicle frame, and the suspension mechanism assembly is fixedly connected to the lower fork arm; The four-wheel independent steering module includes: a steering motor controller, a steering motor, a planetary gear reduction mechanism, an angle sensor and a steering shaft; the steering motor controller is fixedly connected to the steering motor, and the steering motor is rotationally connected to the upper fork arm through a housing; the planetary gear reduction mechanism is located below the steering motor and is fixedly connected to the steering motor, the output shaft of the planetary gear reduction mechanism is connected to the steering shaft through a key, the angle sensor is fixedly installed at the output shaft position of the planetary gear reduction mechanism, the steering shaft is fixedly connected to the housing of the hub motor to drive the wheel to rotate, and the lower end of the steering shaft is rotationally connected to the lower fork arm; the steering motor controller controls the steering motor to drive the planetary gear reduction mechanism to rotate by receiving the steering control signal sent by the active anti-rollover-side slip controller, thereby driving the rotation of the steering shaft to control the rotation of the wheel; the steering motor controller also receives the angle signal collected by the angle sensor to adjust the control steering torque; The differential brake module includes: a brake disc, a brake caliper assembly, a brake motor controller, a brake motor and a ball screw reduction mechanism; the brake disc is fixedly connected to the built-in rotor of the wheel hub motor, the brake caliper assembly is fixedly connected to the housing of the wheel hub motor, the ball screw reduction mechanism is fixedly connected to the brake caliper assembly, the output shaft of the brake motor is connected to the input shaft of the ball screw reduction mechanism, and the brake motor controller is fixedly connected to the brake motor; the brake motor controller receives a brake control signal sent by an active rollover prevention-side slip controller, controls the brake motor to drive the output shaft of the ball screw reduction mechanism to brake, and presses the brake caliper assembly and the brake disc to achieve braking; The six-degree-of-freedom inertial measurement unit is fixedly installed at the center of gravity of the commercial vehicle body, and is used to measure vehicle motion information and generate vehicle motion data, including: longitudinal acceleration, lateral acceleration, yaw angular velocity and roll angular velocity; The active anti-rollover-side slip controller is fixedly installed on the vehicle frame, and includes: an observer unit, a four-wheel independent steering unit and a differential braking unit. The observer unit is used to generate a vehicle motion parameter and a load parameter signal according to vehicle motion data and a longitudinal vehicle speed signal. The four-wheel independent steering unit is used to generate a steering control signal according to a signal generated by the observer unit and a braking control signal generated by the differential braking unit. The differential braking unit is used to generate a braking control signal according to a signal generated by the observer unit and a steering control signal generated by the four-wheel independent steering unit.

2. A method for controlling the lateral and longitudinal coordinated stability of a commercial vehicle distributed skateboard chassis under variable load conditions, based on the chassis of claim 1, characterized in that: The steps are as follows: 1) Establish a commercial vehicle vehicle dynamics model to construct a commercial vehicle vehicle state observation model. Through the observation model, a commercial vehicle distributed skateboard chassis load state-motion state observer based on an improved volumetric Kalman filter algorithm is constructed to obtain vehicle load parameters and motion parameters; 2) reconstructing the vehicle dynamics model of the commercial vehicle according to the vehicle load parameters and motion parameters calculated in step 1), updating the reference values ​​and constraints of the vehicle motion parameters, updating the incremental stability constraints and physical execution constraints of the four-wheel independent steering unit and the differential braking unit, and calculating the stability index of the four-wheel independent steering unit and the differential braking unit; 3) Based on the stability index of the four-wheel independent steering unit and the differential braking unit obtained in step 2), the coordination parameters of the four-wheel independent steering unit and the differential braking unit are calculated; in combination with the physical execution constraints, incremental execution constraints and reference values ​​of the vehicle motion parameters of the four-wheel independent steering unit and the differential braking unit obtained in step 2), the cost function of the four-wheel independent steering unit and the differential braking unit is constructed, and a comprehensive cost function with convex function properties is constructed in combination with the coordination parameters of the four-wheel independent steering unit and the differential braking unit, the Pareto solution of the execution strategy of the four-wheel independent steering unit and the differential braking unit is calculated, and the optimal control strategy of the four-wheel independent steering unit and the differential braking unit is obtained, and information is exchanged between the four-wheel independent steering unit and the differential braking unit and sent to the steering motor controller and the brake motor controller to execute the optimal control strategy.

3. The method for controlling the lateral and longitudinal coordinated stability of a commercial vehicle distributed skateboard chassis under variable load conditions according to claim 2, characterized in that: The step 1) of establishing a vehicle dynamics model for a commercial vehicle includes a vehicle body dynamics model, a magic tire model and a wheel dynamics model; The vehicle body dynamics model is established as follows: In the formula, m s is the sprung mass, which refers to the mass carried by the commercial vehicle suspension; m us is the unsprung mass, which indicates the mass of the commercial vehicle not carried by the suspension; m is the vehicle mass; h s K is the vehicle center of mass roll height; φ is the roll angle stiffness; C φ is the roll angle damping; g is the gravitational acceleration; ΣF x ,ΣF y ,ΣM z They represent the resultant force of the vehicle along the x-axis, the resultant force of the vehicle along the y-axis, and the resultant moment of the vehicle around the z-axis; φ is the roll angle of the vehicle body; is the body roll angular velocity; is the derivative of the body roll angular velocity; a x ,a y ,a ys are the vehicle longitudinal acceleration, lateral acceleration and lateral acceleration of the sprung mass; V x ,V y are the longitudinal and lateral velocities of the vehicle, respectively; are the derivative of the vehicle's longitudinal velocity and lateral velocity, respectively; ω r is the vehicle yaw angular velocity; is the derivative of the vehicle's yaw rate; I xzs is the product of inertia of the sprung mass around the x and z axes; R xs ,R z are the rotation radius of the sprung mass around the x-axis and the rotation radius of the vehicle around the z-axis respectively; I xs ,I z ,I xzs are the moment of inertia of the sprung mass around the x-axis, the moment of inertia of the vehicle around the z-axis, and the joint moment of inertia of the vehicle around the x- and z-axes; O represents the calculation parameter; Through force analysis, it can be obtained that the resultant force of the vehicle body dynamics model along the x-axis direction is: ∑F x =F x1 cosδ1+F x2 cosδ2+F x3 cosδ3+F x4 cosδ4-F y1 sinδ1-F y2 sinδ2-F y3 sinδ3-F y4 sinδ4 In the formula, F xi is the tire longitudinal force; F yi is the tire lateral force; δ i is the wheel angle, i=1, 2, 3, 4 correspond to the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively; The resultant force along the y-axis is: ∑F y =F x1 sinδ1+F x2 sinδ2+F x3 sinδ3+F x4 sinδ4+F y1 cosδ1+F y2 cosδ2+F y3 cosδ3+F y4 cosδ4 The resultant moment about the z-axis is: Where a and b are the distances from the center of mass to the front axle and the distance from the center of mass to the rear axle respectively, and d is the tire center distance; The magic tire formula is used to establish a nonlinear magic tire model as follows: Where Y v is the tire lateral force, tire longitudinal force or tire self-aligning torque; X h is the tire slip angle or tire longitudinal slip rate; x h ,y v is the calculation parameter; D t is the peak factor; B t is the stiffness factor; C t is the curve shape factor; E t is the curve curvature factor; S h S is the horizontal drift of the curve; v Drift in the vertical direction of the curve; In the formula, F z is the vertical ground load on the tire; s is the longitudinal slip rate of the tire; B x ,C x ,D x ,E x ,BCD x are tire longitudinal force calculation parameters; a0, a1, …, a 10 is the fitting parameter; α is the tire slip angle; γ is the wheel camber angle; B y ,C y ,D y ,E y ,BCD y b0, b1, …, b are the tire lateral force calculation parameters; 13 is the fitting parameter; Under the combined conditions of steering and braking, the longitudinal force F of the tire x With lateral force F y Respectively expressed as: In the formula, ε x ,ε y ,ε respectively represent the longitudinal force parameter, lateral force parameter and comprehensive tire force parameter; Establish wheel dynamics model; The vertical load model of each wheel is established as follows: In the formula, F zi represents the vertical load on the wheel, i=1,2,3,4 represents the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively, L represents the wheelbase of the front and rear axles of the vehicle; H1 and H2 represent the height of the vehicle center of mass from the ground and the height of the sprung mass center of mass from the ground respectively; The side slip angle model of each tire is established as follows: In the formula, α i Indicates the side slip angle of each wheel, i=1,2,3,4 represents the left front wheel, right front wheel, left rear wheel, and right rear wheel respectively.

4. The method for controlling the lateral and longitudinal coordinated stability of a commercial vehicle distributed skateboard chassis under variable load conditions according to claim 3, characterized in that: In the step 1), a commercial vehicle state observation model including vehicle load parameters and motion parameters is established based on the commercial vehicle vehicle dynamics model. The load parameters include the sprung mass m s and the vehicle's center of mass roll height h s The motion parameters include the body roll angle φ, the vehicle center of mass sideslip angle β and the vehicle yaw rate ω r ; Establish a commercial vehicle vehicle dynamics augmented model including vehicle load parameters and motion parameters as follows: In the formula, represents the derivative of the sprung mass, represents the derivative of the vehicle's center of mass roll height; Based on the commercial vehicle vehicle dynamics augmented model and sensor data, including: longitudinal vehicle speed V x , yaw angular velocity ω r , roll angular velocity Longitudinal acceleration a x and lateral acceleration a y , establish the system continuous state space model as follows: u=[δ1 δ2 δ3 δ4 F x1 F x2 F x3 F x4 ] T In the formula, X represents the state vector, that is, the vector to be observed; represents the derivative of the state vector; Z represents the measurement vector; u represents the system input vector; g(·), h(·) represent the continuous state function and the continuous observation function respectively; w and v represent the system process noise and the measurement noise respectively, w~N(0,Q), v~N(0,R), both noises are Gaussian white noise; The forward Euler method is used to discretize the continuous state space model of the system, and the discretized state space model of the system is established, that is, the commercial vehicle state observation model, as follows: Where k represents the time; X(k+1) represents the state vector at time k+1, X(k) represents the state vector at time k, Z(k) represents the measurement vector at time k, w(k) represents the system process noise at time k, v(k) represents the system measurement noise at time k, u(k) represents the system input at time k, G(·) and H(·) represent the state equation and observation equation respectively; The state equation and output equation of the commercial vehicle state observation model are expressed as: G(X(k),u(k))=X(k)+g(X(k),u(k))·T Where V x (k) represents the longitudinal velocity of the vehicle at time k, ω r (k) represents the yaw rate of the vehicle at time k, represents the roll angular velocity of the vehicle at time k, a x (k) represents the longitudinal acceleration of the vehicle at time k, a y (k) represents the lateral acceleration of the vehicle at time k, and T is the sampling time of the observation system.

5. The method for controlling the lateral and longitudinal coordinated stability of a commercial vehicle distributed skateboard chassis under variable load conditions according to claim 4, characterized in that: In the step 1), a commercial vehicle distributed skateboard chassis load state-motion state observer based on an improved volumetric Kalman filter algorithm is constructed through an observation model to calculate vehicle load parameters and motion parameters, specifically including: 11) Initialize the system status as follows: In the formula, X0 is the initial value of the system state, is the initial observation value of the system state and satisfies P X,0 is the initial value of the system state error covariance matrix, E(·) represents the expected calculation function; 12) Generate volume point set and calculate system state volume point set as follows: The cubature Kalman filter algorithm is improved by using diagonal transformation to prevent the stability of the algorithm from decreasing due to the loss of positive definiteness of the state error covariance matrix during the iteration process. The following transformation is performed on the error covariance matrix: Setting A d is an n-order real symmetric matrix, then there is an n-order orthogonal matrix V such that V T A d V=V -1 A d V=D,that is A d =V T DV, where D is A d The n eigenvalues ​​of the diagonal matrix are diagonal elements, then the error covariance matrix P of the system state X,k It is expressed as: Where V X,k Denotes the auxiliary matrix of error covariance matrix decomposition, D X,k represents the decomposition matrix of the error covariance matrix, The eigenvalues ​​of the error covariance matrix representing the system state, i p =1,2,…,n, representing the sequence of eigenvalues ​​of the error covariance matrix of the system state; Define the DCP function as: In the formula, S X,k The matrix representing the diagonal decomposition of the system state error covariance matrix; The point set of the calculated system state volume is as follows: In the formula, is the volume point set, i d =1,2,…,n,m d =2n,I n×n represents the n-dimensional identity matrix, m d represents the system volume point set parameter, X id,k is the system state volume point set calculation, is the observed value of the system state at time k; 13) Update the observer system status as follows: 131) The propagation system state volume point set is as follows: In the formula, represents the volume point set of the system propagation state at time k+1, u k represents the system input value at time k; 132) Update the observer system prediction state In the formula, represents the predicted value of the system state at time k+1; 133) Update the system state error covariance matrix P X,k+1|k ; Calculate the initial value of the system state error covariance matrix as follows: In the formula, is the initial value of the system state error covariance matrix at time k+1, Q w,k is the system state noise covariance matrix at time k; Generate the initial value of the system measurement volume point set and calculate the initial value of the system propagation measurement volume point set as follows: In the formula, The initial value of the system state error covariance matrix decomposition matrix prediction at time k+1, is the initial value of the system measurement volume point set at time k+1, is the initial value of the system propagation measurement volume point set at time k+1; Update the observer system measurement prediction initial value as follows: In the formula, Predict the initial value of the system measurement at time k+1; An adaptive fading factor is introduced to adjust the error covariance matrix online to improve the observation accuracy of the load state-motion state observer of the commercial vehicle distributed skateboard chassis; Calculate the system measurement error as follows: In the formula, e k+1 is the system measurement error at time k+1, Z k+1 is the system measurement value at time k+1; Calculate the system measurement calculation factors as follows: In the formula, ι k+1|k is the system measurement calculation factor at time k+1; Calculate the system status calculation factor as follows: In the formula, χ k+1|k is the system state calculation factor at time k+1; Calculate the system H factor as follows: In the formula, H k+1 Calculate the factor of system H at time k+1; Calculate the system M calculation factor as follows: Where M k+1 Calculate the factors for system M at time k+1; Calculate the System V factor: In the formula, e0, are the initial system measurement error and the transpose of the initial system measurement error, respectively. θ, θ∈(0,1) are the forgetting factors. V k is the calculation factor of the system V at time k, V k+1 Calculate the factor for system V at time k+1; Calculate the system N calculation factor as follows: In the formula, ζ,ζ≥1 is the weakening factor, R v,k is the system measurement noise covariance matrix at time k, N k+1 Calculate the factor for system N at time k+1; Calculate the adaptive fading factor λ k+1 ,as follows: λ k+1 =max(1,tr[N k+1 ] / tr[M k+1 ]) Calculate the system state covariance matrix P with adaptive fading factors X,k+1|k ,as follows: P X,k+1|k =λ k+1 (P k+1|k -Q w,k )+Q w,k ; 14) Update observer system measurements; 141) Generate a system measurement volume point set and calculate a system propagation measurement volume point set as follows: In the formula, S X,k+1|k is the system error covariance matrix decomposition matrix at time k+1, is the system measurement volume point set at time k+1, is the system propagation measurement volume point set at time k+1; 142) Update the observer system measurement prediction value as follows: 143) Update the system measurement error covariance matrix as follows: Where P Z,k+1|k is the measurement error covariance matrix of the system at time k+1, R v,k is the system measurement noise covariance matrix at time k; 15) Update the observer state and measurement cross covariance matrix P XZ,k+1|k ,as follows: 16) Update the observer gain matrix and the system state error covariance matrix as follows: In the formula, K k+1 is the observer gain matrix at time k+1, is the updated value of the system state at time k+1, P X,k+1 is the updated value of the system state error covariance matrix at time k+1; The sprung mass m is obtained by observation s , vehicle center of mass roll height h s , vehicle lateral speed V y , vehicle longitudinal speed V x , yaw angular velocity ω r , the body roll angle φ, as follows: [V x V y ω r φm s h s ] T =C ICKF X In the formula, C ICKF represents the observation selection matrix, X represents the state vector; The vehicle's lateral velocity V y , vehicle longitudinal speed V x The required vehicle center of mass sideslip angle β is calculated as follows:

6. The method for controlling the lateral and longitudinal coordinated stability of a commercial vehicle distributed skateboard chassis under variable load conditions according to claim 5, characterized in that: The step 2) specifically includes: 21) Based on the sprung mass m obtained in step 1) s and the vehicle's center of mass roll height h s , the established commercial vehicle vehicle dynamics model is reconstructed, and the reconstructed vehicle dynamics model is characterized as: In the formula, represents the system state variable, Represents the derivative of the system state variable, u1=[δ1 δ2 δ3 δ4] T Indicates the input variable of the four-wheel independent steering unit, u2=[F x1 F x2 F x3 F x4 ] T Represents the input variable of the differential brake unit; The reconstructed vehicle dynamics model is linearized and the system state increment Δx at time k is defined as k , Δx k =x k -x k-1 , where x k is the system state at time k, x k-1 is the system state at time k-1, and the control input increment Δu of the four-wheel independent steering unit at time k 1,k , Δu 1,k =u 1,k -u 1,k-1 , where u 1,k is the control input variable of the four-wheel independent steering unit at time k, u 1,k-1 is the control input variable of the four-wheel independent steering unit at time k-1, and the control input increment Δu of the differential brake unit at time k 2,k , Δu 2,k =u 2,k -u 2,k-1 , where u 2,k is the control input increment of the differential brake unit at time k, u 2,k-1 The control input increment of the differential brake unit at time k-1 forms the increment equation as follows: Discretize the incremental state space equation and obtain the incremental recursive relationship as follows: Δx k =AΔx k-1 +B1Δu 1,k-1 +B2Δu 2,k-1 In the formula, Where A represents the discrete system state matrix; B1 represents the four-wheel independent steering unit input matrix; B2 represents the differential brake unit input matrix; is the Jacobi matrix, ΔT is the sampling time; 22) Based on the commercial vehicle chassis load parameters obtained in step 1), the reference values ​​and constraints of the vehicle motion parameters are updated: The control targets of the active rollover and sideslip controller are yaw rate, vehicle center of mass sideslip angle and lateral load transfer rate. The yaw rate reference value is selected as the nominal value of the two-degree-of-freedom vehicle model. The nominal value of the yaw rate is corrected according to the lateral load transfer rate. The lateral load transfer rate LTR is limited as follows: In the formula, LTR max represents the maximum lateral load transfer rate; Considering the anti-rollover control, the yaw rate reference value ω r_ref Set to: In the formula, K represents the stability factor of the two-degree-of-freedom model, μ represents the road adhesion coefficient, and δ represents the target front wheel turning angle; The reference value of the vehicle's center of mass sideslip angle is set as the steady-state response value of the vehicle's two-degree-of-freedom model. Considering the limitation of the road adhesion on the maximum value of the vehicle's center of mass sideslip angle, the nominal value of the vehicle's center of mass sideslip angle β ref Corrected to: In the formula, K r represents the rear wheel cornering stiffness, and L represents the vehicle wheelbase; 23) Based on the commercial vehicle chassis load parameters obtained in step 1), the incremental stability constraints and physical execution constraints of the four-wheel independent steering unit and the differential braking unit are updated, including: the physical execution constraints of the four-wheel independent steering unit and the differential braking unit and the execution constraints in the incremental form defined in step 21), and the stability index of the four-wheel independent steering unit and the differential braking unit is calculated; The design of the four-wheel independent steering control system needs to meet the requirements of vehicle lateral motion stability. The execution constraints of the four-wheel independent steering unit are expressed as follows: Δu 1min ≤Δu1≤Δu 1max ,u 1min ≤u1≤u 1max In the formula, represents the maximum and minimum turning angles of the steering motor in the four-wheel independent steering module per unit time, where k1 and k2 represent the constraint calculation parameters of the four-wheel independent steering unit, and e is a constant; u 1max =-u 1min represents the physical constraint of the steering motor turning angle; ψ1, ψ2 represent coefficients; σ1 represents the vehicle side slip stability index, represents the upper bound of the vehicle's sideslip stability index, where k3 and k4 represent the calculation parameters of the stability index of the four-wheel independent steering unit. The lateral motion of the vehicle needs to satisfy the β-ω r The constraints of the vehicle's lateral stability region constituted by the phase plane; For the design of differential brake control system, the constraints include brake force saturation constraint limited by actuator power and the requirement that the vehicle can always operate under the control of The constraints within the safe area of ​​the vehicle roll motion represented by the phase diagram are specifically expressed as: Δu 2min ≤Δu2≤Δu 2max ,u 2min ≤u2≤0 In the formula, Indicates the maximum and minimum requirements of the wheel longitudinal force increment, where k5 and k6 represent the constraint calculation parameters executed by the differential brake unit; u 2min =-F x,max Indicates the minimum requirement for the longitudinal force of the wheel; represents the coefficient; σ2 represents the vehicle rollover stability index, represents the upper limit of the critical rollover stability index of the vehicle, where k7 and k8 represent the calculation parameters of the stability index of the differential brake unit; according to the lateral force and related state quantities of the current working point of the wheel, the maximum single wheel longitudinal force F under the adhesion ellipse constraint is approximately obtained. x,max , which is calculated as follows: Where ρ is the curvature parameter of the friction adhesion ellipse.

7. The method for controlling the lateral and longitudinal coordinated stability of a commercial vehicle distributed skateboard chassis under variable load conditions according to claim 6, characterized in that: The step 3) specifically includes: 31) Based on the vehicle stability indexes σ1, σ2 in step 23), the coordination parameters of the four-wheel independent steering unit and the differential brake unit are calculated respectively. The relationship between the coordination parameters and the stability index is expressed by the following formula: In the formula, κ p are the coordination parameters κ p The upper and lower bounds of The stability index σ p The upper bound of 32) constructing a cost function of the four-wheel independent steering unit and the differential braking unit based on the vehicle motion parameter reference value calculated in step 22) and the coordination parameters of the four-wheel independent steering unit and the differential braking unit and the shared information between the units calculated in step 31), and constructing a comprehensive cost function with a convex function property based on the cost function of the four-wheel independent steering unit and the differential braking unit; For the four-wheel independent steering unit, to ensure the yaw stability of the vehicle, the cost function J1(k) is expressed as: In the formula, x ref =[β ref ω r_ref ] T represents the reference state vector, C1 is the state selection matrix, N p For the prediction time domain, Q1 represents the process state matrix of the four-wheel independent steering unit, R1 represents the input cost matrix of the four-wheel independent steering unit, and P1 represents the terminal state matrix of the four-wheel independent steering unit; For the differential brake unit, the roll stability of the vehicle during steering is improved, and the cost function J2(k) is expressed as: In the formula, x ref =φ ref represents the reference state, C2 is the state selection matrix, Q2 represents the differential brake unit process state matrix, R2 represents the differential brake unit input cost matrix, and P2 represents the differential brake unit terminal state matrix; According to the coordination parameters of the four-wheel independent steering unit and the differential brake unit calculated in step 31), a linear convex combination of the control objective function is designed as a comprehensive cost function J(U) as follows: In the formula, J q represents the cost function of the four-wheel independent steering unit and the differential brake unit; Solve the optimal cost function to obtain the execution strategy of the four-wheel independent steering unit and the differential braking unit, and get the Pareto solution strategy U as: 33) According to the Pareto solution obtained in step 32), the sharing strategy of the four-wheel independent steering unit and the differential braking unit at the next moment is constructed, that is, the optimal control strategy U of the four-wheel independent steering unit and the differential braking unit n,q ,as follows: Where U n,q (k|k-1) represents the shared strategy of the four-wheel independent steering unit and the differential brake unit at time k-1 to construct time k, K x represents the system state feedback gain, x m represents the terminal state of the system, υ q Indicates the control input of the four-wheel independent steering unit and the differential brake unit; 34) Perform information exchange between the four-wheel independent steering unit and the differential brake unit, and send the steering motor controller of the four-wheel independent steering module and the brake motor controller of the differential brake module for control execution.

Images